1. Introduction
Tax evasion poses a major public policy issue worldwide due to the resulting losses of revenues that could be used by governments to maintain adequate social services and sustainable economies [1,2,3,4,5]. One critical social objective of taxation is the redistribution of income and wealth to reduce inequality [3,4,5,6]. As noted by Wenzel [5], tax evasion will increase inequality if the extent of evasion increases with income levels. It is often observed that higher income individuals have more resources to facilitate tax evasion, such as access to expensive tax and legal services provided by accounting and law firms (e.g., [3,4]). You and Khagram [7] also note that the wealthy have greater motivation and resources to engage in corruption (including tax evasion). Recent empirical studies also document higher rates of tax evasion among the wealthy (e.g., [8]). Thus, it appears quite likely that tax evasion exacerbates economic inequality and consequently poses a fundamental threat to the sustainability of fair and just socioeconomic systems.
Given the significance of the public policy issues involved, it seems surprising that previous research has not recognized that some taxpayers may have ideological preferences for the persistence of economic inequality, and such preferences may motivate tax evasion. Research in social psychology suggests that support for inequality is a core component of certain socioeconomic and political belief systems. For instance, Azevedo, Jost, Rothmund and Sterling [9] concluded that the political belief systems of U.S. and U.K. citizens are primarily organized around support for vs. opposition to social and economic forms of inequality. Findings such as these highlight the critical role of support for inequality in individuals’ ideological belief systems, and the need to better understand the influence of such ideologies in various realms of decision making including tax compliance.
The lack of attention paid to the effects of preferences for sustained inequality in the taxation literature appears consistent with a similar dearth of research on inequality and poverty in the business literature recently noted by organizational scholars (e.g., [2,10]). In contrast, in the broader social science literature, significant and growing attention has been devoted to the causes, consequences and potential remediation of inequality in recent decades. For instance, Wilkinson and Pickett [11] provide extensive evidence documenting that greater inequality is associated with a litany of social problems across societies, including drug use, poor mental and physical health, reduced life expectancies and higher rates of violence and imprisonment. Indeed, they demonstrate that the healthiest and happiest societies tend to have high levels of equality. Stiglitz [12] notes that the U.S. has the highest level of economic inequality among developed nations and argues that entrenched inequality poses a threat to the sustainability of the U.S. economy and democracy. In his influential book, Piketty [13] examines evidence on long-term trends in inequality in many countries (including the U.S.) to illustrate the changes in the concentration of income and wealth. His data indicate that since the 1970s the U.S. has had the largest increase in income inequality among developed countries. He argues that disparities in income and wealth will continue to grow in the future in the absence of more effective government intervention. To rectify the problem of growing inequality and place societies on a more sustainable path, Stiglitz [12] and Piketty [13] propose a more progressive system of taxation on income as well as a wealth tax.
Both Stiglitz [12] and Piketty [13] argue that economic inequality arises and is perpetuated through the influence of power elites who have an outsized influence on social and economic policies. These arguments are consistent with the results of numerous studies documenting an apparent disparity between the preferences of a majority of U.S. citizens and ongoing trends toward increased inequality. For example, McCall [14] and Page and Jacobs [15] document that the majority of U.S. citizens are concerned about increasing income and wealth inequality, and favor government redistribution programs that increase economic opportunities for the underprivileged. However, due to the continuing influence of neoliberalism, calls for the reduction of inequality will surely be contested.
For instance, Tsui, Enderle and Jiang [16] recently argued from economic, social psychological, and ethical perspectives that corporations should take a more active role in preventing extreme income inequality by reducing top executive pay and paying employees a living wage to ensure they can maintain their well-being and social dignity. To enhance the prospects for just and sustainable societies, they also advocate a stronger role for governments and significantly higher taxes on the wealthy, citing Piketty’s [13] position that in developed economies the optimal top tax bracket is probably above 80 percent. These arguments were met with a retort from Agarwal and Holmes [17], who questioned the wisdom of intervention in market processes, citing foundational neoliberal texts such as Hayek [18] and Friedman [19].
Such debates make it clear that attitudes toward economic inequality and how it should be addressed (if at all) depend critically on individuals’ beliefs or ideologies. As noted above, research in social and political psychology documents that some individuals are motivated by ideological belief systems to oppose equality in principle [9]. For instance, individuals who possess a social dominance orientation tend to support the persistence of inequality and hierarchical dominant/subordinate relations among social groups [20,21,22]. Social dominators tend to prefer unjust social systems they feel they can manipulate for their own benefit [23].
In the current study, we were particularly interested in the association between support for economic inequality and tax evasion. Ideological support for economic inequality and related resentment of redistribution efforts could reduce taxpayers’ willingness to pay a fair share of taxes, and thus perpetuate the growing gap between the rich and poor and related social problems. However, the influence of support for economic inequality in the context of tax avoidance/evasion has not been investigated. Such beliefs may reduce the perceived legitimacy of progressive tax systems, and prior research has shown that taxpayers are more likely to engage in noncompliance if they question the legitimacy of the system [5]. Because this is a potentially significant but understudied issue, it is the primary focus of our study.
We also propose that individuals who are ideologically opposed to equality will be likely to exhibit Machiavellian tendencies. Machiavellianism is associated with antisocial attitudes and traits such as low levels of empathy and a tendency to view fellow humans as dishonest, weak, and deserving of their fate [24,25]. Thus, individuals who are ideologically motivated to oppose equality should be likely to adopt such attitudes as a means of rationalizing their preferred conclusion that societies should be largely indifferent to the plight of the poor and disadvantaged. However, the association between support for economic inequality and Machiavellianism has not previously been addressed. Consistent with much previous research in business and the social sciences, high Machiavellians should also be prone to unethical and illegal actions such as tax evasion; consequently, Machiavellianism should partially mediate the association between support for inequality and tax evasion intentions.
1.1. Overview of Tax Avoidance Research
A number of researchers have recounted the historical development of the tax avoidance/evasion literature in a similar fashion [5,26,27,28]. Early work in the field focused primarily on economic incentives for compliance such as audit and penalty rates (e.g., [29]). However, over time, it became apparent that economic considerations could not adequately explain taxpayer behavior. For instance, voluntary tax compliance rates were found to be higher than economic models predicted [30,31,32]. As noted by Brizi et al. [26], this “puzzle of compliance” led researchers to seek new explanations of taxpayer behavior, shifting the focus of much compliance research from “rational” to “social” taxpayers.
One explanation offered to explain the puzzle of compliance was the assumption that taxpayers possessed an intrinsic motivation or perceived moral and social responsibility to pay taxes [26]. This construct has often been referred to as “tax morale”, and empirical research has found that higher levels of tax morale are associated with higher rates of compliance [33]. Another social psychological variable that has received significant attention in the past two decades is social norms, or taxpayer perceptions of the general expectations or moral standards regarding tax compliance among various social collectives, such as all taxpayers or valued others. In general, research suggests that taxpayer perceptions of social norms are associated with compliance decisions [34,35,36,37,38].
Recent studies of tax compliance have also examined issues relating to economic inequality. Durante et al. [6] concluded that although taxpayers were to some extent willing to pay to reduce inequality between themselves and others, self-interest was the dominant determinant of demand for income redistribution. Brizi et al. [26] found that taxpayers’ social value orientations, defined as a broad motivational construct regarding how resources or outcomes should be divided between the self and others, significantly influence their tax avoidance/evasion decisions. Specifically, individuals with a prosocial orientation tend to possess higher levels of tax morale and be less likely to engage in noncompliance.
Studies have also documented that relatively simple measures of political orientation (e.g., self-classification as liberal or conservative) are associated with compliance decisions, with more liberal taxpayers generally being more likely to comply [6]. However, little attention has been given to the potential influence of socioeconomic ideologies such as support for inequality on tax compliance. Socioeconomic ideologies may be related to broad political classifications, but detailed studies of ideological beliefs hold promise for providing a more nuanced understanding of taxpayer motivations. Ideological beliefs may be associated with undesirable attitudes (e.g., prejudice and racism) that influence taxpayer compliance in ways not often recognized.
To place our study in perspective, we introduce our discussion of support for economic inequality and its relevance to tax evasion by discussing the somewhat broader and widely recognized construct of social dominance. A social dominance orientation includes preferences for the persistence of inequality and the closely related preferences for hierarchical dominant/subordinate social relations. Social dominance appears to be fundamentally at odds with conceptions of social responsibility and socioeconomic justice. We discuss this construct and its implications in the following section, with a particular emphasis on support for the persistence of economic inequality.
1.2. Social Dominance and Support for Inequality
Social dominance has been extensively studied in social, political and environmental psychology in recent decades (e.g., [9,20,21,23,39,40,41,42]). As noted by Jost and Thompson [20], social dominance is positively associated with political and economic conservatism, authoritarianism, nationalism, cultural elitism, racism and sexism and negatively associated with support for women’s rights and pro-environmental actions. The association of social dominance with such a wide variety of social, political and economic attitudes suggests that it is a fundamental ideological belief system.
Jost and Thompson [20] found that the social dominance scale of Pratto et al. [21] comprised two related elements: support for dominance of one’s social group and opposition to equality (though the two factors had a positive correlation above 0.6). Due to our interest in the effects of ideological beliefs regarding the acceptability of disparities in wealth and income on tax evasion, we chose to focus our empirical study on the effects of beliefs regarding economic inequality. Altemeyer [23] developed the Social Inequality Scale to obtain a better understanding of why individuals with a social dominance orientation prefer inequality. He concluded based on a survey of undergraduate psychology students and their parents that support for inequality was not based on the philosophical or pragmatic considerations often voiced by social dominators, (such views include, for instance, belief that unregulated markets provide a fair or meritocratic basis for the distribution of economic resources and that providing someone with financial assistance will reduce incentives for work and thereby encourage indolence [23]) but rather on a simple “preference for an unfair and unjust social system that they believe benefits them”, p. 425 [23].
This conclusion is supported by an analysis of items included in the economic inequality scale (taken from the Social Inequality Scale of Altemeyer [23], see Appendix B, Panel A). For instance, item 1 (“Economic equality is a bad idea dreamed up by muddle-headed ‘do-gooders’. It would be a big mistake to pursue it.”) is a broad and non-substantive dismissal of the desirability of equality coupled with insults of those who support it. Items 3 and 4 (“People have no right to economic equality. All of us should get as much as we can, and if some don’t get enough, that’s their problem” and “If the natural forces of supply and demand and power make a few people immensely wealthy and millions of others poor, so be it.”) appear to reflect an underlying Social Darwinist perspective [43] consistent with support for unregulated capitalism regardless of the consequences it may produce for the less-privileged. Item 7 (“Nobody should get extra help improving his place in society. Everyone should start off with what his family gives him, and go from there.” (emphasis added)) expresses direct opposition to programs of social uplift, and also reflects support for an aristocratic society in which private capital is accumulated and passed from one generation to another. This statement provides a sharp contrast to the reverse-scored statements 5 and 6, which advocate social programs that benefit the poor and disadvantaged. As these items illustrate, those who support the persistence of economic inequality feel it is acceptable for the relatively affluent to receive financial assistance in the form of family gifts and inheritances, but unacceptable for the poor to receive assistance through government redistribution.
It is apparent from the above discussion that Altemeyer’s [23] measure of ideological support for the persistence of economic inequality reflects support for laissez-faire capitalism and opposition to government assistance for the needy, which are hallmarks of political–economic conservatism and neoliberalism. Consequently, it should come as no surprise that support for economic inequality is closely associated with these constructs, as documented by Azevedo et al. [9]. These authors, p. 2 [9], note that ‘All these ideological forms project immense confidence in market forces to solve social as well as economic problems and reflect deep suspicion of governmental efforts to regulate business and to provide for the welfare of its citizens.’ Thus, those who more strongly endorse the persistence of economic inequality should favor a largely unregulated market system that makes few concessions to those at the bottom of the economic pyramid. Individuals who possess such ideological beliefs should be likely to question the legitimacy of government redistribution through taxation, and arguably be more prone to tax evasion.
There are other reasons to believe that individuals who endorse the persistence of economic inequality will oppose government redistribution of income and question the legitimacy of tax systems. Because support for inequality is an integral component of a social dominance orientation, such individuals should be strongly motivated to improve their own economic and social status and the status of their in-group but have relatively little empathy for members of out-groups [20,23]. Additionally, because social dominance is positively correlated with racism and most minority groups in the U.S. disproportionately fall into lower income brackets [20], social dominators should tend to resent government policies they believe will benefit minorities at their personal expense.
Indeed, it is well-documented that socially dominant groups and individuals often engage in the marginalization or vilification of minority groups to rationalize social and economic injustice. Suddaby et al. [10] provide historical illustrations of how carefully chosen forms of “talk” may create and sustain inequality. They discuss influential works such as Malthus’ Essay on Population and Ayn Rand’s Atlas Shrugged which helped transform the collective perception of inequality from a failing of socioeconomic systems to a failing of individuals, p. 386 [10]. They discuss how dominant groups have historically used rhetorical strategies to disseminate myths that portray minorities as ‘avaricious, lazy… and [possessing] a host of other shortcomings that further served to separate minorities, while simultaneously promoting myths that exaggerated the homogeneity of the dominant group’, p. 390 [10]. They note that ‘At this particular moment in history, we seem to be fixated on wealth as the focal variable for determining inequality… The absence of wealth is a signifier for a host of related sins—a presumed lack of motivation, ambition or ability’, p. 390 [10].
Individuals who prefer to maintain or increase current levels of inequality should be particularly susceptible to myths regarding the personal failings of members of out-groups (minorities or the poor) due to motivated social cognition, or the tendency to perceive social situations in a way that serves individual psychological needs [44,45]. Thus, it seems quite likely that supporters of inequality will believe that a portion of their tax dollars are used to benefit “undeserving” groups. Such beliefs should increase their opposition to government redistribution with an attendant increase in the likelihood they will engage in tax evasion. Based on the foregoing arguments, we propose the following hypothesis.
Higher levels of ideological support for economic inequality will be associated with a greater likelihood of committing tax evasion.
1.3. Implications of Machiavellianism
The Machiavellianism construct has been influential in the social science and business literatures for many years. In their widely cited early work, Christie and colleagues [24,46,47] identified three key characteristics of Machiavellianism: (1) a high propensity for the use of manipulation and deceit; (2) a tendency to view others as cowardly, weak and easily manipulated; and (3) a low level of concern with conventional standards of morality and ethics. A variety of other antisocial traits have also been associated with Machiavellianism, such as selfishness, greed, callousness, low levels of empathy, and low levels of guilt or remorse for one’s actions [25]. High levels of Machiavellianism are associated with a tendency to distrust others and cynically view others as manipulative and primarily motivated by self-interest [24,48,49,50].
A case can clearly be made for the existence of a positive association between support for economic inequality and Machiavellianism. Indeed, as noted by Altemeyer [23] social dominators, like high Machiavellians, tend to be manipulative and have relatively little concern for others. This would be expected, given that social dominators are motivated to achieve high social and economic status at the expense of others. Altemeyer [23] also concluded that supporters of economic inequality are motivated primarily by preferences for unjust social and economic systems they believe will benefit them personally. This is consistent with the greed, selfishness and low levels of concern for morality associated with high Machiavellians. The desire to maintain socioeconomic systems that promote inequality and injustice is also consistent with the callousness and lack of guilt or remorse often displayed by high Machiavellians [25]. In addition, the general lack of concern for the plight of the poor and disadvantaged exhibited by social dominators is consistent with the Machiavellianism tendency to view others as dishonest, manipulative and deserving of their fate [24,48,49,50,51]. For instance, supporters of sustained economic inequality may feel that individuals receiving government assistance are making dishonest or exaggerated claims that allow them to “live off the government.” Such cynical views of human nature are characteristic of high Machiavellianism.
We also propose that individuals who support economic inequality and closely related ideologies such as political–economic conservatism and neoliberalism will tend to become more Machiavellian as a result. Machiavellianism has traditionally been viewed as a personality trait or disposition that is an antecedent of manipulative and unethical behavior (e.g., [24,25,52]). However, it is apparent that personal experiences as well as adherence to certain beliefs or ideologies may lead individuals to become more or less Machiavellian over time. Indeed, psychologists have suggested that Machiavellianism is rooted in both genetic (nature) and environmental (nurture) influences (e.g., [53,54]). Some recent business studies also support the conception of Machiavellianism as a malleable personality characteristic. Valentine and Fleischman [55] argue that negative experiences or influences may lead individuals to develop more abrasive, insensitive and ruthless personality traits that are characteristic of high Machiavellians. Their empirical results support this argument, indicating that in a sample of U.S. sales and business employees those who reported experiencing more workplace bullying in the past exhibited higher levels of Machiavellianism. Ruiz-Palomino, Bañón-Gomis and Linuesa-Langreo, p. 186 [56] proposed that ‘in contexts where peers behave unethically…Machiavellian mindsets might be forged.’ Consistent with this contention, they documented that employees’ who had been exposed to more unethical peer behavior possessed higher levels of Machiavellianism, which in turn were associated with intentions to engage in similar behavior.
Given that Machiavellianism appears to be a somewhat malleable personality trait, there are reasons to believe that individuals who possess high levels of support for economic inequality and related ideologies may develop higher levels of Machiavellianism. As noted by Suddaby et al. [10], processes of indoctrination and socialization are largely responsible for the adoption of such ideologies. These processes themselves are likely to pressure many people to adopt more cold-hearted and ruthless mindsets. Once adopted, belief systems such as social dominance and support for inequality often become deeply held convictions and important components of self-identify [20,23]. Yet they are clearly controversial, often contested, and viewed by many as immoral and antisocial [9,20]. In the face of overwhelming and accumulating evidence of the negative socioeconomic consequences of such ideologies (e.g., [11,12,13,57,58,59,60]), the adoption of stronger Machiavellian tendencies seems likely as a means of rationalizing and reinforcing one’s beliefs.
Indeed, many of the criticisms of authors who have highlighted the escalating social problems created by inequality may be viewed as ideologically driven and Machiavellian in nature. Piketty’s [13] work in particular has been widely critiqued across the social sciences. Scholarly critique of a work of such broad import is obviously desirable and necessary for the advancement of scientific knowledge and understanding. Yet, many of the criticisms of Piketty [13] seem rooted in ideological support for the maintenance of existing social, political and economic systems that have contributed to growing levels of inequality. (We prefer to refrain from citing particular papers on this point because we do not want to explicitly criticize individual authors or suggest that their work is ideologically biased. Interested readers can easily locate a large number of articles that criticize Piketty’s [13] work from multiple perspectives by searching academic databases or the Internet.) Piketty’s [13] adoption of inequality as his primary focus has been denounced, and the practical significance of inequality has been questioned. Scholars have argued that there is nothing inherently wrong with the rich accumulating larger amounts of capital, provided the poor are not becoming poorer or living in abject poverty. Others have argued that Piketty’s [13] work has little practical significance because his proposals for wealth taxes and higher marginal income tax rates are simply untenable in most contemporary societies. Such arguments reflect rationalizations that attempt to discredit Piketty’s [13] work and policy prescriptions as misguided, trivial or impractical. In particular, suggestions that it is acceptable for power elites to control most of the world’s accumulated capital while the poor struggle simply to maintain a decent standard of living clearly reflect low levels of caring and empathy for others and a lack of concern for moral issues such as social and economic justice. These are widely recognized markers of Machiavellianism; consequently, we suggest that such arguments effectively illustrate the proposed relationship between ideological support for economic inequality and Machiavellianism.
There are many other examples of the use of carefully crafted and manipulative arguments to justify support for inequality and closely related ideologies such as neoliberalism. As observed by Azevedo et al. p. 12 [9] ‘At worst, neoliberalism is a victim-blaming ideology that treats poor people as responsible for their own poverty and, in doing so, exonerates social, economic and political institutions and arrangements under capitalism. In this sense, it is a system-justifying ideology par excellence.’ Victim-blaming is a classic strategy for rationalizing inequality consistent with Machiavellian arguments that the poor possess many shortcomings and consequently are deserving of their position in society cf. [10]. Piketty [59] also provides many historical examples of how political regimes have rationalized the existence of inequality using manipulative and deceitful arguments. Based on the foregoing discussion, we propose the following hypothesis.
Greater support for economic inequality will be associated with higher levels of Machiavellianism.
An argument may also be made for the existence of a positive relationship between Machiavellianism and tax evasion. High Machiavellians’ predisposition for manipulation and deceit make them more likely to engage in a wide range of unethical and illegal practices, a fact that has been extensively documented in the business [61,62,63,64,65] and accounting [66,67] literature.
Because high Machiavellians have relatively little concern for conventional social norms of morality or ethics, they tend to focus more on “winning” in social contexts [47]. As noted by Simmons, Shafer and Snell [68], an obsession with winning or gaming the system through guile and deceit should be relatively easy to rationalize in financial or business contexts, where the importance of ethics and social responsibility is often contested and the outcomes are closely tied to personal success and economic well-being. Indeed, Shafer and Simmons [69] documented a direct positive association between Machiavellianism and tax avoidance among professional tax advisors in Hong Kong. Shafer and Wang [27] found a similar association using a sample of U.S. taxpayers. We aim to replicate the latter finding in the current study, as indicated in the following hypothesis.
Higher levels of Machiavellianism will be associated with a higher likelihood of committing tax evasion.
Based on Hypotheses 2 and 3, we anticipate that support for inequality will be indirectly associated with tax evasion intentions through its association with Machiavellianism, while Hypothesis 1 suggests that support for inequality will also have a direct positive association with evasion intentions. Thus, we anticipate that Machiavellianism will partially mediate the relationship between support for economic inequality and tax evasion, as indicated in the following hypothesis and illustrated in Figure 1.
Machiavellianism will partially mediate the relationship between support for economic inequality and the likelihood of committing tax evasion.
2. Research Methods
2.1. Sample
To test the hypotheses, we conducted an online survey using Qualtrics. The instrument was distributed by Qualtrics to a random sample of U.S. taxpayers who had recently filed tax returns. Qualtrics assures that the sample is random and representative. In the cover letter, participants were informed that completion of the survey was voluntary, that the survey was anonymous and the researchers would not be able to determine individuals’ identity, and that the results would only be disclosed in a summarized statistical format.
A total of approximately 500 participants completed the survey. After deleting incomplete responses and those who failed consistency checks, the final sample was 269. Table 1 provides a summary of demographic information for participants. Because our participants played the role of small business owners in the tax evasion scenario, we evaluated the comparability of the sample with U.S. small business owners. Although the data are not directly comparable due to the use of different cutoff points, the distribution of our sample by age appears to provide a rough approximation of the population of small business owners based on information reported by the U.S. Small Business Administration [70] Approximately 53 percent of our sample was over the age of 45, whereas 51 percent of U.S. small business owners are over the age of 50. The remaining 47 (49) percent of our respondents (U.S. small business owners) were below 45 (50), and in both our sample and the population, the distribution appears to taper off rapidly among the lower age ranges. For example, only 4.5 percent of our sample was below the age of 25, and only 16 percent of the population of small business owners is below the age of 35 [70]. Slightly over 70 percent of our participants were female. Given that the sample was randomly selected from the general population of taxpayers, this latter result suggests that females are more conscientious regarding participation in surveys of this type. However, as discussed later, gender was not significantly associated with tax evasion intentions. Over 60 percent of respondents indicated that they file their own tax returns, and almost 40 (75) percent had over 30 (10) years of filing experience. These results indicate that our sample consisted predominantly of highly experienced taxpayers who have experience in filing tax returns.
2.2. Survey Instrument
We adopted a brief tax evasion scenario similar to those used in several previous studies [35,71,72]. Participants were asked to play the role of a small business owner faced with a decision of whether to fraudulently overstate their deduction for the business use of a personal car (see Appendix A). Respondents estimated the likelihood they would overstate the deduction on a seven-point Likert scale where higher numbers indicated a greater probability of committing fraud.
Following much previous research, we used the Mach IV instrument to measure Machiavellianism. Responses were provided on a seven-point scale with higher numbers indicating stronger Machiavellian tendencies. We also adopted eight items from the Social Inequality Scale developed by Altemeyer [23] that measure attitudes toward economic inequality. Consistent with other measures, we used a seven-point scale where higher numbers indicate greater support for inequality.
To control for potential social desirability bias, we adopted the Balanced Inventory of Desirable Responding (BIDR) impression management scale, which contains 20 propositional statements [73] measured on seven-point scales where seven indicates more favorable or socially desirable responses. The propensity for impression management is measured by counting the number of extreme responses, including responses of six or seven, to this scale. For each participant, the total number of extreme responses serves as the measure of their tendency to portray themselves in a more favorable light.
3. Results
3.1. Preliminary Analysis
We performed an exploratory factor analysis for the eight items that measure attitudes toward economic inequality. As presented in Appendix B Panel A, the eight items loaded on two distinct factors. The first factor included four reverse-coded items that, as written, indicated opposition to inequality (labeled “opposition”). The other four items indicate support for inequality (labeled “support”). With one exception, the eight items all had factor loadings in excess of 0.8. Both factors had strong internal reliability, with coefficient alpha values above 0.80. Based on these findings, we constructed two separate subscales for attitudes toward economic inequality (opposition and support).
We also conducted an exploratory factor analysis for the Machiavellianism scale, finding that eight of the original items had factor loadings above 0.5 on a single dimension. The remaining items all had factor loadings below 0.5 and were thus deleted. Our eight-item scale (see Appendix B Panel B) had reasonably strong internal reliability, with a Cronbach’s alpha of 0.80.
3.2. Descriptive Statistics and Correlations
Table 2 presents descriptive statistics and correlations for the variables of interest. Participants estimated an average intention of committing tax evasion of 2.5 on the seven-point scale. The mean value for Machiavellianism was 3.4, slightly below the midpoint of the scale. The mean level of support for inequality was somewhat lower at 3.1. The Pearson correlation coefficients among the variables are based on two-tailed tests. As anticipated, Machiavellianism and support for economic inequality were positively correlated with intentions of committing tax fraud, and these relationships were highly significant. However, the opposition to inequality factor was not associated with fraud intentions and consequently was eliminated from further analysis. Machiavellianism was also significantly correlated with support for inequality. As anticipated, participants who supported economic inequality tended to be more Machiavellian. The impression management measure was negatively associated with intentions to commit fraud, indicating that participants’ reported intentions were biased to portray themselves as more ethical. Age and years of filing experience were also negatively correlated with tax evasion intentions, suggesting that older and more experienced taxpayers tend to be more ethical. Due to the fact that age and filing experience were highly correlated (r = 0.759), we eliminated filing experience from our multivariate analysis to avoid possible multicollinearity issues. Somewhat surprisingly, gender was not significantly associated with tax evasion intentions.
3.3. Mediation Analysis
To test our hypotheses, we conducted mediation analysis using the Hayes [74] SPSS PROCESS macro. Table 3 summarizes the results of a mediation model with support for inequality as the independent variable and participants’ reported intentions to commit tax fraud as the dependent variable. Machiavellianism is included as a mediating variable. Impression management, age and gender are included as controls.
Both regression models in the mediation analysis were highly significant and explained a significant portion of the variation in the dependent measures. The model with Machiavellianism as the dependent measure indicates a highly significant positive association with support for economic inequality, which provides strong support for Hypothesis 2. These results suggest that individuals who have ideological preferences for sustained economic inequality tend to be selfish, manipulative, and have relatively little empathy for others. Age and gender had significant negative associations with Machiavellianism, indicating that older and female participants were less Machiavellian. Impression management also had a significant negative association with Machiavellianism, suggesting that participants who were more prone to impression management tended to portray themselves as less Machiavellian.
The full model with tax fraud intentions as the dependent measure indicates that, contrary to Hypothesis 1, greater support for inequality did not have a significant direct association with the likelihood of fraud. Consistent with Hypothesis 3, Machiavellianism had a highly significant positive association with intentions to commit fraud. Thus, the relationship between support for economic inequality and tax evasion appears to be fully mediated by Machiavellianism. Age had a significant negative association with fraud, consistent with the correlation results, indicating again that older taxpayers appear to be more ethical. However, gender was not associated with fraud intentions. Impression management had a significant negative association with fraud intentions—as would be expected individuals who were more concerned with presenting themselves in a favorable light tended to bias their reported fraud intentions downward. (The strength of the correlations between impression management and Machiavellianism raises some concern regarding multicollinearity in the regression model. To address this issue, we ran a separate linear regression model and examined the variance inflation factors (VIFs) for the independent variables. None of the VIFs in this model significantly exceeded 1.0, thus mitigating concerns regarding possible multicollinearity.).
To formally test the direct and indirect effects of support for inequality on tax evasion intentions, following Hayes [74] we constructed 95 percent bias-corrected confidence intervals based on 5000 bootstrapped samples. The use of bootstrapping is preferred over techniques such as Sobel tests because bootstrapping makes no assumptions regarding the sampling distributions of the measures. The confidence interval for the direct effect of support for inequality on tax fraud straddled zero, again indicating that this association was not significant at conventional levels. In contrast, the confidence interval for the indirect effects of support for inequality on fraud intentions through Machiavellianism did not straddle zero, indicating that we can be 95 percent confident that support for inequality is associated with fraud intentions through its relationship with Machiavellianism. A summary of the empirical results is provided in Figure 2.
4. Conclusions and Discussion
Our study is subject to a number of limitations that should be kept in mind when interpreting the results. The cross-sectional survey design does not provide a sound basis for inferring cause-and-effect relationships. This is an inherent limitation of research of this nature, because it would be quite challenging to develop meaningful experimental manipulations of fundamental ideological viewpoints and personality traits such as those investigated herein. Our sample was randomly chosen; however, there is always a potential for nonresponse bias in surveys. This concern is mitigated somewhat by the fact that our respondents were approximately representative of the population of U.S. small business owners in terms of age, because age and years of filing experience (which were highly correlated) were the only demographic variables that were significantly associated with tax evasion intentions. Approximately 70 percent of participants were female; thus, the sample is not representative in terms of gender breakdown. However, gender was not associated with evasion intentions.
As anticipated, the findings document strong positive associations among ideological support for economic inequality, Machiavellianism and intentions to commit tax evasion. Based on mediation analysis, support for inequality was significantly associated with Machiavellian tendencies, which in turn were strongly associated with intentions to engage in tax evasion. However, contrary to our expectations, the direct relationship between support for inequality and evasion intentions was not significant in the mediation model. Consequently, this association appears to be fully mediated by Machiavellianism. The significant association between support for economic inequality and Machiavellianism suggests that ideologies may lead individuals to develop a variety of immoral and antisocial attitudes to justify or rationalize their preferred belief systems.
The finding that support for inequality was positively associated with a higher self-reported likelihood of tax evasion is revealing, suggesting that such ideological views are strong enough to motivate taxpayers to engage in illegal activities to minimize the redistribution of their income to less advantaged members of society. Prior studies of tax evasion have not addressed this association; thus, the current study provides a unique contribution to our understanding of the motivations that underly tax evasion.
We acknowledge that our research focuses on a problem that will be challenging to address. At the societal level, it is well known that power elites and governments may employ indoctrination and socialization techniques to encourage the adoption of certain preferred socioeconomic ideologies such as support for inequality and neoliberal capitalism. As observed by Suddaby et al. p. 837 [10], ‘Through socialization (i.e., talk of what is acceptable, normal and valued in a society) we gradually begin to see inequality as natural and maybe even a desirable state of the human condition.’ Consequently, inequality may become institutionalized, taken for granted and hardly questioned. It will obviously be difficult to counter such entrenched narratives in the short term. Of course, changes in the balance of political power occur periodically, and governments with more prosocial orientations may be able to attenuate prevailing ideologies through the use of public education campaigns that promote more egalitarian views and emphasize citizens’ duty to pay their fair share of taxes. We feel that ongoing scholarly work that increases awareness and understanding of the issues involved is also an important step. Suddaby et al. [10] emphasize that scholars should be very careful in how they choose to talk about such issues, noting that over the past 150 years, discussions of the potentially horrific consequences of poverty have gradually morphed into relatively sanitized discussions of “inequality”. Thus, if scholars continue to document and more explicitly discuss the negative socioeconomic consequences of unregulated capitalism and support for inequality, it may increase awareness and help lead to a shift in collective consciousness.
One significant source of indoctrination is undoubtedly business education. Indeed, business schools have been deeply implicated in the promotion of neoliberal capitalism and support for inequality [75,76,77,78]. Khurana [76] explains that the original purpose of modern business education was to create an elite class of busines and management practitioners bound by professional/ethical ideals regarding the role of business in society, but this purpose was largely subverted by the increasingly dominant forces of neoliberalism over the ensuing decades. These trends have culminated in a situation where, despite the presence of a few business ethics courses in most curricula, serious consideration of the interests of stakeholders such as workers, consumers and society are often neglected or portrayed as a threat to business success. Fotaki and Prasad [75] urge business educators to counter this trend by incorporating more effective techniques of raising awareness of the negative consequences of unregulated capitalism.
Future research on the effects of socioeconomic and political views on taxpayer behavior should be expanded to examine the influence a broader range of ideologies and related attitudes. For instance, because support for inequality is a key component of a social dominance orientation [23], our findings imply that prejudice may significantly influence tax evasion decisions. Social dominators are known to possess relatively high levels of prejudice against individuals whom they perceive to be outside their social group, coupled with a strong desire to dominate such outsiders [20,23]. This suggests that if social dominators define their in-group based primarily on economic status, they are likely to be prejudiced against the poor and disadvantaged and consequently may engage in tax evasion to reduce the amount of their income subject to redistribution. However, social dominance exists throughout income distributions [20], which implies that some individuals define their in-group primarily based on other considerations such as race. If social dominators possess high levels of racism toward minority groups, their anti-egalitarian impulses combined with stereotypical perceptions of certain minorities as unwilling to work hard to improve their socioeconomic status may motivate them to oppose income redistribution and engage in tax evasion. (We did not gather information on factors such as participants’ perceived economic/social status or race because we viewed these as potentially sensitive issues. We acknowledge this as a limitation of our study because such factors may be associated with support for economic inequality.)
This discussion applies not only to taxpayers and business owners/managers, but also to professional tax advisors (accountants and attorneys). To the extent that socioeconomic ideologies such as political–economic conservatism, neoliberalism and support for inequality are present among tax professionals, they should arguably be more likely to aid and abet their clients’ desires to engage in noncompliance. Sikka [3,4] has highlighted the deep involvement of professional accountancy firms in facilitating tax avoidance/evasion by their clients and emphasized how such socially irresponsible behavior has contributed to worsening economic inequality. We suggest that economic elites who support the persistence of inequality and the maintenance of unjust social and economic systems should be likely to seek out and cultivate relationships with like-minded tax professionals (and vice versa), resulting in a relatively high likelihood of noncompliance. However, these issues have not been addressed by previous research.
More broadly, research on ideological support for inequality and closely related belief systems such as political–economic conservatism and neoliberalism has the potential to contribute to our understanding of the creation and perpetuation of socioeconomic conditions that are unjust and unsustainable. To better understand the role of organizations in contributing to inequality, Bapuji et al. [2] proposed a general framework that focuses on value creation, appropriation, and distribution. They encourage researchers to focus on the issue of value distribution, noting that if the distribution is skewed toward a few claimants, it will lead to greater inequality. The claimants to the value created by organizations include employees (with a distinct category for executives), shareholders, governments and society (unidentified stakeholders). Bapuji et al. [2] discuss the management actions with respect to value distribution that have contributed to the current high levels of inequality in neoliberal societies, such as minimizing union power and wages paid to lower level employees while awarding lavish compensation to executives, distributing increasing amounts of value to shareholders, lobbying for tax incentives and engaging in tax avoidance/evasion, and minimizing corporate philanthropy or deploying philanthropic distributions only in a strategic fashion to advance or protect organizational interests.
This framing helps place tax evasion in context as only one aspect of organizational efforts to minimize value distributions to governments. There are obviously many other areas of organizational decision making that are likely to be influenced by ideological belief systems such as support for inequality. Some recent work supports this contention. For example, Gupta and Wowak [79] found that organizations with more politically conservative boards tend to give larger compensation packages to CEOs than those with more politically liberal boards. Chen, Hambrick and Treviño [80] found that CEOs’ political ideologies (measured by political donations) influenced their firms’ corporate social responsibility (CSR) practices. Among other findings, they documented that more politically conservative CEOs place less emphasis on CSR and are more likely to curb CSR efforts when financial performance declines. The results of such studies indicate that political conservatism among boards and CEOs is associated with antisocial behavior that is likely to increase income inequality and threaten sustainability. However, much more work could be done in this area. For instance, future studies could examine the influence of a broader range of ideological belief systems on boards’ and executives’ propensity to engage in a variety of value distribution decisions that contribute to inequality, such as those outlined by Bapuji et al. [2].
Author Contributions
Conceptualization, W.E.S.; methodology, W.E.S., Z.W., T.-S.H.; validation, W.E.S., Z.W., T.-S.H.; formal analysis, Z.W., T.-S.H.; investigation, Z.W., T.-S.H.; resources, W.E.S.; data curation, Z.W., T.-S.H.; writing—original draft preparation, W.E.S.; writing—review and editing, W.E.S., Z.W., T.-S.H.; visualization, Z.W., T.-S.H.; supervision, W.E.S.; project administration, W.E.S. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Conflicts of Interest
The authors declare no conflict of interest.
Appendix A
CASE SCENARIO
Assume you have your own small business, and you own a car that you use for both business and personal reasons. You prepare your own tax return. The tax laws provide that automobile expenses are deductible to the extent the automobile is used for business purposes. In preparing your income tax return, you determine that the automobile was used 60% for business. However, you also calculate that if you FALSELY claimed it was used 80% for business, your deduction would increase by USD 3000, and you would save USD 700 in taxes. This is a significant amount of your total tax liability.
Appendix B
SCALE ITEMS
Table A1
Panel A: Attitudes toward economic inequality (α = 0.85) (α = 0.82).
| Items | Factor 1 | Factor 2 |
|---|---|---|
| Economic equality is a bad idea dreamed up by muddle-headed “do-gooders.” It would be a big mistake to pursue it. | 0.814 | |
| Although it may be ultimately impossible, economic equality is a very worthy goal that we should definitely strive for.* | 0.808 | |
| People have no right to economic equality. All of us should get as much as we can, and if some don’t get enough, that’s their problem. | 0.838 | |
| If the natural forces of supply and demand and power make a few people immensely wealthy and millions of others poor, so be it. | 0.825 | |
| Everyone should have an equal opportunity for economic success. Those born into poor circumstances should be given extra help to make the “playing field” level for them.* | 0.855 | |
| “Access programs” to higher education, which give people from poor backgrounds extra financial support and counseling while in university, are a good idea.* | 0.842 | |
| Nobody should get extra help improving his place in society. Everyone should start off with what his family gives him, and go from there. | 0.754 | |
| Tax money should be used to make sure everyone has an adequate standard of living.* | 0.807 |
* Reverse coded.
Table A2
Panel B: Machiavellianism (α = 0.80).
| Items | Factor Loadings |
|---|---|
| Never tell anyone the real reason you did something unless it us useful to do so. | 0.628 |
| The best way to handle people is to tell them what they want to hear. | 0.717 |
| It is safest to assume that all people have a vicious streak and it will come out when they are given a chance. | 0.549 |
| Generally speaking people won’t work hard unless they are forced to do so. | 0.528 |
| The biggest difference between most criminals and other people is that the criminals are stupid enough to get caught. | 0.503 |
| It is wise to flatter important people. | 0.638 |
| It is hard to get ahead without cutting corners here and there. | 0.504 |
| Most people forget more easily the death of a parent than the loss of their property. | 0.504 |
Figures and Tables
Table 1
Demographic information.
| Age | Frequency | Percent |
| <25 | 12 | 4.5 |
| 25–44 | 114 | 42.4 |
| 45–64 | 88 | 32.7 |
| >65 | 55 | 20.4 |
| Total | 269 | 100.0 |
| Gender | Frequency | Percent |
| Male | 79 | 29.4 |
| Female | 190 | 70.6 |
| Total | 269 | 100.0 |
| Filing Experience in Years | Frequency | Percent |
| <5 | 33 | 12.3 |
| 6–10 | 33 | 12.3 |
| 11–15 | 36 | 13.4 |
| 16–20 | 18 | 6.7 |
| 21–25 | 22 | 8.2 |
| 26–30 | 24 | 8.9 |
| >30 | 103 | 38.3 |
| Total | 269 | 100.0 |
| Prepare Own Return | Frequency | Percent |
| No | 107 | 39.8 |
| Yes | 162 | 60.2 |
| Total | 269 | 100.0 |
Descriptive statistics and correlation coefficients.
| Mean |
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | |
|---|---|---|---|---|---|---|---|---|---|
| 1.Likelihood of fraud | 2.5 |
1 | |||||||
| 2. Machiavellianism | 3.4 |
0.428 ** | 1 | ||||||
| 3. Impression management | 7.4 |
−0.350 ** | −0.326 ** | 1 | |||||
| 4. Support for inequality | 3.1 |
0.221 ** | 0.403 ** | −0.070 | 1 | ||||
| 5. Opposition to inequality | 3.0 |
−0.009 | −0.063 | −0.093 | 0.505 ** | 1 | |||
| 6. Age | −0.261 ** | −0.289 ** | 0.164 ** | −0.003 | 0.132 * | 1 | |||
| 7. Gender | −0.085 | −0.137 * | 0.106 | −0.041 | 0.043 | −0.023 | 1 | ||
| 8. Years of filing experience | −0.291 ** | −0.304 ** | 0.196 ** | −0.041 | 0.133 * | 0.759 ** | −0.013 | 1 | |
| 9. Prepare own return | 0.025 | 0.105 | −0.107 | 0.101 | 0.108 | −0.144 * | −0.090 | −0.111 |
** Correlation is significant at the 0.01 level (two-tailed). * Correlation is significant at the 0.05 level (two-tailed).
Table 3Mediation analysis.
| Consequent | ||||||
|---|---|---|---|---|---|---|
| Machiavellianism | Tax Evasion | |||||
| Antecedent | Beta | SE | P | Beta | SE | P |
| Constant | 3.95 | 0.25 | 0.000 | 2.11 | 0.58 | 0.000 |
| Machiavellianism | ----- | ----- | ----- | 0.44 | 0.10 | 0.000 |
| Support for inequality | 0.29 | 0.04 | 0.000 | 0.12 | 0.07 | 0.268 |
| Impression management | −0.06 | 0.01 | 0.000 | −0.08 | 0.02 | 0.000 |
| Age | −0.32 | 0.06 | 0.000 | −0.31 | 0.12 | 0.009 |
| Gender | −0.24 | 0.12 | 0.048 | −0.09 | 0.21 | 0.664 |
| Model F (p-value) | 31.0(0.000) | 18.2(0.000) | ||||
| Adjusted R Square | 0.32 | 0.26 | ||||
© 2020 by the authors.