Tax evasion is a world-wide problem. This paper explores this phenomenon from an economic and a social psychological perspective. Both theoretical and empirical studies are reviewed to discover the underlying assumptions, decisions and countervailing efforts. Linkages between the level of tax evasion and the deterrent methods used by tax authorities are explored. Conclusions are made as to the most effective ways of endevouring to reduce the incidence of tax evasion.




CONTACT: Prof N Ronan, E:MAIL,
National University of Lesotho
Department of Business Administration
Roma box 180,



Taxation is a major source of public finance, and the outflow of public expenditure is the provision of various public/social services by the state for the benefit of citizens and non-citizens within its jurisdiction. The principal reason that governments levy taxes is to raise revenue to provide resources for the provision of such services, in turn presumably motivated by the wish to promote outcomes such as reduced poverty, maintenance of law and order and higher living standards. Other reasons include redistribution of income and wealth, economic regulations and macroeconomic stabilization.
There are various types of taxes but all of them are usually classified as direct and indirect taxes. This classification is mainly based on the incidence of the particular taxes. Direct taxes are more visible to the taxpayer and they include such taxes as income tax, estate duty and property tax levied on the person (whether individual or non individual) with no possibility of shifting the incidence to another person. Indirect taxes are generally taxes on consumption and outlay. They are often less visible to the taxpayer and collected by an intermediary. Because tax is levied on goods and services, the tax yield depends directly on the level of consumption of the population on the particular taxed commodity or service or other outlay.
The payment of tax is compulsory in nature. This does not mean that all tax revenue is paid unwillingly, but the will of the taxpayer is legally immaterial. Although some people pay their taxes out of a desire to do good for their governments, and the fear of the consequences of non-payment, majority of people do not enjoy paying taxes, regardless of the benefits they may realize from taxation to the society generally and to themselves. Such people are often referred to as “free riders”. Slemrod and Bakija (2001) noted that each citizen has a very strong incentive to ride free on the contributions of others because one’s own individual contribution is relatively immaterial to what one gets back from the government. For these reasons, paying taxes must be made a legal responsibility for each citizen. In this connection, most governments are faced with a great challenge in enforcing tax laws to induce people to pay their taxes. Some dutiful people will undoubtedly pay what they owe, but others would not. Consequently, over time the ranks of the dutiful will shrink as they realize how they are being taken advantage of by others.
Tax evasion is a global phenomenon. Recently, it has attracted a great deal of official and public interests, but it is difficult, if not impossible for tax authorities to eliminate it entirely. For example, in the United States, failure to pay one’s taxes timely is a civil offence, subject to a variety of penalties. Fraud can expose the citizen to criminal charges and jail sentences, although this is a very rare occurrence. Even in the face of those penalties, substantial evasion persists, [Slemrod and Bakija (2001)]. Nonetheless, most governments are continuously striving to reduce this problem of tax evasion because without taxation survival of any government is at stake. Therefore, there is a need to properly understand tax evasion dimensions and develop a sound strategy to tackle it.
This paper aims to explore the ramifications and phenomenon of tax evasion. It is discussed from both an economic and social psychology perspective.:

Tax evasion is not a new phenomenon; it has been in existence for a long time and still continues to impose growing challenges on tax authorities and governments. It raises issues relevant to a broad range of social science – accounting, psychology, political science, and sociology – and it is of enormous importance in policy design [P. Webley, H. Robben, H. Elffers and D. Hessing (1989)]. A common definition of tax evasion is the minimization of one’s tax liability by methods that violates the provisions of the tax codes. It is therefore an offence that if discovered, could lead to the imposition of criminal proceedings against the taxpayer.
In legal terms there is a distinction between tax evasion and tax avoidance. Tax avoidance involves every attempt by legal means to use loopholes in order to minimize ones tax burden. However, from economic point of view, and focusing on the consequences of both, in terms of tax collection, evasion and avoidance are similar even if one is illegal and the other one is not. In most countries there are many perfectly legal tax avoidance strategies for arranging financial affairs to minimize taxes. However, these strategies can often provide attractive alternatives for tax evasion. Most taxpayers contract the services of tax consultants to avoid taxes. But most tax consultants tend to exploit uncertainty where it exists, hence engaging in tax evasion. This suggests that professional advice is inextricably linked with the practice of tax evasion. Taking measures against tax evasion without introducing corrective measures against professionals who encourage evasion, is insufficient to enlarge the tax base.

It is also vital to draw a distinction between tax evasion and noncompliance. Most writers often use the term of noncompliance to characterize the intentional or unintentional failure of taxpayers to pay their taxes correctly. David J. Pyle (1989) describes non-compliance as a more neutral term than evasion since it does not assume that an inaccurate tax return is necessarily the result of an intention to defraud tax authorities, and it recognizes that the inaccuracy may only result in overpayment of taxes. In evading taxes one is knowingly breaking the law, and there are social and psychological consequences such as stigma and guilt attached. However, in practice, it is difficult if not impossible to be certain of intent and as such the term noncompliance is often used.
It is only recently that economists have begun to construct theoretical models of tax evasion decision. The first rigorous theoretical analysis was made by Allingham and Sandmo (1972), based on the work of economics of crime [Becker (1968) and second, the work on the economics of risk and uncertainty [Arrow (1970)]. This framework considers man as amoral taxpayer. Based on moral connotation, the assumption is that people will commit an offence, any offence; if by so doing they maximize their utility. This model simply assumes that behavior is influenced by factors such as, tax rate (which determine the benefit of tax evasion) and penalties for fraud and probability of detection (which determine costs). Individuals have a choice of how much income to declare and may report none, some, or all of it. This model generally produced unsurprising results. For instance, an increase in penalty rates and an increase in the probability of detection both result in more income being declared. But simultaneously, with penalties on tax evaded rather than income concealed and decreasing absolute risk aversion, both reasonable possibilities, the model predicts that evasion decreases when the tax rate rises.
On the other hand, Kinsey’s (1984) review suggests that for criminal behavior in general, penalties are less deterrent than probability of being caught, and this pattern is also found in the literature of compliance. This assertion is also supported by Mason and Calvin (1978) report in their survey study that the highest correlation with admitted evasion was the perceived probability of not being caught and many studies have found that evaders and participants in the underground economy perceive a lower probability of detection than others [Vogel (1974), Scott & Grasmick (1981), Grasmick and Scott (1982), Van Eck and Kazemier (1988)].
Using IRS records from 1969, Dubin and Wilde (1988) also found that audits had a deterrent effect. This raises the question of what influences people’s perceptions of the probability of detection. Klepper and Nagin (1989) found that age and itemization of deductions were associated with perceiving lower probabilities of detection overall, but more interestingly, perceived probabilities of detection were also found to vary according to the nature of the specific declaration made. For instance, detection was seen as likely for income subject to third party reporting and more likely if tax was evaded on a large proportion of an item (e.g. charitable deduction).
There have been a variety of extensions for this simple model. For instance, interactive (game-playing) models [Corhon (1984), Benjamini and Maital (1985)] stem from recognition that a taxpayer is not taking decisions in isolation and there are other ‘players’ in the ‘game’. For instance, tax situation is treated as two-person game involving the taxpayer and the revenue authorities. The taxpayer has two choices; either comply or not comply. The revenue authorities also have two choices; they can either investigate the taxpayer or not. Clearly, there is no equilibrium in this model. If the taxpayer is complying it is best for authorities not to waste money in investigating but, if the authorities are known not to be investigating it is best for the taxpayer not to comply. But, there is equilibrium if both parties use mixed strategies. In this situation the probability of evasion increases with marginal cost of investigation and decreases with the size of the penalty.
The reasons why people evade taxes can also be viewed from the social psychological point of view. For psychologists, social psychological variables such as stigma, reputation and social norms have a great impact in taxpayer’s decision to comply or not. Grasmick and Scott (1982) found that while the relationship between the threat of legal punishment (detection probability) and intention to evade taxes in the future was statistically significant, anticipated feelings of guilt and possible social stigma attached to tax evasion were more strongly associated with deterrence. They concluded from their data that policies increasing the public’s sense of moral duty to comply should be the most effective strategy for improving compliance. In a similar vein, Mason and Mason (1992) have drawn on the moral development literature to argue that an appeal to conscience or civic duty, if correctly targeted should improve compliance over and above a fear of sanction or threat. This implies that people will pay taxes because they; are morally committed to, and out of fear of stigma which may involve informal sanctions like ridicule or ostracism.
Community responsiveness to a tax system and tax authority is multidimensional, and changeable, and has much to do with social relationships as with technical and administrative procedures. Braithwaite (2003) defines community responsiveness as the evaluation that individuals or groups make of the tax authority in their community, as well as the actions that taxpayers take in response to the expectations of this authority. The fundamental dimensions of community responsiveness are tax attitudes and tax behavior.
Tax attitudinal kind of responsiveness can be measured in terms of motivational postures; which describe the stance of taxpayers that must be managed when a tax authority seeks to change or wants an explanation for taxpaying behavior. There are five motivational postures that can be identified in the context of taxation compliance; (i) commitment, (ii) capitulation, (iii) resistance, (iv) disengagement and (v) game playing.
Commitment reflects beliefs about the desirability of the tax system. Capitulation reflects acceptance of the tax office as the legitimate authority and the feeling that the office is a benign power as long as one acts properly and defers to its authority. These two postures reflect an overall positive orientation to authority. In contrast, the last three postures are of defiance. Resistance reflects doubt about the intentions of the tax office to behave cooperatively and benignly towards those it dominates and provides rhetoric for calling on taxpayers to be watchful, to fight for their rights, and to curb tax office power. Disengagement is also a motivational posture that communicates resistance, but here disenchantment is more widespread, and individuals and groups have moved beyond seeing any point in challenging the tax authorities. Lastly, game playing is a particular kind of attitude to law which perceives law as something to be moulded to suit one’s purposes rather than as something to be respected as defining the limits of acceptable activity.
Motivational postures are useful markers of degree of consent, cooperation and commitment that underlines the human system as it comes into contact with the administrative/technical tax system. When commitment and capitulation are high, the conditions for introducing measures to improve compliance are optimal. This measure may involve setting up social contexts where tax issues can be contested in a constructive and dialogic fashion, and where tax administrators and citizens can co-design tax systems to make the work better for everyone. However, when defiant postures of resistance, disengagement, and game playing are high, a truce will need to be negotiated in all likelihood before any meaningful attempts at the co-design of the tax system can proceed. Thus, if non compliance has occurred, response by the authority is required. From the perspective of an authority, part of dealing with an individual’s non-compliance is to ensure that it will not happen again, and part is to show the community that compliance standards are high and will be maintained. The challenge for tax administrators is therefore to play a two handed game: to deal with the wrong doing today, while nurturing consent for tomorrow.
Lewis and Cullis (1985, 1988) put forward the concerns of individual and concerns of authorities. From the individual perspective factors such as fiscal attitudes and perceptions (which include the individual’s support for government policies, perceptions of the tax system and burden feelings of alienation and inequity); perceptions of enforcement and opportunity; and characteristics of taxpayers (demographics and personal traits), determine individual decision whether to evade tax or not. As far as the authorities are concerned, three factors are identified. These are government’s fiscal policy and tax enforcement policy (which affect each other) and the policy maker’s assumptions about taxpayers. These assumptions are affected by enforcement structure and themselves weakly influence fiscal policy. The tax enforcement structure also affects the perceived opportunities for evasion. Vogel (1974) also identified the individual’s exchange relationship with the government (tax burdens minus services received) as factors having effect on tax evasion. However, he shows that, knowing others who evade tax is also an important factor.
In general, the above theories, though the authors argue from different perspectives, seem to have more or less one thing in common. Taxpayers behavior to decide as to evade or not is influenced by a variety of factors. To understand tax evasion, authorities have to study such factors in order to deal with evasion accordingly.
On the basis of the tax evasion theoretical theory provided in the preceding section, this section attempts to examine empirical studies conducted by different researchers in understanding why people evade taxes.
While there are several research approaches to understand why people evade taxes, the utility maximizing theory in which taxpayers are engaged in a cost benefit analysis (i.e. by weighing up the risks of detection and punishment for noncompliance against probabilities of successfully evading taxes) is still dominant. Several rigorous econometric studies have been conducted by a number of researchers to test the effects of sanctions variables on tax compliance. First, Witte and Woodbury (1983, 1985), using the 1969 TCMP data base provided by the United States Internal Revenue Service (IRS), confirmed the theoretical results that tax audits have an effect on deterrence. The dependent variable in this study is an IRS measure of voluntary compliance, calculated following a series of audits of a stratified random sample of taxpayers for 1969 TCMP. A large selection of independent variables was used, including before tax income, measures of the probability and severity of punishment, a measure of return complexity, proxies for knowledge of tax laws, awareness of IRS and attitude to government, etc.
The results were found to be strong though they differed markedly by audit class. The principal results were as follows: First, higher overall probabilities of audit were associated with higher compliance rates, except for low-income non-business taxpayers. However, the effects of increased probability of civil and criminal sanctions were not as expected. Where these variables were significant (which was not often) they were generally associated with lower compliance rates. Second, fear of detection (measured by the proportion of taxpayers receiving warning notices) was found to increase compliance. Third, taxpayers with high proportions of labour income in their total income were also found to be more likely to comply with tax regulations than those with higher proportions of non-labour income.
On the other hand, Dubin and Wilde (1988) also studied federal income taxation procedures in the United States. Likewise, their results revealed that there is a deterrence effect associated with increases in the audit rate. Though, their selection of returns for audit differed from that of Witte and Woodbury (1983, 1985), because their estimates suggested that audit rate is an endogenous variable while Witte and Woodbury (1983, 1985) treated it as and exogenous variable, they revealed that there is a significant negative time trend in the audit rate and in compliance. Dubin and Wilde (1988) also demonstrated that any increase in the resources of tax authorities implies an upsurge in audits for all categories of taxpayers. They also undertook a second type of measure in parallel, implying an effort to target those most likely to attempt evasion, such as self employed taxpayers.
In addition to the latter, Beron, Tauchen and Witte (1988) analyzed the effects of repressive action on two alternative forms of tax evasion, evasion by understatement of income and evasion by overstatement of the different authorized deductions. They indicated that, for all audit classes, more extensive auditing decreases tax noncompliance. However, this efficiency was more marked for evasion by overstatement of deduction than by understatement of income and this is essentially middle income classes of taxpayers. Thus depending on their occupation and professional status taxpayers face various opportunities to evade.
In general, some points need to be emphasized about the impact of detection efforts on tax evasion decisions. Whilst the majority of evidence points to a significant deterrence effect of increases in the probability of detection/auditing, there is less evidence that more severe penalties act in the same way. In addition, for an efficiency of specific audit policies, there are certain requirements to be met. First, tax audit has to be correctly targeted. Second, suppressive measures have to be efficient. For example, it is necessary that they are implemented competently so that evasion is revealed and punished otherwise taxpayers will not be impressed and their decision with respect to evasion will not change as their decision is mostly influenced by past audit values. Lastly, the tax authority’s action must be completed by a general action implying the omnipresence of the threat of detection of illegal behavior.
Song and Yarbrough (1978) conducted a sample study in North Carolina in 1975 to explore both attitudinal and behavioral aspects of ethics. Scores for the statements that they received were summed and converted into a percentage. They asserted that scores higher than fifty percent indicate positive tax ethics, and scores below fifty percent indicate negative tax ethics. They found unsurprising results,
because little over two thirds of the respondents revealed positive tax ethics, one fifth had negative tax ethics, and the rest exhibited neutral tax ethicsIt appears that, whilst most people feel that the tax laws should be obeyed, they do not regard violations as serious crime meriting a severe punishment. Simple correlations between scores for tax ethics and various socio-economic, demographic and other variables were also undertaken. These showed that individuals with high levels of income and education exhibited high level of tax ethics. However, people who generally distrusted others, felt alienated, or felt that local politicians/officials would not listen to them generally had low score of tax ethics.
Dean, Keenan and Kenny (1980) also conducted a survey on taxpayers’ attitudes in Fife (Scotland) in 1977. The vast majority of respondents (ninety three percent) thought that ‘income tax in this country is (much/little) too high’ while a substantial proportion (sixty two percent) did not think that the government spends taxpayers’ money wisely. On this latter point many thought there was evidence of considerable waste and inefficiency in the public sector, which combined with what they regarded as incorrect spending priorities, caused dissatisfaction. In addition many respondents felt they were paying too much income tax compared with other people, even when compared with those on similar incomes. This survey also included the questions concerning perceptions of opportunities for evasion and it is interesting that two thirds of respondents thought that all or most taxpayers would exploit an opportunity for small scale evasion if they thought they could get away with it. Likewise, nearly a quarter of respondents thought that all or ost taxpayers would attempt large scale evasion if they felt that it would go undetected
This paper has explored the world-wide phenomenon of tax evasion. It has been estimated that compliance rates among tax-payers in the developing countries may be as low as 50%. Clearly it is a major problem for governments trying to balance budgets while at the same time experiencing severe pressures to social and economic investment.
We have discussed the variety of perceptions which taxpayers bring to the issue of evasion. One of the clear messages to emerge from the discussion is that strong tax auditing and heavy penalties for tax evasion are strong deterrents against tax evasion. But more research is needed into the actual extent of tax evasion. Secondly, we need to know where the significant losses are incurred in tax evasion. It may be that multi-national companies may evade tax due to transfer pricing techniques. It is usually impossible to discover these practices yet there is some suspicion that it is taking place. Also, the informal sector does not usually pay taxes. A number of governments have made soundings to tax this sector. We need accurate information on the extent to which the informal trader would actually have a liability to pay tax , if the tax net was extended to them. We may find the tax take would be relatively small and the costs of collection may exceed the revenue gained. Finally, we may observe that tax evasion is a highly complex phenomenon which requires a multi-disciplinary approach to its study.
Allingham M. G. and A. Sandmo. (1972) “Income Tax Evasion: A Theoretical Analysis”. Journal of Public Economics. 1 (3/4) pp. 323-340.

Becker G.S. (1968) “Crime and Punishment: An Economic Approach”. Journal ofPolitical Economy. 76: pp. 169-217.

Braithwaite Valerie. (2003) Taxing Democracy, Understanding Tax Avoidance and
Evasion. The Australian National University Canberra. Ashgate Publishing Company U.S.A.

Dubin, Jeffrey, Michael Graetz, and Louis Wilde. (1990) “The Effect of Audit Rates on the Federal Individual Income Tax”. National Tax Journal, 43(4) pp. 395-409.

Gramsmick H.G. and Scott W.J. (1982) “Tax Evasion and Mechanism of Social Control. A Comparison with Grand and Petty Theft”. Journal of Economic Psychology. (2) pp. 213-230.

Klepper S. and D.S. Nagin. (1989) “The Anatomy of Tax Evasion”. Journal of Law,
Economics and Organization, 5(1) pp.1-24.

Pyle David J. (1989) Tax Evasion and the Black Economy. London: Macmillan Press LTD.

Slemrod Joel and Jon Bakija. (2001) Taxing Ourselves. A Citizen’s Guide to the Great Debate over Tax Reform, U.S.A: MIT Press.

Song Y. D. and T. E. Yarbrough. (1982) “Tax Ethics and Taxpayer Attitudes Survey”. Public Administration Review.

Webley P. H. Robben, H. Elfers, and D. Hessing. (1991) Tax Evasion: An
Experimental Appraoch –European Monograph in Social Psychology.
Cambridge University Press.

Witte R. D. and Woodbury. (1985) “The Effects of Tax Laws and Tax Administration on Tax Compliance: The Case of The U.S. Individual Income Tax”. National Tax Journal.

Journal Comments