ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846
-A A +A

Role of Trust and Power in Tax Compliance

Evidence from India

Sudhanshu Kumar ( teaches economics at the Indian Institute of Management Bodh Gaya.

The interplay of power and trust in tax compliance in India is analysed by using the slippery slope framework. Its hypotheses are tested using survey data and by employing the ordered logit estimation methodology to determine the role of trust and power in voluntary and enforced compliance. Trust plays a positive, significant role in voluntary compliance; it is highest in a high trust, high power scenario.

Part of the work was completed when the author was associated with the National Institute of Public Finance and Policy (NIPFP), New Delhi. Support to carry out the survey to collect data for this research was provided by the NIPFP. The author gratefully acknowledges the very useful comments received from the anonymous reviewer; Shubham Kumar Verma for his assistance in entering data from questionnaire to the computer; and D P Sengupta, Hippu Salk Kristle Nathan, Sumanjeet, and Santosh Dash for their useful comments at different stages of this research work. The author is especially grateful to all the participants who volunteered to respond to the questionnaire.

Tax compliance can be voluntary or enforced. Voluntary tax compliance depends on the trust people have in their government. Enforced tax compliance depends on the power exercised by tax authorities. The early literature on the subject considered the audit rate, tax rate, and fines as instruments for improving tax compliance. Allingham and Sandmo (1972) and Srinivasan (1973) attempted to set the tax compliance problem within the general theory of criminal behaviour as proposed by Becker (1968). They treated tax evasion as purely an economic decision under risk; hence, they considered, tax compliance can be improved by fine-tuning these

instruments. The literature relying on this framework considers factors related to the tax authorities’ effectiveness in enforcing tax laws and emphasises their power in ensuring tax compliance. However, it is also observed that the decision to comply depends not only on the authorities’ enforcing power but also on motivational orientations, including trust in authorities. It is argued that as long as citizens trust that the government is using tax revenue sensibly, they will pay taxes voluntarily (Feld and Frey 2007). Thus, compliance depends on both tax authorities’ power and trust in government.

The interplay of power and trust in tax compliance can be analysed using the slippery slope framework (SSF) (Alm 2012; Kirchler et al 2008). The SSF analyses compliance by assuming that people might pay their taxes because they are motivated enough, or they are forced to pay, or because they are both motivated and forced to pay. The framework integrates economic and psychological factors, and it is useful in exploring the determinants of tax compliance. The literature suggests that trust increases voluntary compliance; power increases enforced compliance (Alm 2012); and the interactions between tax authorities and taxpayers determine tax compliance (Adams and Webley 2001; Braithwaite 2003, 2009; Cullis and Lewis 1997). Tax authorities have the power to coerce and ensure tax compliance; taxpayers consider how the government is using their money.

Following Becker’s (1968) economic approach to criminal behaviour, Allingham and Sandmo (1972) and Srinivasan (1973) developed models to measure the evasion of income tax. In these models, the decision to pay tax is a decision made under uncertainty—the safe option is to report income honestly, because evasion is risky—and taxpayers report their income honestly or evade partially or fully depending on income, tax rate, audit probability, and penalties or fines. Their expected utility theory model predicts a negative relationship between marginal tax rates and tax evasion—the “Yitzhaki puzzle” (Yitzhaki 1974). Kirchler et al (2008) provide an overview on the empirical validity of these models, but the empirical and experimental evidence is mixed; some studies even suggest a positive relation between tax rate and tax evasion. Webley et al (1991) find that the probability of an audit affects tax compliance, and the amount of fines has no significant impact, but Alm et al (1992) observe a very weak effect of fines.

Prospect theory has been used as an alternative—some find that it plays a role in explaining tax evasion (Dhami and al-Nowaihi 2007; Bernasconi and Zanardi 2004; Yaniv 1999); Hashimzade et al (2013) find alternative specifications that do not resolve the Yitzhaki puzzle; and Piolatto and Rablen (2017) present a set of conditions under which the puzzle holds. Recently, many researchers have presented other explanations for tax behaviour, acknowledging that people are motivated not only by self-interest but also by sociological and psychological factors and group notions like social norms, fairness, trust, reciprocity, and tax morale (Dell’Anno 2009; Cullis et al 2012; Balestrino 2010).

The SSF proposed by Kirchler (2007) and Kirchler et al (2008) considers the relationship between taxpayers, the government, and tax authorities, and identifies their interactions as determinants of taxpayer compliance behaviour. The framework suggests that trust and power play a key role in enhancing tax compliance (Kirchler et al 2008). It hypothesises that voluntary compliance depends primarily on trust in the government—the belief that the government is benevolent and works beneficially for the common good, for which it needs tax revenue—and enforced compliance depends on the power attributed to and exercised by tax authorities, or the perception that the authorities are capable of detecting tax evasion and punishing it.

The design of the SSF utilises the interactions between taxpayers and the tax authority and allows us to distinguish between a synergistic and an antagonistic relationship between the two (Muehlbacher et al 2011). The antagonistic climate is reflected through the degree of coercive power assigned to the tax authority, whereas a synergistic climate is captured through the level of trust between taxpayers and tax authorities. Thus, in the SSF, tax payments are influenced by the combination of trust and power. If both are low, tax payments are assumed to be low, since taxpayers have no motive to comply except their personal morality. When trust in authorities increases, so do expected tax payments. Tax payments also increase as tax authorities become more powerful. Tax compliance is assumed to be highest if governments are considered trustworthy and they are assisted by a powerful tax authority. In tax compliance literature, these assumptions have been tested using either experiments or surveys of real-world taxpayers (Wahl et al 2010; Muehlbacher et al 2011; Kogler et al 2013), with promising results.

Two main hypotheses derived from the SSF can be tested using data: trust in tax authorities is a predictor of voluntary tax compliance; and the power attributed to tax authorities determines enforced compliance. Voluntary compliance may also be affected by power, and enforced compliance may also be connected to trust, but these linkages are hypothesised to be weaker. Tax compliance can be achieved by encouraging voluntary action, or enforcing compliance, or by encouraging voluntary action and enforcing compliance. Different combinations of trust and power lead to different levels of tax compliance. Figure 1, reproduced from Kirchler et al (2008), depicts the two compliance behaviours as a function of trust and power, as hypothesised in the SSF, and shows the theoretical plane of tax compliance. In this framework, tax compliance is maximum when the highest level of trust is combined with the highest level of power. A fall in either trust or power leads to a sudden fall in tax compliance, as depicted by the slippery slope of the compliance plane. This article analyses tax compliance behaviour as proposed by the SSF and tests its hypotheses.

Data and Methodology

This article uses data collected by direct interaction with respondents through surveys conducted in different parts of India. The survey participants are educated persons who understand that citizens are required to pay income taxes if their income crosses a threshold, but not all participants pay income tax, and income levels vary significantly. Respondents participated on request; they were not paid or rewarded otherwise. All responses based on one of the four randomly assigned scenarios in the questionnaire were collected in article form by the author. Most participants were unknown to the author; they responded to the questions in their first meeting. All participants had graduate-level education in the sciences, arts, commerce, engineering, or management.

This study used four sets of questionnaires depicting different scenarios that describe a virtual country. These questionnaires are adapted from an experiment presented by Wahl et al (2010) and implemented by Kogler et al (2013) and by Kaplanoglou and Rapanos (2015). Each respondent was given only one questionnaire selected randomly out of the four. The four set of questionnaires differ only in terms of the description of the country; all the questions were the same.

The questionnaire described fictitious countries. Participants were asked to read the description, imagine they were citizens of these countries, and respond to the questions. The questionnaire was edited to ensure that the size, population, economy and the literacy rate of these countries were similar to that of India and prevent quick, unintended comparisons. The variations in the four scenarios were to manipulate the perceptions of trust in the government and the power of tax authorities. The participants were asked to imagine that they live, work, and pay taxes in that particular country. The data were collected between July 2016 and December 2017.

Four different scenarios correspond to different combinations of trust and power. In the high trust scenario, people feel that political corruption is low. They trust their government, describe it as transparent, trustworthy, and service-oriented, and they feel that the government utilises taxpayer money properly. The low trust scenario has none of these qualities. In a high power scenario, tax authorities work efficiently to implement tax laws and detect and prosecute tax evasion as per the law. In the low power scenario, tax authorities are not efficient in their duties and are lenient towards tax evaders, and taxpayers know that there is little chance of being prosecuted for tax evasion. These descriptions are combined in four different scenarios: high trust and high power (Scenario 1); low trust and high power (Scenario 2); high trust and low power (Scenario 3); and low trust and low power (Scenario 4). To avoid any bias, one of the scenarios was given as a questionnaire randomly to each respondent. All the participants were asked to fill in the same set of 28 questions, whose answers were formatted on a scale of 1 (highly disagree) to 9 (highly agree).

Manipulation Check

In the questionnaire, one group of six questions—three each for trust and power—is used to check whether respondents have correctly perceived the trust and power dimensions. The manipulation check is carried out for the country’s trust and power description.

Table 1 shows the means and standard deviations of each trust and power manipulation check item for all four scenarios. For the first manipulation check question on trust (MCt1), about whether the government acts fairly towards citizens, participants assigned the high trust scenario had more trust in the government than those with low trust. The second manipulation check question on trust (MCt2) was posed in the opposite direction to that of the first question with respect to trust in the government. The third question on trust MCt3 had the same direction as the MCt1. Similarly, for power manipulation, participants who were told that authorities are powerful perceived them as being more powerful than participants who were told that authorities’ power is weak. MCp1, MCp2, and MCp3 are three manipulation check questions for power description. Questions for MCp1 and MCp3 are designed to indicate in the same direction with respect to power of the authorities but MCp2 in the opposite direction. A correct response to manipulation checks indicates that participants understood the manipulation for power and trust in different scenarios. Only those questionnaires that passed the manipulation check—963 of 1,328—were analysed.

Results and Analysis

Studies that use the SSF find that voluntary compliance is expected to vary depending on the trust individuals have in their government, whereas enforced compliance is expected to depend on tax authorities’ power to penalise tax evaders. This article tests whether these findings hold for India.

Given that the dependent variables—intended compliance, voluntary compliance, enforced compliance, and strategic tax evasion—are ordered categorical variables, the ordinary least squares estimates would be biased and inefficient (Park 2009). To test the SSF hypothesis, this article uses ordered logit estimates. Ordered logit models estimate relationships between an ordinal dependent variable that is both categorical and ordered and a set of independent variables. The estimated odds ratios are reported and analysed in this article using STATA, the statistical software package.

A set of five proxies is used to measure voluntary and enforced compliance on a scale of 1 to 9. These proxies use different questions to measure the same thing. Four dummy variables—d1, d2, d3, d4—are created to capture the four different scenarios, respectively, Scenario 1, Scenario 2, Scenario 3, and Scenario 4. Gender, age, and a measure of dishonesty, along with scenario dummies, are used as explanatory variables. Dummy variable d4, which corresponds to the low trust, low power combination (Scenario 4), is used as the reference in estimation. To avoid the issue of multicollinearity, d4 is dropped from the estimated equation. The odds ratios estimated using the ordered logit model are reported in Table 2 (for explaining voluntary compliance) and Table 3 (for explaining enforced compliance). All the equations are estimated with and without the individual dishonesty variable. Estimates suggest that participant gender or age does not significantly affect compliance behaviour.

Results from Table 2 suggest that trust in the government plays a significant role in improving voluntary compliance, which increases as trust improves. It does not increase with an increase in power only, and it is highest where trust and power are high. This finding is in line with the SSF hypothesis and with earlier empirical studies (Hammar et al 2009; Wahl et al 2010; Muehlbacher et al 2011; Kogler et al 2013; Kastlunger et al 2013; Kaplanoglou and Rapanos 2015). If taxpayers trust that the government and tax authorities will use their money sensibly to provide public goods and services, they will voluntarily pay tax (Feld and Frey 2007).

Another finding that emerges from the Indian data is that in the absence of trust, taxpayers perceive an increase in the power of tax authorities as coercive, and a climate of antagonism develops. This is in line with the theoretical argument of Turner (2005), which distinguishes between legitimate and coercive power. In a low trust environment, taxpayers consider power as coercive, whereas in a high trust environment they perceive it as legitimate. Voluntary compliance is maximal if citizens trust the government and tax authorities are powerful.

Estimates for explaining enforced compliance are reported in Table 4. These are in line with the SSF, which suggests that enforced compliance is a function of power. However, result suggests that power
impacts enforced compliance more in low trust environments. Wahl et al (2010) find similar results for Austria. Kaplanoglou and Rapanos (2015) explain that in low trust environments, taxpayers perceive power as coercive, and they comply only to the extent that tax authorities force them.

Tax evasion through strategic taxpaying: Respondents were asked about their use of strategic taxpaying as a tax evasion tool. Figure 2 presents the responses under the four scenarios. Data suggest that power significantly reduces evasion in high trust environments but not in low trust environments, where strategic taxpaying is highest. High trust and power can discourage strategic taxpaying. Table 4 presents the ordered logit estimate to explain the role of trust and power. Five questions are asked and estimates reported. Dummies for the scenarios, along with age, gender, and individual dishonesty, are used as explanatory variables. Scenario 4 (low trust, low power) is used as the reference. Strategic taxpaying is discouraged only in a high trust, high power environment. Also, individual dishonesty significantly encourages strategic taxpaying.

Intention towards tax compliance: In the study questionnaire, intended tax compliance is measured by three variables—INTTC1, INTTC2, and INTTC3—corresponding, respectively, to the questions “Would you pay your tax completely honestly?”; “Would you declare your annual income completely honestly?”; and “Would you retain part of your taxes?” The variables INTTC1 and INTTC2 assess honesty in reporting income to the tax authority; the variable INTTC3 assesses whether respondents would pay all of their tax. The average score on the last question is lower than in the other two (Table 5). The Pearson Chi2 probability for the pairwise test is also reported in the table. It indicates that the distributions for the reported mean values are significantly different from each other. The high power scenario significantly and positively impacts the intention to pay taxes honestly, and improving both trust and power raises taxpayers’ intention towards tax compliance the highest.

The estimates for trust and power as determinants of intended tax compliance are reported in Table 6. Scenario 1 is used as the benchmark in the estimated equation with the dummy variables for the scenarios along with the respondents’ gender and age variables as explanatory variables. The results suggest that the intention towards tax compliance is the highest in high trust, high power environments, and it drops if trust or power falls.


Honesty a preferred choice: The data suggest that revealing taxable income dishonestly is generally, although not completely, disliked. In the sample of analysis, two peaks are observed in the distribution for the revealed individual level of dishonesty (Table 7). The kernel density plot is depicted in Figure 3.


The first peak in the distribution—the higher of the two—is around the maximum level of revealed honesty. The second peak occurs at the middle, which implies indifference towards dishonesty. On a scale of 1 (lowest) to 9 (highest), the average score for individual dishonesty is 3.74. Thus, many prefer honesty; some others at times like or dislike dishonesty, but they do not seem to prefer it. The revealed individual level of dishonesty is highest in the low trust condition.

The last question of the questionnaire asks respondents which scenarios they find similar to that of India (Table 8). This question was asked in the end as a last question, to avoid the responses of other questions being influenced by respondents’ perception about India while answering them or using the virtual description of the country as their reference. The response for overall similarity is matched with the highest for Scenario 2 (low trust, high power). Trust similarity is also highest for the low trust combination (Scenario 2) followed by Scenario 4. Power description similarity is the highest with that of high level of power (Scenario 1), followed by Scenario 2. Thus, on an average India is seen as a country similar to the description of low trust and high power.



Tax evasion is a concern worldwide, and it is widely studied in both the theoretical and empirical literature on taxation. The early literature emphasised that the authorities needed to ensure tax compliance by using instruments such as the audit rate, tax rate, and fines. Recent research holds that taxpayer trust also influences tax compliance. The SSF examines the interplay of trust and power in explaining both voluntary and enforced compliance. In this framework, tax compliance is maximal when trust and power are highest; a fall in trust or power leads tax compliance to fall, as depicted by the slippery slope of the compliance plane.

Data collected through direct interaction with respondents from different parts of India are used to test the various hypotheses of the SSF. The ordered logit estimation methodology is used to estimate the role of trust and power in voluntary and enforced compliance. In India, trust plays a positive and significant role in voluntary compliance. In a low trust environment, increasing power does not improve voluntary compliance; it is highest in the high trust, high power scenario.

This study’s estimates support the hypothesis that enforced compliance improves with power. In a low trust environment, enforced compliance improves if tax authorities’ power increases, but taxpayers perceive it as coercive. To improve tax compliance overall, taxpayer trust must be improved. Tax evasion through strategic taxpaying is discouraged only in a high trust, high power environment, which also maximises the intention towards honest tax compliance. The data suggest that taxpayers prefer to declare taxable income honestly, and respondents consider India a low trust, high power environment.

Establishing the specific reason for these findings is beyond the scope of this article. The attempts to simplify tax filing and enhance assessment methodologies are expected to improve tax compliance up to some extent only; overall, success depends on improving citizens’ trust and the tax authorities’ power.


Adams, C and P Webley (2001): “Small Business Owners’ Attitudes on VAT Compliance in the UK,” Journal of Economic Psychology, Vol 22, No 2, pp 195–216.

Allingham, Michael and Agnar Sandmo (1972): “Income Tax Evasion: A Theoretical Analysis,” Journal of Public Economics, Vol 1, pp 323–38.

Alm, James (2012): “Measuring, Explaining, and Controlling Tax Evasion: Lessons from Theory, Experiments and Field Studies,” International Tax and Public Finance, Vol 19, pp 54–77.

Alm, James, G H McClelland and William D Schulze (1992): “Why Do People Pay Taxes?” Journal of Public Economics, Vol 48, pp 21–38.

Balestrino, A (2010): “Tax Avoidance and the Endogenous Formation of Social Norms,” The Journal of Socio-Economics, Vol 39, pp 601–09.

Becker, Gary S (1968): “Crime and Punishment: An Economic Approach,” The Journal of Political Economy, Vol 76, pp 169–217.

Bernasconi, M and A Zanardi (2004): “Tax Evasion, Tax Rates and Reference Dependence,” FinanzArchiv, Vol 60, No 3, pp 422–45.

Braithwaite, Valerie (2003): “Dancing with Tax Authorities: Motivational Postures and Non-Compliant Actions,” Taxing Democracy, Understanding Tax Avoidance and Evasion, V Braithwaite (ed), Aldershot: Ashgate Publishing Limited, pp 15–39.

— (2009): Defiance in Taxation and Governance: Resisting and Dismissing Authority in a Democracy,” Cheltenham, UK: Edward Elgar.

Cullis, J and A Lewis (1997): “Why People Pay Taxes: From a Conventional Economic Model to a Model of Social Convention,” Journal of Economic Psychology, Vol 18, Nos 2–3, pp 305–21.

Cullis, J, P Jones and A Savoia (2012): “Social Norms and Tax Compliance: Framing the Decision to Pay Tax,” The Journal of Socio-Economics, Vol 41, pp 159–68.

Dell’Anno, R (2009): “Tax Evasion, Tax Morale and the Policy Maker’s Effectiveness,” The Journal of Socio-Economics, Vol 38, pp 988–97.

Dhami, S and A al-Nowaihi (2007): “Why Do People Pay Taxes? Prospect Theory versus Expected Utility Theory,” Journal of Economic Behavior and Organization, Vol 64, No 1, pp 171–92.

Feld, Lars P and Bruno S Frey (2007): “Tax Compliance as the Result of a Psychological Tax Contract: The Role of Incentives and Responsive Regulation,” Law & Policy, Vol 29, pp 102–20.

Hammar, Henrik, Sverker C Jagers and Katarina Nordblom (2009): “Perceived Tax Evasion and the Importance of Trust,” Journal of Social-Economics, Vol 38, No 2, pp 238–45.

Hashimzade, N, G D Myles, and B Tran-Nam (2013): “Applications of Behavioural Economics to Tax Evasion,” Journal of Economic Surveys, Vol 27, No 5, pp 941–77.

Kaplanoglou, G and Vassilis T Rapanos (2015): “Why Do People Evade Taxes? New Experimental Evidence from Greece,” Journal of Behavioral and Experimental Economics, Vol 56, pp 21–32.

Kastlunger, Barbara, Edoardo Lozza, Erich Kirchler and Alfred Schabmann (2013): “Powerful Authorities and Trusting Citizens: The Slippery Slope Framework and Tax Compliance in Italy,” Journal of Economic Psychology, Vol 34, pp 36–45.

Kirchler, Erich (2007): The Economic Psychology of Tax Behaviour, Cambridge: Cambridge University Press.

Kirchler, Erich, Erik Hoelzl and Ingrid Wahl (2008): “Enforced versus Voluntary Tax Compliance: The ‘Slippery Slope’ Framework,” Journal of Economic Psychology, Vol 29, No 2, pp 210–25.

Kogler, Christopher, Larissa Batrancea, Anca Nichita, Jozsef Pantya, Alexis Belianin and Erich Kirchler (2013): “Trust and Power as Determinants of Tax Compliance: Testing the Assumptions of the Slippery Slope Framework in Austria, Hungary, Romania and Russia,” Journal of Economic Psychology, Vol 34, pp 169–80.

Muehlbacher, Stephan, Erich Kirchler and Herbert Schwarzenberger (2011): “Voluntary versus Enforced Tax Compliance: Empirical Evidence for the ‘Slippery Slope’ Framework,” European Journal of Law Economics, Vol 32, No 1, pp 89–97.

Park, Hun M (2009): “Regression Models for Ordinal and Nominal Dependent Variables Using SAS, Stat, Limdep and SPSS,” The University Information Technology Services Center for Statistical and Mathematical Computing, Indiana University, Working Paper.

Piolatto, A and M Rablen (2017): “Prospect Theory and Tax Evasion: A Reconsideration of the Yitzhaki Puzzle,” Theory and Decision, Vol 82, No 4, pp 543–65.

Srinivasan, T N (1973): “Tax Evasion: A Model,” Journal of Public Economics, Vol 2, No 4, pp 339–46.

Turner, John C (2005): “Explaining the Nature of Power: A Three-process Theory,” European Journal of Social Psychology, Vol 35, pp 1–22.

Wahl, I, S Muehlbacher and E Kirchler (2010): “The Impact of Voting on Tax Payments,” Kyklos, Vol 63, No 1, pp 144–58.

Webley, P, H S J Robben, E Elffers and D J Hessing (1991): Tax Evasion: An Experimental Approach, Cambridge, England: Cambridge University Press.

Yaniv, G (1999): “Tax Compliance and Advance Tax Payments: A Prospect Theory Analysis,”National Tax Journal, Vol 52, No 4, pp 753–64.

Yitzhaki, S (1974): “A Note on Income Tax Evasion: A Theoretical Analysis,” Journal of Public Economics, Vol 3, No 2, pp 201–02.

Updated On : 24th May, 2019


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top