The Power of Tax Authority and Tax Compliance: Does Tax Literacy Matter?

Purposes: Our research aims to prove the influence of tax authority power (coercive and legitimate), tax morality, perception of fairness, and religiosity on tax compliance. Also, we test the moderating effect of tax literacy on the influence of the relationship between perceived fairness and coercive power on tax compliance. Methods: This research used 199 respondents determined by convenience sampling, snowball sampling technique. Data were analyzed using Partial Least Squares based Structural Equation Modeling (PLS-SEM). Results: We report that legitimacy power, tax morale, and religiosity positively influence tax compliance. On the other hand, coercive power and perceived fairness do not influence tax compliance. Moreover, our study reports that tax literacy has a role in reducing the influence of perceived fairness on tax compliance. Novelty: Our study contributes to expanding literature in two important ways. First, we re-examined the influence of the power of tax authority on tax compliance of university employees who experienced a change in university status from a public university to a state university with legal entity status (PTN-BH)-causing a change in tax rates. Second, we expand the slippery slope framework by proving the role of tax literacy in increasing the influence of the power of tax authority on tax compliance.


INTRODUCTION
Taxes are still the main source of income for financing government spending.As the need for state funding increases, the tax authority will strive to increase taxes as government income through increasing taxpayer compliance and expanding taxpayers (Wu & Teng, 2005).However, governments of various countries report tax revenues that are lower than the target, indicating that taxpayer tax compliance is low (Iraman et al., 2022;Inasius et al., 2020).Thus, tax authorities still face challenges in reducing tax non-compliance (Kastlunger et al., 2013) and efficient tax management, especially in developing countries (Carnahan, 2015).
One of the factors that researchers debate to explain tax compliance is tax authority power, including in the context of studies in Indonesia (Saptono & Khozen, 2023;Khasanah et al., 2019).The self-assessment system adopted by the state, tax revenues received, emphasizes taxpayers' awareness of paying taxes according to their income.This system allows taxpayers to report taxes that are less than the bill they have to pay.Following Saptono & Khozen (2023), Lisi (2014), Kastlunger et al. (2013), and Kogler et al. (2013), tax authority power can increase tax compliance and reduce tax avoidance.However, Bakar et al. (2022) and Taing & Chang (2021) reported that tax authority power is not effective in increasing tax compliance.Following Kirchler et al. (2008), the mixed result of prior studies is because taxpayers' compliance with taxes has various motives: economic and non-economic.The economic approach states that taxpayers will comply with tax because they calculate the costs for non-compliance as being too high, while the non-economic approach states that awareness as members of society causes them to comply (Kirchler et al., 2008).Taxpayers for economic reasons, the application of inspection rates and fines will increase tax compliance for two reasons (Mohdali et al., 2014).First, taxpayers try to avoid taxes every time they pay them (Alstads et al., 2022;Inasius et al., 2020).Second, most taxpayers seem to doubt the legitimacy of the tax system, and tax collection does not consider the possibility of taxpayers avoiding taxes (Kirchler et al., 2008).Thus, tax regulation settings, different taxpayer motives, and tax authority may impact tax compliance differently.
The contribution of this research is in two important ways.Firstly, we re-examine the effect of tax authority power on tax compliance, but with the background of a different social environment subject to tax rate changes.We used university employees who experienced a change in university status from a public university to PTN-BH.This change in university status causes changes in tax rates (Hamid & Christine, 2019).Changes in social settings due to tax rates can influence taxpayer psychology, trust, and perceived fairness and thus have an impact on taxpayer compliance (Jimenez & Iyer, 2016).This is in line with the slippery slope argument, which contends that taxpayers' compliance with the law is determined by their perceptions of the government's legitimacy and level of trust in it (Inasius et al., 2020).
Second, we expand the slippery slope framework introduced by (Kirchler et al., 2008)-highlighted to understand compliance behaviors: trust of taxpayers and power of tax authorities.Discussion of both factors to increase tax compliance presents mixed results.The effectiveness of the slippery slope framework to explain taxpayers must be differentiated in two different taxation environments: synergistic and antagonistic (Inasius et al., 2020).However, we see that the slippery slope framework assumes that taxpayers understand the taxation system used.Trust can be enhanced by increased accountability of tax authorities.Thus, the role of tax literacy needs to be considered when explaining tax compliance.Tax literacy can be defined as efforts to increase knowledge and understanding of the tax-related system for society and increase tax compliance (Agusti & Rahman, 2023;Alexander & Balavac-Orlic, 2022;Çetin Gerger et al., 2019).In the context of our research, which emphasizes taxpayers who experience changes in tax rates, literacy is needed to improve compliance.Following Alexander & Balavac-Orlic (2022), we argue that tax literacy is a moderating effect in the model to explain tax compliance.
Our paper reminder is structured into interrelated sections.The second section provides a literature review of tax compliance based on the slippery slope framework and theory of planned behavior, followed by research methods.The fourth section briefly overviews the research results, followed by a discussion.The final section presents the conclusions.Kirchler et al. (2008) presented the slippery slope framework to explain taxpayers' tax compliance.Based on this framework, tax compliance depends on two major dimensions: perceived trust in the authorities and perceived power of the authorities.The perceived trust of taxpayers in the tax authority fosters voluntary compliance, and conversely, perceived power leads to enforced compliance.Based on the slippery slope framework, the possibility of audits and fines is important to increase tax compliance by regulating citizen behavior to comply with taxes.Batrancea et al. (2019) call it increasing tax compliance through increasing tax authority power, which enforces compliance on tax evasion.However, recent empirical evidence shows mixed results regarding the influence of power and coerced compliance on tax avoidance.Saptono & Khozen (2023), Lisi (2014), Kastlunger et al. (2013), andKogler et al. (2013) report that tax authority power can increase tax compliance and reduce tax avoidance and vice versa Bakar et al. (2022) and Taing & Chang (2021) report that tax authority power is not effective in increasing tax compliance.The difference in the results of this research is because there is an opinion that heavy prosecution and coercion will increase strategic tax-paying behavior.In contrast, taxpayers will behave oppositely once they see the possibility of their behavior not being caught by the tax authorities (Kastlunger et al., 2013).

The Slippery Slope Framework and Hypothesis
The slippery slope framework differentiates between the power of tax authority, which is coercive and legitimate (Kastlunger et al., 2013).Based on the slippery slope framework view, coercive power can be defined as coercive power by providing punishment, prosecution, and fines as a determining factor in tax compliance.This approach holds that the only sustainable way to increase government revenue is by coercion, which is why sanctions and punishments (fines, penalties, and jail time) are used to force citizens to pay taxes (Bakar et al., 2022).Furthermore, when tax authority strength is low, taxpayers who purposefully violate tax regulations to maximize their outcomes do so, reducing voluntary tax compliance's influence.According to the paper of Bayissa & Dijke (2016) and Sadress et al. (2018), coercive authority has a favorable effect on enforced compliance.Additionally, Frecknall-Hughes et al. (2023) document that coercive power increases taxpayers' interest in using tax practitioners in determining the taxes they must pay.The second type of power is legitimacy power.Legitimate power is based on the perception that the taxpayer has the right to pay tax, typically due to a form of responsibility in society.An institution's authority must be legitimate for it to have legitimate power, which comprises the supply of necessary information and the ability of tax authorities to persuade individuals to identify their reasons for voluntarily paying taxes (Bakar et al., 2022).which highlights the importance of legitimacy.power as regulation of correct, legitimate, and effective behavior rather than coercive power.The tax authority is legitimately empowered if taxpayers perceive it as one that maintains a fair allocation of tax loads and benefits among all taxpayers.The taxpayers trust the tax authority to do the right things and have some level of confidence that the authorities do not abuse their power and exploit them.Thus, taxpayers will comply.Gangl et al. (2020) and Kastlunger et al. (2013) report that legitimate power is effective in increasing tax compliance.
H 2 : Legitimate power has a positive influence on tax compliance.
Taing & Chang (2021), Taing & Chang (2021), Damayanti et al. (2015), and Bobek & Hatfleld (2003) use the social psychology theory of planned behavior (TPB) to explain tax compliance.Social psychology's TPB describes how attitudes toward behavior, subjective norms, and perceived behavioral control all have an impact on behavioral intention (Ajzen, 1991).TPB is an expansion of the earlier social psychology theory of Reasoned Action (TRA), which incorporates behavioral intentions as significant motivators for predicting the actual behavior of humans, and it assumes that people have volitional control over their behavior (Taing & Chang, 2021).Generally, a person is more likely to practice a behavior if they are more inclined to engage in it.Damayanti et al. (2015) used TPB and found that attitude on tax compliance, subjective norms, and perceived behavioral control affect tax compliance.An extension of TPB to tax compliance by Taing & Chang (2021) report that tax morale, fairness, and complexity have a statistically significant influence on the tax compliance intention of the citizens.In contrast, the power of authority, trust in government, tax information, and tax awareness do not affect tax compliance.This argument is based on opinion of Luttrell et al. (2016) that attitudes are based on moral foundations.Similarly, Bobek & Hatfleld (2003) they modified the TPB by adding a moral obligation and stating that the TPB is the right theory to explain tax compliance.Also, they report that moral obligation significantly explains the tax non-compliance.
Following Taing & Chang (2021) and Bakar et al. (2022), we expand the TPB by using tax morale as a factor to explain tax compliance.Tax compliance is a form of behavior from taxpayers who are motivated by other "ethical" factors, such as morality, altruism, and fairness (Luttmer & Singhal, 2014).Moreover, the ability to pay taxes indicates a person's support for tax morale, regardless of government spending or the actions of other taxpayers (Ortega et al., 2016).Taing & Chang (2021) and Hassan et al. (2021) also, confirm that it has a statistically significant influence on the tax compliance intention of the taxpayer.
H 3 : Tax morale has a positive influence on tax compliance.Taing & Chang (2021) added the perception of fairness as a factor influencing tax compliance and their effort to modify the TPB theory.In the TPB approach, the perception of fairness forms a subjective norm because the perception of fairness is an individual's perception of certain behavior, which is influenced by the assessment of those closest to them, the tax management system, and tax accountability managed by the government.Moreover, Alexander & Balavac-Orlic (2022) and Hassan et al. (2021) report that the perception of fairness in a tax system is found to be an important factor in tax morale, which has a positive impact on tax compliance.Also, the perception of fairness in the tax system increases a taxpayer's trust in the government and, consequently, positively impacts tax compliance (Alexander & Balavac-Orlic, 2022).Moreover, taxpayers' perceptions of the fairness of the distribution of the tax burden between income groups and participation in public goods, procedural fairness, and social norms as factors forming taxpayers' trust in tax authority are also crucial for increasing tax compliance.Saptono & Khozen (2023) and Taing & Chang (2021) confirm that tax fairness has a statistically significant influence on the tax compliance intention of the taxpayer.

H 4 : Perception of fairness has a positive influence on tax compliance.
Religion is seen as a real fact of the human experience and believed to be an influential factor in the development of an individual's personality and a critical part of life (Said et al., 2018).Religiousness is defined as "believing strongly in the existence of a God.Religion has holy books that teach good and bad deeds that its adherents must obey.Thus, religion can influence behavior.Researchers have proven the influence of religiosity on the behavior of fraud (Said et al., 2018;Rohman et al., 2023;Purnamasari & Amaliah, 2015), depositor (Aysan et al., 2018), customer (Mokhlis, 2009;Mohdali & Pope, 2014), and tax compliance (Davidescu et al., 2022;Mohdali & Pope, 2014;Torgler, 2006).
Based on the perspective slippery slope framework, tax morale is an individual characteristic of taxpayers and influences tax compliance.Davidescu et al. (2022) report that religiosity and religion have an effect on tax morale, and an individual's religiosity reduces tax evasion and increases the degree of tax morale.Mohdali & Pope (2014) report that religiosity has a statistically significant positive impact on voluntary tax compliance.Also, Torgler (2006) stated that religiosity raises tax morale.Religion impacts people's personalities, beliefs, and practices that are applied daily, affecting tax compliance.

METHODS
This study aims to help taxpayers at Universitas Negeri Semarang (UNNES) whose status has changed from that of a public university to that of PTN-BH.Under SE-34/PJ/2017 states that the status of universities as PTNBH causes universities not to meet the criteria for certain units of government bodies, which are exempt from being subject to domestic tax, as also regulated in article 2 paragraph (3) letter b of the Income Tax Law.This regulation is the basis for PTNBH changing its organizational status to taxable entrepreneur (PKP) and PTNBH employee income tax rates using progressive tax rates.
In 2023, UNNES will have 1,886 employees.The sample was determined using convenience sampling to determine respondents because researchers did not have the ability to select samples based on probability sampling -not all taxpayers were willing to be respondents.The number of samples is determined based on sampling quotas, and Hair et al. (2021) and Kock & Hadaya (2018) recommend the minimum sample size when using SEM-PLS is the '10-times rule' method.This means that the sample size employed in an empirical study should be equal to 10 times the number of independent variables in the most complex regression in the PLS path model (Hair et al., 2021;Kock, 2018).In the context of our research, there are seven variables used in our research model and require a minimum sample size of 70 taxpayers.
The development of the questionnaire began by conducting a pilot study, which we developed simply from broader survey questions and circulated to a small number of potential respondents.We tested the validity of the instrument and revised the instrument in accordance with input for improvements from respondents in this pilot study.Respondents who receive the questionnaire via Google Form, which we distribute via WhatsApp, will then be asked to send the questionnaire link to their friends who meet the requirements as respondents for this research (snowball sampling technique).Data collection was carried out for two months, July to August 2023, and resulted in 199 respondents.This number is greater than the minimum sample size recommended by (Hair et al., 2021).
The tax compliance of taxpayers (TC) variable is measured with six indicators adopted from Kirchler et al. (2008).Tax morale (TM) variable is measured with four indicators adopted from Bakar et al. (2022), Torgler (2007), and Hana & Vitezić (2023), and also tax authorities' power (coercive power-CP and legitimate power-LP) measured with six and five indicators, respectively, adopted from (astlunger et al. (2013).Religiosity (REL) is measured with six indicators adapted from Rohman et al. (2023) and perception of fairness (PF) is measured with four indicators adopted from Taing & Chang (2021).Finally, we use six indicators adopted from Alexander & Balavac-Orlic (2022) to measure the tax literacy (TL) variable.We present all indicators for measuring variables in Appendix 1. Research variables were measured by using a 5-point Likert scale questionnaire, scoring from 1 to 5 and identifying from "strongly disagree" to "strongly agree."

RESULTS AND DISCUSSIONS Descriptive Analysis
Table 1.presents a description of respondents based on gender and age.Table 1.shows that 54.31% of respondents were female and 45.69% were male.Based on educational background, our data shows 59.39% of respondents have postgraduate education, and 40.61% have graduate education.The majority (41.71%) were of productive age (30-40 years), and Table 1.presents 27.64% of respondents aged 27.64%.
Appendix 1 presents a descriptive analysis of all variables, and it can be seen that the average TC score is 4.00.These results indicate that respondents have a good level of tax compliance.However, the power authority indicator (coercive and legitimate power) produces an average score of 2.53 and 3.67, respectively.This score shows that the coercive power of tax authority is low.This finding may be because Indonesia uses a self-assessment system in its tax collection system, so they are rarely subject to sanctions and punishments (e.g., fines, penalties, and jail time).Appendix 1 also reports the average scores of individual psychological variables, such as taxpayer, tax morale, and religiosity, of 4.01 and 4.56.These results also show that our respondents have a very good level of adherence to their religious teachings, which also encourages good tax morale.However, the perception of fairness and tax literacy produced low average scores (3.10 and 3.21, respectively).These results also show that respondents have insufficient knowledge about the tax system, which may cause their perception of fairness to be low.Respondents believe the low tax literacy is due to system changes and new tax rates that have not been socialized to taxpayers.

Measurement Model Assessment.
This research uses warp-PLS as a statistical tool to process research data.This software presents several indicators for evaluating model fit and quality indices, and the standard score has been determined.The p-value of the Average path coefficient (APC), Average R-squared (ARS), and Average adjusted R-squared (AARS) are 0.023, <0.001, and <0.001, respectively.A P-value of these three indicators for model fit test results was lower than 0.05 and indicated that our model had met the indicators of the goodness of fit model (Kock, 2019).The average block VIF (AVIF) score is 3.013, which indicates that our model is acceptable because the score is lower than five and lower than the ideal score (3.3) (Kock, 2019).
The goodness of fit (GoF) referred to as Tenenhaus GoF is a global fitness measure (Alotaibi & Alshahrani, 2022).GoF values under 0.10 can be considered small, values between 0.10 and 0.25 can be considered medium, and values to 0.35 can be considered significant (Kock, 2019).In this study, GoF was 0.555, meaning it was sufficiently large enough to ensure the validity of the PLS model (Alotaibi & Alshahrani, 2022).Sympson's paradox ratio (SPR) is a measure of the extent to which a model is free from Simpson's, and an instance of Simpson's paradox occurs when a path coefficient and a correlation associated with a pair of linked variables have different signs (Kock, 2019).Our model results in an SPR value of 0.714 (greater than 0.7) and indicates that our model is free from Simpson's paradox.The R-squared contribution ratio (RSCR) value is 0.998 (greater than 0.9), indicating that the sum of positive R-squared contributions in a model is 99.8 percent of the total sum of the absolute R-squared contributions in the model.Following Kock (2019), Nonlinear bivariate causality direction ratio (NLBCDR) index is a measure of the extent to which bivariate nonlinear coefficients of association provide support for the hypothesized directions of the causal links in a model.Our model result in NLBCDR value is 0.929 and indicates that in 92.9 percent of path-related instances in a model, the support for the reversed hypothesized direction of causality is weak or less (Kock, 2019).
The loading factor describes the correlation score between the indicator and its latent construct, and the reliability of an item is considered acceptable when its factor loading is greater than 0.6.Higher values of factor loading indicate higher levels of reliability.The reliability values between 0.60 and 0.70 are considered "acceptable in exploratory research, " values between 0.70 and 0.90 range from "satisfactory to good, " Values of 0.95 and higher indicate that the items are redundant, thereby reducing construct validity (Hair et al., 2019).Our analysis result showed that all measurements are significant, with a loading factor of more than 0.60, indicating that the measurement is acceptable.

Structural Model Assessment and Discussion
Table 2. describes path coefficients, p-values, and decisions on hypothesis support.The first hypothesis shows that the path coefficient and p-value are 0.001 and 0.497, respectively.This fact indicates that the first hypothesis, which states that coercive power positively influences tax compliance, is rejected.The results of this research are in line with the findings of Bakar et al. (2022), Taing &Chang (2021), andHofmann et al. (2014), which reported that coercive power is not effective in increasing tax compliance.Hofmann et al. (2014), using a sample of sixty-two taxpayers recruited from acquaintances of university members, reported that tax authority power was ineffective in increasing tax compliance.Also, Bakar et al. (2022) report that there is no influence of coercive power on tax compliance.Coercive power is power based on fear.In this approach, the government increases tax compliance for taxpayers by providing punishment or coercion, e.g., fines, penalties, and jail time (Bakar et al., 2022).Severe prosecution and coercion against tax violators may increase strategic tax-paying behavior.However, on the other hand, taxpayers will behave oppositely if they see the possibility of their tax violation behavior not being caught by the tax authorities (Kastlunger et al., 2013).
Based on Table 2. our study receives the second hypothesis that legitimate power positively influences tax compliance.The results of the second hypothesis test produced a path coefficient score of 0.320 and p-value <0.001.Our research results are in line with the findings of Gangl et al. (2020) and Kastlunger et al. (2013) that legitimate power is effective in increasing tax compliance.Our study supports the slippery slope framework that legitimate power is vital to increase taxpayers' awareness of fulfilling their obligations.Legitimate power is the power the people legally give to the government or tax authority.This legitimate power causes taxpayers to feel responsible and respect tax authorities; taxpayers realize that the state needs taxes to finance public interests.Hence enhancing the tax compliance of taxpayers.Also, legitimate power is based on the perception that taxpayers have the right to pay taxes, typically due to a form of responsibility in society.An institution's authority must be legitimate in order for it to have legitimate power, which includes the supply of necessary information and the ability of tax authorities to persuade individuals to identify their reasons for voluntarily paying taxes (Bakar et al., 2022).Additionally, legitimate power seems to have a positive impact on a large set of indicators such as trust, climate, motivation, and also tax compliance (Gangl et al., 2020).
We found that the correlation between tax morale and tax compliance produces a path coefficient of 0.221 and a p-value of 0.007.These results indicate that tax morale has a positive influence on tax compliance.The results of this research are in line with Hassan et al. (2021), McCulloch et al. (2021), Taing &Chang (2021), andDamayanti et al. (2015), which found a positive relationship between tax morale and tax compliance.Tax compliance is a form of behavior from taxpayers who are motivated by other "ethical" factors, such as morality, altruism, and fairness (Luttmer & Singhal, 2014).The ability to pay taxes indicates a person's support for tax morale, regardless of government spending or the actions of other taxpayers (Ortega et al., 2016).Castañeda (2023) highlighted its importance on taxpayer views about public goods provision, an important influence on moral tax, increasing taxpayers' willingness to pay taxes.Taing & Chang (2021) and Hassan et al. (2021) also confirm that they have a statistically significant influence on the intention of taxpayers to comply with tax regulations.
Table 2. shows the effect of the perception of fairness on tax compliance results with a path coefficient of 0.044 and a p-value of 0.321.This result indicates that the fourth hypothesis is rejected, and the decision is made that the perception of fairness does not affect tax compliance.Perceptions of fairness of the tax system refer to any element of the environment perceived by taxpayers or collectives as fair according to previous norms or standards (Roeters, 2014).In the TPB approach, the perception of fairness forms a subjective norm because the perception of fairness is an individual's perception of certain behavior, which is influenced by the assessment of those closest to them, the tax management system, and tax accountability managed by the government.Thus, the perception of fairness in the tax system has a positive impact on tax compliance (Alexander & Balavac-Orlic, 2022;Hassan et al., 2021).However, from the test results of this research, we do not show a positive correlation between the perceived fairness of the tax system and tax compliance.The absence of this relationship may be because taxpayers still do not understand the tax system, especially in our research design, which uses taxpayers who experience changes in tax rates.Therefore, literacy of the new tax system is really needed for taxpayers, as a basis for them to increase their understanding of the new system.Understanding the tax system is their basis for evaluating the fairness of the tax system and subsequently increasing tax compliance (Rahmawati, 2022).Thus, the perception of fairness of taxpayers is conditional on the level of tax literacy (Alexander & Balavac-Orlic, 2022).
Finally, our study shows that religiosity has a path coefficient and p-value of 0.270 and 0.001, respectively.Our study shows that religiosity has a positive impact on tax compliance.This finding supports the findings of Davidescu et al. (2022), Mohdali &Pope (2014), andTorgler (2006) that religiosity is a vital factor in increasing taxpayers' tax compliance.Religion contains permitted and prohibited behavior as a guide to behavior and subsequently shapes behavior, including behavior to fulfill tax obligations.Also, religion is seen as a real fact of the human experience and is believed to be an influential factor in developing an individual's personality and a critical part of life.(Said et al., 2018).In this context, taxpayers with high religiosity are taxpayers who believe strongly in the existence of a God and implement the role of religion.Thus, religion can influence behavior and, in turn, influence tax compliance (Davidescu et al., 2022;Mohdali & Pope, 2014;Torgler, 2006).
The impact of religion can also be viewed from a slippery slope framework perspective, where religion shapes taxpayer morale, creates individual characteristics of taxpayers, and, therefore, influences tax compliance.Davidescu et al. (2022) support this argument and report that religiosity and religion have an effect on tax morale.There is a religious teaching that obeying the government as the leader of the country (ulul amri) is mandatory, and tax compliance is a form of obeying ulul amri.Davidescu et al. (2022) document that an individual's religiosity reduces tax evasion and increases the degree of tax morale.Mohdali & Pope (2014) report that religiosity has a statistically significant positive impact on voluntary tax compliance.Also, Torgler (2006) state that religiosity raises tax morale because religion impacts people's personalities, beliefs, and practices that are applied daily and consequently affects tax compliance.Prior literature has explained the role of two types of power (coercive and legitimate) in explaining tax compliance behavior.Although single relationships between model variables are well documented, the overall framework cannot report inconsistencies in research results.Apart from that, the absence of a relationship between coercive power and the perception of fairness we report means that other variables can explain the relationship between coercive power and tax compliance.In the context of our research, our sample is taxpayers who experienced changes in tax rates due to changes in the status of the organization where they work, so we argue that tax literacy is needed to increase their understanding of the tax system.The consequence will be increased tax compliance (Agusti & Rahman, 2023;Alexander & Balavac-Orlic, 2022;Çetin Gerger et al., 2019).

Additional Test
Table 2. shows that coercive power and perception of fairness do not affect tax compliance.Following Alexander & Balavac-Orlic (2022), we use tax literacy as a moderating effect in the model used to explain the influence of coercive power and perception of fairness, which has no effect on tax compliance.We present the test results of the moderating effect of tax literacy on the relationship between coercive power and perception of fairness on tax compliance in Table 3.
Table 3. shows that tax literacy*perceived fairness produces a path coefficient of -0.226 and p-value of 0.006, which indicates that tax literacy has a significant influence on reducing the influence of perceived fairness on tax compliance.The Slippery Slope Framework can explain tax compliance, assuming taxpayers understand the tax system used.Trust, an important factor of the Slippery Slope Framework, can be improved by increasing the accountability of tax authorities.Tax literacy can be defined as efforts to increase knowledge and understanding of the tax-related system for society and increase tax compliance (Agusti & Rahman, 2023;Alexander & Balavac-Orlic, 2022;Çetin Gerger et al., 2019).In the context of our research-emphasizing taxpayers who experience changes in tax rates, the role of literacy is needed to improve compliance (Alexander & Balavac-Orlic, 2022).Thus, Gangl et al. (2020) suggest to tax authorities that they provide clear and transparent information on tax procedures and foster a positive reputation as a service provider.This can increase citizens' trust and cooperation and increase tax compliance.
Table 3. shows that Tax literacy*coercive power produces a path coefficient of -0.001 and p-value of 0.494 and indicates that tax literacy does not affect reducing the influence of coercive power on tax compliance.In the Slippery Slope Framework approach, the effectiveness of coercive and legitimate power to explain tax compliance depends on different taxation environments: synergistic and antagonistic (Inasius et al., 2020).In the context of our research, which uses taxpayers in Indonesia using a self-assessment tax system, synergism between taxpayers and tax authorities is the main requirement in tax management.This is what underlies our research results: knowing and understanding the tax-related system of taxpayers does not cause an increase in tax compliance in antagonistic tax environments.

CONCLUSIONS
Our research aims to prove the influence of tax authority power (coercive and legitimate), tax morality, perception of fairness, and religiosity on tax compliance.Using a sample of 199 taxpayers, we show that legitimacy power, tax morale, and religiosity positively influence tax compliance.On the other hand, coercive power and perceived fairness do not influence tax compliance.The self-assessment system emphasizes a synergistic approach between tax authorities and taxpayers rather than antagonistic.The power of authorities that originates from coercive power produces a tax environment that tends to be antagonistic; the provision of sanctions and punishments, such as fines, penalties, and jail time, may increase tax compliance.However, when taxpayers perceive that their tax avoidance behavior is safe, they will reduce tax compliance by doing what is considered safe.In addition, the synergistic approach to the tax environment is more effective in increasing tax compliance because of their commitment and awareness that paying taxes is the entire country's obligation.A synergistic tax environment causes high tax morale.Moreover, high religiosity encourages them to behave according to religious teachings and tax compliance is mandatory for religious followers.
The next stage is testing the moderating effect of tax literacy on the relationship between cognitive power and perceived fairness on tax compliance.Surprisingly, our study reports that tax literacy has a role in reducing the influence of perceived fairness on tax compliance.However, regarding the relationship between coercive power and tax compliance, we did not find a role for tax literacy in the relationship between the two.Tax literacy can increase taxpayers' knowledge and understanding of the tax-related system, increase taxpayer analysis in tax management, and further increase perceived fairness.A high perception of fairness in the tax system increases tax compliance.
This study provides recommendations to the tax authority to increase tax literacy among the public, especially taxpayers who experience changes in tax rates.UNNES, together with tax authorities, can provide literacy about changes in tax rates to its employees.Their knowledge and understanding can increase perceived fairness and tax compliance.It is also important to implement a synergistic tax environment in managing the tax system because it can increase tax morale and thus increase tax compliance.The tax authorities can also increase tax compliance through increasing religiosity, for example, by increasing religious events, because it positively impacts the awareness that paying taxes is an obligation and subsequently increases tax compliance.
This study focuses on taxpayers at UNNES; other research objects may produce different findings.Other researchers can expand this study to other tax objects and not limit it to tax objects affected by changes in tax rates.Apart from that, other researchers can further analyze the influence of coercive power on tax compliance by paying attention to the role of tax system information disclosure, which might increase the influence of coercive power on tax compliance.

H 1 :
Coercive power has a positive influence on tax compliance.

H 5 :
Religiosity has a positive influence on tax compliance.

Table 2 .
Results of test of hypotheses

Table 3 .
Moderation effect of tax literacy