Effects of taxpayer compliance with taxpayer consciousness as intervening variable

This study analyzes (1) the effect of tax authorities services on taxpayer awareness; (2) the effect of tax sanctions on taxpayer awareness; (3) the effect of tax audits on taxpayer awareness; (4) the effect of taxpayer awareness on mandatory compliance; (5) the effect of tax authorities services on taxpayer compliance; (6) the effect of sanctions on taxpayer compliance; (7) the effect of tax audits on taxpayer compliance; (8) the role of mediating the awareness of taxpayers in the influence of tax authorities services on taxpayer compliance; (9) the role of mediating the awareness of taxpayers in the effect of tax sanctions on taxpayer compliance; (10) the role of mediating taxpayer awareness in the effect of tax audits on taxpayer compliance. The population of this study is an individual taxpayer who performs free work registered at the North, South and West Makassar Primary Tax Office with 234.104 effective taxpayers still active. The research sample was 100 active taxpayers in the regional office of the directorate general of taxes in South Sulawesi. The sampling technique is purposive sampling. Data were analyzed using the Structural Equation Modeling method with the help of the Smart-PLS 3.0 program. The results of the study show that tax sanctions and tax audits have a direct or indirect effect on taxpayer compliance, while the services of tax authorities do not affect direct or indirect taxpayer compliance. The interesting result in this study is that the service of the tax authorities is not significant towards taxpayer compliance, but the positive influence fully mediates the effect of tax authority’s services on taxpayer compliance through taxpayer awareness.


INTRODUCTION
The development cannot be implemented if the government does not have the availability of funds. The implications of non-implementation of development are that efforts to realize people's welfare are hampered. This shows that the availability of funds for the government is so important to facilitate the development process. One source of funding for the government comes from taxes. The Directorate General of Taxes of the Republic of Indonesia in 2016 noted The low level of taxpayer compliance can be seen from the return rate associated with the delivery of the Notice. The following is a table that explains the level of taxpayer compliance at the Pratama Makassar Tax Office from 2013 to 2016. Based on Table 1. above, it can It is seen that from 2013 to 2017, the level of compliance of Individual Taxpayers at Makassar KPP Pratama declined. This certainly requires a study so that it does not occur protractedly. Therefore, these conditions provide motivation for conducting research on several factors that influence the level of taxpayer compliance at KPP Pratama Makassar.
Efforts to improve taxpayer compliance must lead to trust, awareness and socialization efforts with various programs, and not only can rely on the role of the Directorate General of Taxes (tax officials), but also the active role of the taxpayers themselves. The tax system reform has been carried out by the government, the taxation system from the official assessment system becomes a self-assessment system which is a tax collection system that raises issues relating to competence, honesty, trust, awareness, ability and readiness of taxpayers to accept the burden to calculate the tax amount payable and submit tax reports accurately (Palil et al., 2013). The relationship between taxpayer compliance and awareness in fulfilling tax obligations has been analyzed by several scholars (Saad, 2010;Sapiei, 2012;Sapiei & Kasipillai, 2012). The role of sanctions and examinations in increasing tax compliance shows a positive influence (Alm J & McKee, 2016), but it is different from the results of other undergraduate studies that tax audits have not made a big impact on private tax in Nigeria (Kennedy & Anyaduba, 2014).
Other researchers examined taxpayer compliance, their findings provide evidence those fines, tax services (Fiscus) and tax awareness has a significant positive effect on tax compliance (Jatmiko, 2006;Muliari & Setiawan, 2011). Tax audit, service quality and tax sanctions have a positive effect on taxpayer compliance, while service quality and tax sanctions do not affect taxpayer compliance (Masruroh, 2013). Tax law and tax sanctions have a major influence on tax compliance (Marvina, 2015).
Studies that rely on legal science perspectives are carried out by other researchers with the assumption that either through the threat of fear it is possible to be examined or then subject to taxation sanctions will force the taxpayer to comply with their obligations (Allingham & Sandmo, 1972). Meanwhile, most of the research related to the relationship of increasing tax services to compliance was examined using a persuasive and cooperative approach between taxpayers and taxation authorities (Feld & Frey, 2002;Alm et al., 2010;Vossler et al., 2011;De Simone, 2013;Hauptman et al., 2014). The effectiveness of this collaboration is also influenced by the calculation of profit and loss, both on the taxpayer's side and the tax authority. Therefore, persuasive efforts must be balanced with consistent law enforcement actions. Some other scholars examine the relationship between awareness and trust with the government (Wahl et al., 2010;Mas'ud et al., 2014;Doerrenberg, 2015;Hartl et al., 2015;Gangl et al., 2015). The influence of personal norms such as awareness is positively related to tax compliance (Coricelli et al., 2010;Christian & Alm, 2014).
The discourse of the various research results has made it difficult to draw definite conclusions about how these factors affect taxpayer compliance. An objective method is needed to deduce various research results that may be saturated, but with inconsistent results. The research method using survey is a useful alternative to conclude various studies with inconsistent results. This research has added value in the form of a more comprehensive study compared to previous tax compliance studies, the factors that determine the level of tax compliance in Indonesia from various different points of view. This study also tries to describe the factors that have more influence on taxpayer compliance in Indonesia than other factors.
In general, the problems to which the answers are sought are important factors that cause the existence of several taxpayers who obediently pay taxes, but on the other hand there are also taxpayers who do not comply with paying taxes. One explanation of compliance with paying taxes is explained in a model called slippery slope framework (Kirchler et al., 2008;Prinz et al., 2014) as illustrated below:

Figure .1 Slippery Slope Framwork
The framework it is explained that the willingness of people to pay taxes is because they consider that the cost (risk) of non-compliance will be very high (enforced tax compliance) or because of the awareness that paying taxes is indeed a duty as a citizen (voluntary tax compliance). Enforced tax compliance is affected by power of authorities, mainly because of the government's ability to enforce tax rules through aspects of inspection and tax sanctions. Voluntary tax compliance is influenced by the level of government service to taxpayers.

LITERATURE REVIEW Expected Utility Theory
Expected Utility Theory (EUT) is a theory of decision making in situations of uncertainty (Harless, 1988;Moscati, 2016). This theory states that when there are several alternative possibilities of a risky event or uncertain outcome, one will make a decision based on the highest expected utility (EU). This EU value is the sum of the weighted utility rates calculated by summing the utility values of each choice after multiplying by probability or probability of occurrence (Mongin, 1997;Moscati, 2016). Social Contract Theory Rooted in the thoughts of Socrates & Plato about the importance of avoiding coercion regarding relations between the state and citizens. Hobbes (1651) citen in Van Gelder (1998), popularizes what is then called social contract theory (SCT). This concept begins with the real conditions faced by humans, which initially lived and multiplied into many, but without the government and the laws that governed them. The risk of this situation is the risk of chaos, where there is a possibility that the rights possessed by someone will be violated by other people, without something preventing it (Friend, 2016). In addition to social contract theory (SCT), other theories used in building this research model refer to the Theory of planned bahaviour explains that a person's behavior is planned based on the information he receives or their perceptions of existing conditions or possibilities that will occur (Ajzen, 1991).

Taxation services (Fiscus)
Service is a level of excellence to fulfill customer desires.(Fiscal servicesTaxation services) are assessed based on the perception of taxpayers who compare expectations to receive services and actual experience of services received (Alm et al., 2010;Ayuba et al., 2015). There are five dimensions that determine service quality, namely: physical evidence, reliability, responsiveness, assurance of certainty and empathy (Parasuraman, 1985).

Tax Penalties
Penalties or sanctions are actions in the form of punishment given to people who violate the rules. Sanctions are deemed necessary so that rules and regulations are not violated (Allingham & Sandmo 1972;Cho et al. 1996;Keller 1997;Arum, 2012). Indicators used in measuring tax sanctions use measurement of tax sanctions, namely: (a) Giving sanctions that are reasonable, (b) Imposing sanctions to educate taxpayers, (c) Tax sanctions to non-tolerant violators, (d) Imposing sanctions on taxes can be negotiated (Purnamasari et al., 2016;Muliari & Setiawan, 2011).

Tax audit
Tax audit is a series of activities, namely collecting and processing data, data or evidence information carried out objectively and professionally in accordance with the standard compliance tax audit examination examination aimed at implementing the provisions contained in applicable tax laws (Muehlbacher et al., 2012;Dulleck et al., 2016). Tax audits are measured by (a) taxpayer accounting checks, (b) examination of records and documents, (c) interviews and physical testing, (d) cross-testing with third parties, (e) promising something with the examiner's decision (Suhendra, 2010) Taxpayer awareness Taxpayer awareness is the behavior or attitude of an object which involve assumptions and feelings and a tendency to act on the object (Li et al., 2018). Thus it can be said that the awareness of taxpayers in paying taxes is a taxpayer's behavior in the form of views or feelings involving knowledge, beliefs, and reasoning along with a tendency to act according to the rules given by the system and the tax provisions, (Riahi-Belkaoui, 2004;Muliari & Setiawan, 2011;Christian & Alm, 2014: Castro & Scartascini, 2015. Taxpayers are said to have tax awareness if in accordance with the following: (a) Knowing taxation provisions, (b) Knowing tax functions, (c) Understanding taxation obligations, (d) Understanding tax functions (Muliari & Setiawan, 2011).

Taxpayer compliance
Taxpayer compliance is the awareness of the public to fulfill their tax obligations (McKee et al., 2018), namely registering themselves to obtain the taxpayer's ID number (NPWP), submits an notification letter (SPT) with complete and correct calculation of all tax objects, paying taxes based on the actual amount and on time. Taxpayer compliance indicators, namely compliance with paying taxes, compliance making financial statements according to actual conditions and compliance with reporting Notification Letter (SPT). Indicators about taxpayer compliance are: (a) calculation of tax rates, (b) timely payments, (c) tax reporting, (d) obtaining a tax ID number (NPWP). These measurements develop measurements of taxpayer compliance (Bradley, 1994;Siahaan, 2012).
Based on the description above, the conceptual framework of the research can be illustrated in the following figure:

Tax services (fiscus) affects on the taxpayer's compliance (H1)
One of the factors, namely normative beliefs related to tax services (Taxation services (Fiscus)), that is when going to do something, individuals will have beliefs about normative expectations from others and motivation to fulfill those expectations and with good service from tax officers, an efficient and effective taxation system, as well as tax counseling that motivates taxpayers to obey taxes, will make taxpayers have confidence or choose tax compliance. The results of previous studies provide evidence that tax services (Taxation services (Fiscus)) have a positive effect on taxpayer compliance (Jatmiko, 2006;Muliari & Setiawan, 2011).

Tax penalties effects on the taxpayer's compliance (H2)
The perspective of the law or the enforcement approach, Allingham & Sandmo (1972) begin by assuming that the threat of sanctions, both through the threat of fear of the possibility of being examined and then subject to taxation sanctions will force the taxpayer to comply with their obligations.

Tax audits effects on the taxpayer's compliance (H3)
Enforced tax compliance is affected by power of authorities, mainly because of the government's ability to enforce tax rules through aspects of tax audit. In general, the role of inspection in increasing tax compliance shows a positive influence, as shown in some previous studies (Kasper et al., 2015;Alm et al., 2016).

Taxpayer awareness affects on the taxpayer's compliance (H4)
Based on the belief subjective norm concept in planned behavior theory (TPB), the behavior of a taxpayer whether to comply or not in fulfilling his tax obligations, is influenced by the norms he believes in. These norms originate from the environment in which the taxpayer is located or originates which is often referred to as social norms (Ajzen, 1985).

Tax services (fiscus) affects on the taxpayer's Awareness (H5)
Taxpayers awareness in fulfilling the obligation to pay taxes depends on how tax officers provide the best quality service to taxpayers (Alm et al., 2012). Tax services can be associated with normative beliefs which are one of the factors of Theory of Planned Behavior. Normative beliefs explain when going to do something, individuals will have beliefs about normative expectations from others and motivation to fulfill those expectations.

Tax penalties affects on the taxpayer's Awareness (H6)
The use of social contract theory (SCT) and the slippery slope framework explains that besides enforced tax compliance, voluntary tax compliance also plays an important role in the success of tax collection (Benk et al., 2012;Gangl et al., 2015). Moreover, if the system used is asystem self-assessment, where the WP is given the trust to calculate, calculate, pay and then report the tax owed. Conclusions that are not much different also appear in various other studies (Kastlunger et al., 2013;Kogler et al., 2013;Damayanti et al., 2015).

Tax audit affects on the taxpayer's Awareness (H7)
Expected Utility Theory (EUT) is a theory of decision making in uncertain situations (Harless, 1989;Moscati, 2016). This theory states that when there are several alternative possibilities of a risky event or uncertain outcome, one will make a decision based on the expected utility (EU) highest. Examination is the obedience or awareness of taxpayers in carrying out asystem selfassessment that is filling, calculating, calculating, collecting, cutting, and reporting all tax obligations in accordance with applicable tax law regulations (Marvina, 2015).

Tax services (fiscus) effect on taxpayer's compliance: Taxpayer awareness as an intervening variable (H8)
Based on the social contract theory, if the taxpayer has a perception that the government is positive, such as by providing good service to taxpayers so that it makes taxpayers easy to carry out their obligations, will encourage taxpayers to be aware of doing accordingly by complying with their tax payment obligations. This is also in accordance with what was stated by Alm et al. (2012), that tax authorities have often treated taxpayers as criminals who must be monitored and punished if they violate.

Tax penalties effect on taxpayer's compliance: Taxpayer awareness as an intervening variable (H9)
Tax penalties or tax sanctions also encourage people to be more obedient, then using expected utility theory, the amount of tax sanctions that will be imposed on tax evaders will also prevent taxpayers from disobedient behavior. Some researchers confirm a positive relationship between the amount of sanctions and compliance behavior of taxpayers. Doran (2009) based on a study of tax behavior in America concluded that tax sanctions not only encourage tax compliance, but also tax sanctions determine the standard of behavior that the law is enforced against taxpayers.

Tax audits effect on taxpayer's compliance: Taxpayer awareness as an intervening variable (H10)
Theoretically, as explained by expected utility theory, the possibility that someone will be examined will cause concern for him. Therefore, the higher a person's concern is to be examined, the lower his desire to do tax evasion. According to Alm et al. (2012), individual taxpayers often seem to be too worried about the possibility of being examined, even though data from various countries shows that the examination coverage ratio on average is low. Several other studies are also in line with the theory, where examinations will reduce the desire of taxpayers to embezzle taxes (Alm et al., 2012;Kastlunger et al., 2009;Muehlbacher et al., 2016).

METHODS
This study uses a scientific approach, which uses theoretical structures to form hypotheses and use facts or empirical data to test hypotheses and draw conclusions. The population of this study is individual taxpayers who carry out free work recorded at Tax Offices (KPP) Pratama Makassar (KPP Pratama (North), KPP Pratama (South), and KPP Pratama West Makassar). Whose status is still active, until January 2017 there was 234,104 effective taxpayers?
The determination of which individual taxpayer sample to choose is to use a purposive sampling method with the type of sample selection based on consideration the judgment sampling, which is a type of sample selection that is not random whose information is obtained using certain considerations generally adjusted to the objectives or research problems (Indriantoro & Supomo, 2013). The criteria of the sample used in this study refer to Hair at al. (2010), which states that the minimum number of samples that must be taken is 15 to 20 times the number of variables used. The number of variables used in this study is 5 variables so that the minimum number of samples that must be taken is 5 X 20 = 100

RESULTS AND DISCUSSION Descriptive Statistics
In this study all constructs will be measured by at least four indicators with question items. Some of the criteria used to measure the reliability of a construct are the coefficient of Alpha Cronbach reliability, the coefficient of composite reliability and the proportion of extracted variance. The unidimensionality as a tool of reliability analysis and construct validity (Anderson & Gerbing, 1988). A construct will be considered reliable if the Cronbach Alpha coefficient value at least 0.6. Unidimensionality can also be measured by the convergent validity and reliability of the Bentler-Bonnet delta (Sjahruddin & Sudiro, 2013;Sekaran, 2014).
Descriptive statistics provide a description or descriptive of a data that is seen from the minimum, maximum, average, and standard deviation values as follows: The construct of the taxation services (Fiscus) indicates the value of the minimum score 12 and score a maximum value of 20. The average value for the construct of compliance services tax of 17.30> standard deviation of 2.022. This statistical result illustrates that respondents have the attitude that tax services (Taxation services) are a pleasant thing. Constructs of tax sanctions show a minimum score of 16, and a maximum score of 25. The average value for the tax sanction construct is 21.28> standard deviation of 2.256, this result illustrates that respondents have a high awareness of tax sanctions, whether individuals or groups agree tax compliance. In addition, this statistical value illustrates that respondents are motivated to adjust to compliance with tax obligations.
The examination construct shows a minimum score of 12 and a maximum score of 20. The average value for the inspection construct is 16.98> standard deviation of 2.127. These statistical results illustrate that respondents have a strong level of awareness in tax audit behavior. The construct of awareness and tax compliance shows a minimum score of 15 and 13, and a maximum awareness of 20, and compliance 20. The average value for the construct of awareness and compliance is 17.70,> standard deviation is 1.772 and 17.12> standard deviation is 1.976. These statistical results illustrate that respondents have high compliance in fulfilling their functions, namely calculating, paying, and reporting the amount of tax owed.

Evaluation Model
This research model uses five constructs, namely tax service (Taxation services (Fiscus)), tax sanctions, tax audit, taxpayer awareness and tax compliance. Evaluation of the Smart-PLS model is done by evaluating the measurement model (outer model) and structural model (inner model).

Convergent Validity Test
The rule of thumb is commonly used to assess convergent validity, namely the value of loading factor > 0.7 for studies confirmatory and the value of loading factors between 0.6-0.7 for studies exploratory and the value of AVE must be greater than 0.5 (Sholihin & Ratmono, 2013: 16). Even for the initial research stage the value of the loading factor of 0.5 -0.6 is considered sufficient (Chin 1988; cited in Ghozali & Latan, 2015: 74).

Test of Discriminant Validity
Measurement of discriminant validity by looking at the value of loading indicators to the measured construct is greater than loading other constructs (Sholihin & Ratmono, 2013: 16).

Testing the Structural Model (Inner Model)
Testing the structural model (inner model) is done to measure the level of variation in changes in exogenous constructs towards endogenous constructs (Jogiyanto, 2011: 72). The structural model in PLS was evaluated using R-square (R 2 ) for endogenous variables.

Direct effect
Direct effect is the influence that is directly measured from one variable to another. There are seven direct influences that can be tested in this study which are shown in following the table.  Table 3 shows the results of tests using the data processing program Smart-PLS. There are seven hypotheses of direct influence tested in this study. As shown in the table, there are two hypotheses of the direct effect of rejecting (H1 & H5) t-statistical value (t <1.96) at an alpha level of 5% with positive coefficient values of 0.076 and 0.011.

Indirect effect
Indirect influence is a measured influence of one variable (in this research used by X1, X2 and X3) on the variable taxpayer compliance (Y) through the mediator variable (Z), namely the awareness of the taxpayer. In testing the mediation effect, the output of significant test parameters is seen in the total effect table. If the t-statistic value on taxpayer awareness is <1.96, it is stated that it is not successful in mediating the effect of tax services (Fiscus) variable, tax sanction, and tax audit on the variable of taxpayer compliance. Conversely, if the value of t-statistic of taxpayer awareness variable> 1.96, it is stated that it is successful in mediating the effect of tax services (Fiscus) variables, tax sanctions, and checks on variable tax compliance. There are three indirect influences tested in this study, but research shows that H8 is rejected because statistically the t value of 0.106 is smaller than the t-value of statistics (t <1.96). The following is the value of the total indirect effect with the influence of mediation in this study: