THE IMPACT OF ADMINISTRATIVE SANCTION AND UNDERSTANDING OF INCOME TAX LAW ON CORPORATE TAXPAYER S COMPLIANCE

Transcription

1 THE IMPACT OF ADMINISTRATIVE SANCTION AND UNDERSTANDING OF INCOME TAX LAW ON CORPORATE TAXPAYER S COMPLIANCE Yanah Faculty of Shariah, Bandung Islamic University (UNISBA) Jl. Rangga gading No.8 Bandung yanah_cirebon@yahoo.co.id Abstract Tax is a source of potential revenue for the state in supporting economic development. But it is not easy to impose taxes on the people because if too high then people will be reluctant to pay taxes. Conversely, if too low it will be short of funds so that construction will not run. Therefore, the taxpayer voluntary compliance is the key to success in tax collection. Based on attribution theory taxpayer behavior can be caused by internal factors, external factors or a combination of internal-external factors in compliance paying income tax. Research purposes to determine the impact of the implementation of administrative sanctions and understanding of tax laws on corporate taxpayer s compliance. The method used was a survey method, kind of explanatory research. Sampling technique used is cluster sampling with a sample of corporate taxpayers as much as 30 respondents consisting of 11 state and 19 private enterprises. Instrument research using questionnaires. The analysis used is multiple linear regression. Hypothesis testing using the F test and t test. The results showed that there is a positive correlation between attitude taxpayer on the implementation of administrative sanctions and understanding of tax laws on taxpayer s compliance. This suggests that compliance of corporate taxpayers may be caused by external factors, namely the implementation of administrative sanctions, and is caused by internal factors, namely an understanding of the taxpayer's income tax laws. Thus, the higher the administrative penalty imposed the tax payers will respond by increasing compliance in paying income tax as if the violation is detected then the penalty to be paid is much greater than the income that was not reported. The higher understanding of the corporate taxpayer s income tax laws, then the taxpayer s compliance will increase because taxpayers can fill out, calculate and report tax return correctly, complete and timely. Keywords: Administrative Penalties, Attribution Theory, Taxpayer Compliance, Understanding of Income Tax Laws I. INTRODUCTION Taxes have a very important role in the implementation of development as a source of revenue to finance all expenditures including construction spending, however, is not easy to impose taxes on the people because if too high then people will be reluctant to pay taxes, on the contrary, if too low, it will be short of funds so that construction will not run. There is consensus in the literature that the public finances in developing countries many cases of tax evasion. This is done by not reporting or reporting, but not according to the actual situation on the income that can be taxed. This tax evasion has caused so much loss of potential tax revenue that could be used to reduce the burden on the state budget deficit. This is reinforced by evidence that the January 2010 announced the list of 100 largest tax delinquent by the Directorate General of Taxation ( Changes in the tax system of self assessment official assessment to which the taxpayer was entrusted to register, calculate, pay and report tax obligations taxpayers make 55

2 voluntary compliance as the key to the success of tax collection. For more data on tax receipts from in detail can be seen in Figure 1. Evaluation of Tax Revenue Period % Growth PPh Non migas PPN & PPnBM PBB BPHTB Source: The weekly report DJPb tax revenue through (processed data) (accessed 20/12/2010) Figure 1. Evaluation of Tax Revenue Period Taxpayers reduce their tax liability in two ways, namely legal and illegal. As proposed by Alm, Bahl and Murray (1990:605) ; Tax avoidance is legal any activity that lowers taxes, worker such as substitution between wage and nonwage compensation. Tax evasion is the reduction in tax liabilities by illegal means, such as underreporting income on tax returns. According Milack cited by Santoso (2008:86), there are three functions of tax administration in the self-assessment system, namely: 1. Education (counseling) 2. Services (customer service) 3. Supervisory (enforcement) To improve tax compliance, tax administration need to approach or treatment (compliance strategies) are different to the taxpayer by taxpayer's behavior towards compliance based on risk level. Compliance strategies that can be done to taxpayers based on risk level, can be seen in Table 1. Table 1. Risk level, Behavioral Compliance and Compliance Strategies Risk Level Behavioral Compliance Compliance Strategies High Tax payer s intention to deliberately disobey Full Enforcement (Inspection) Tax payers who do not comply but would Obey if any supervision Medium Taxpayer attempted to comply but Assistance to be obedient have not been successful (Counseling) Low Taxpayers who voluntarily obey Easements in the implementation Of tax obligations (Servicing) Source : Santoso, 2008 Attribution theory is a theory that describes the behavior of a person. A person's behavior may be caused by dispositional factors (factors within / internal), such as the nature, 56

3 character, attitude and so on, or caused by external circumstances, such as pressure situations or circumstances that force someone to do a particular act. According to Santrock (2003:647), attribution theory is views people as motivated to discover to the underlying causes of behavior as part of their effort to make sense of the behavior. According to Sparkman & Locander (1980:220), attribution theory is divided into two, namely correspondent inference and dimension four models, derived from the concept of Heider (1958) in explaining the causal relationship. This theory was developed with different objectives, namely; 1. Correspondent inference theory deals with inferences that are made about the actor whose behavior is observed. 2. The four dimension model deals with the inferences made about the environment in which the actor is behaving. According to Jones (1998) cited by Santrock (2003:647), human attribution divides into three dimensions, namely: 1. Internal or external causes Internal attribution includes all causes that are internal to a person, such as personality traits, intelligence, attitude and health. External attributions while covering all external causes of a person, such as social pressure, social aspects of the situation, money, weather and luck. Heider (1958) argued that the internal or external dimensions are key issues in attribution. 2. Stable or unstable causes The cause is stable or unstable is also involved in attribution. Whether it causes relatively enduring and permanent or temporary. 3. Controllable or uncontrollable causes Cause perceived as something that can be controlled or can not be controlled by another causality dimension, for example mood as a controllable while luck as a uncontrollable. According to Simon James et al cited by Santoso (2008:88), the notion of tax compliance (tax compliance) is a taxpayer has a willingness to meet its tax obligations in accordance with applicable rules without the need for the holding of the examination, thorough investigation, warning or threat and application of either legal or administrative sanctions. According to Santoso (2008:88), tax compliance is a state where the taxpayer meets all tax obligations and exercise the right of taxation. In the Practice Note on Compliance Measurement published by the OECD (2001), compliance is divided into two categories, namely: 1. Administrative compliance Includes compliance reporting and procedural compliance. 2. Technical compliance Includes compliance in calculating the amount of tax that would be paid by the taxpayer. Based on these two definitions of compliance, we can conclude that the administrative compliance is formal compliance, the compliance associated with the general provisions and procedures of taxation. While the technical compliance is material compliance, ie compliance related to the accuracy of SPT in determining the amount of tax to be paid. In a study of tax compliance, there are two main models that describe the level of tax compliance, namely: 1. Conventional models (first generation model) This model emphasizes the issue of tax evasion from the taxpayer (Taxpayers) and the factors that influence behavior. 2. The second generation model In this model problem taxpayers also determined by the behavior of others, namely the tax man (tax collector). In second generation models, the analysis is done on the behavioral patterns of both parties together to determine their response in the event of a change in tax rates, the probability of being detected, the level of penalty and bonus system for the tax man (Santoso, 2008:88). In this study, the authors use the conventional model approach is the development of models to determine the risk of non-compliance based on the taxpayer's tax reporting behavior by the taxpayer. Authors assume that the taxpayer is 57

4 rational so that the report will take into account the tax obligations of things that might be posed by reporting that he did, because basically all taxpayers objected to paying tax on earned income, but will have to consider the consequences of actions in the aftermath of the accomplishments. There are several studies conducted to determine the level of tax compliance, the studies that have been conducted by Tarjo and Kusumawati (2006), the survey results revealed that the self assessment system has not done well in Bangkalan because taxpayers are still many who do not do their own payment obligations tax payable, but for the level of tax compliance is pretty good but not for the taxpayer awareness but because of penalties. Feildstein Research, (1999), the traditional analysis tends to underestimate the deadweight losses. There are several reasons persistence on the compliance puzzle, but the most important reason for the lack of individual data in detail that provides a complete empirical specification of all the factors that affect adherence. This difficulty is evident when looking for a lot of data about individual decision-compliance evasion. Research conducted by Erard and Feinstein (1994), indicates that the taxpayer reporting behavior is strongly influenced by the tax system. Taxpayers choose how much income reported on their tax return by way of saving tax is by reporting earnings that are lower than actual income. Thus, although the taxpayer honesty there, but not the factor that most significantly affect tax compliance system. Research conducted by Ansori (2005), showed that compliance corporate taxpayers to comply with tax obligations in measuring the size of the tax savings that aims to minimize the tax burden. Done legally tax savings through tax avoidance which must be understood by both the provisions of the tax laws and changes and development. The study says that the understanding of corporate taxpayers regarding financial accounting and tax laws positively correlated to income tax savings, and from the results of analysis show that 90.80% corporate tax payers to save income tax. Research conducted by Irmadariyani (1999) attempt to obtain empirical evidence about the effect of understanding of financial accounting and tax law to corporate taxpayer s compliance in the town of Jember. The survey results revealed that the independent variables used in the study have a significant effect on corporate taxpayer s compliance. Based on research conducted by Jatmiko (2006) using the independent variable on the implementation of the attitude of taxpayers subject to a fine, the service tax authorities and tax consciousness, the individual taxpayer compliance as the dependent variable. It is known that the taxpayer attitudes towards the implementation of the penalty, the taxpayer attitudes towards service tax authorities and the taxpayer attitudes towards awareness of taxation has a significant positive effect on tax compliance. In addition, although the number of taxpayers from year to year, but there are increasing constraints that can hamper efforts to increase the tax ratio, the constraint is tax compliance. Based on research conducted by Allingham and Sandmo (1972), it is known that the level of sanctions (penalties) affect taxpayer compliance because, according to the concept of expected utility, the taxpayer will do underreporting penalty when the expected value is lower than the expected value of income that was not reported. To make the taxpayers are willing to report all of their income, the penalty should be set such that the expected value of the sanction (penalty) is greater than the expected value of income that is not reported so that when the penalty is set high, then the taxpayer will comply because if detected, fines have paid much greater than income that is not reported. This study aims to determine how much impact the implementation of administrative sanctions and understanding of tax laws on corporate taxpayer s compliance either partially or simultaneously. II. METHODOLOGY This study uses survey research for this study took a sample of the population and the use of questionnaires as the main data collection tool as proposed by Singarimbun and Effendi (2006:3). In addition, the survey method is also useful for the investigation conducted to obtain the facts of the existing symptoms and seek factual particulars, 58

5 both about social institutions, economics, or politics of a group or a region as described by Nazir ( 2009:56). Source of data obtained by the authors in this study consisted of primary data obtained through questionnaires by corporate taxpayers in the Cirebon Small Tax Office, and secondary data obtained from the Directorate General of Taxes through go.id, and Cirebon Small Tax Office. The population in this study, i.e all corporate taxpayers registered in Cirebon Small Tax Office. Based on information from the Cirebon Small Tax Office that not all corporate taxpayers domiciled in the city of Cirebon registered in Cirebon Small Tax Office due to changes in the status of the company that resulted in the Tax Office where registered taxpayer must change, so that about 74% as many as 371 corporate taxpayers are housed position in the city of Cirebon registered in Large Tax Office listed companies. Until the end of 2009, there were 130 WP Board consisting of 46 state enterprises and 84 private enterprises. Not all corporate taxpayers became the object of this research by the sheer numbers and the limitations of time and cost. Therefore samples were taken. The sample is part of the number and characteristics possessed by the population (Sugiyono, 2002: 73). Limitations in this study, the authors found difficulty in getting complete information on the individual taxpayer who is a population, because based on the tax code, Cirebon Small Tax Office is not willing to give data on the identity of the taxpayer, so it affects the sampling technique used by the writer. Sampling technique in this study is Cluster Sampling. This technique is used to determine the sample when the object to be examined or the source data is very extensive and complete record of the elementary unit in the population is not obtained, as stated by Sugiyono (2009:65). Cluster sampling technique is used in two stages, the first stage determines the sample area, and the second stage determines the people that exist in the sampling area. Population of the cluster is a subpopulation of the total population. Elements in nature are not homogeneous cluster, which is different from the elementary units in the strata. Each cluster member has a heterogeneous population resembles itself, as described by Nazir (2009:311), the sampling is done by using the formula: where : N = Large of Population Ni = Large of Sub Population Strata-i n = Large of Samples ni = Large of Samples Strata-i (Sugiyono, 2009:81). Sample size determination based on the table in Sugiyono (2009:71), to determine the sample size in stratified populations with N = 130 and standard error 10%, then the required sample size is n = 30, ( ) ( ) Thus, the required samples in this study were = 30 enterprises To obtain the necessary data, then the data collection. In this study, data collection techniques done by documentary study conducted by collecting data at department / institution relevant to the focus of research, and used questionnaires, the authors asked respondents who have been determined to fill out a questionnaire that has been made to inform the truth conditions and perceived by the respondents. Questionnaire form used is the Likert scale to answer questions with alternative answers 1, 2, 3, 4, or 5. Operationalization of the variables in this study are shown in Table 2 59

6 Table 2. Operationalization of Variables Variable Dimension Indicator Scale No Questionnaire Taxpayer s Attitude Penalties Refuse to pay tax Ordinal 1-10 On Implementation Interest Illegal tax evasion Of Administrative Increase Avoid taxes legally Sanction (X 1 ) Forced to obey Obidient voluntarys Understanding of Income tax laws Corporate Taxpayer s Compliance (Y) Tax calculation method Taxable Tax rate Implementation of Self Assessment System Understand how the calculation of income tax Understanding the classification of taxable income Applied for TIN Reported efforts to be registered as a taxable Fill out a tax return correctly Paying taxes on time Reported a tax return on time Provide data relating to taxation Ordinal Ordinal Based on the operationalization of the variables, it can be seen that the form of ordinal scale data measurement, while necessary for the purposes of regression analysis interval measurement scale, the ordinal measurement scales need to be transformed into interval scale measurement. Step-by-step transformation of ordinal data to interval data is as follows: 1. Look at each item respondents of a questionnaire distributed. 2. At each point determined how many people who got a score of 1,2,3,4 and 5 are referred to as frequency. 3. Each frequency is divided by the number of respondents and the result is called proportion. 4. Determine the value of the cumulative proportion by summing value proportions sequentially perkolom score. 5. Use the Normal Distribution table, calculate the value of Z for each cumulative proportion obtained. 6. Determine the high density values for each value of Z obtained (by using High density tables). 7. Determine the value of scale by using the formula: NS = (Density at Lower Limit) (Density at Upper Limit) (Area Below Upper limit) (Area below Lower Limit) 8. Determine the value of transformation by the formula : Y = NS + 1+ NS min (Riduwan & Kuncoro, 2009:30). 60

7 In this study, validity and reliability tests conducted by One Shot method using SPSS software version 19. Analysis of data using multiple linear regression because the author intends to predict how the state (rise and fall) dependent variable (criterion), when two or more independent variables as factors predictor manipulated (dinaik lower value). Considerations in selecting authors as a method of multiple regression analysis because the variables X 1 and X 2 in this study are independent (not related), as the assumption in the use of regression proposed by Gujarati (2009:203), namely To Operate within the framework of the linear regression model, specifically we assume the following: (1) Zero mean value of U i (2) No serial correlation (3) homoscedasticity (4) Zero covariance between U i and each X variable (5 ) No Specification bias (6) No exact collinearity between the X variables. To ensure that the variables are normally distributed and there are no stochastic disturbances or stochastic error term, it is necessary to test the normality of the data, multikoleniaritas, heteroscedasticity and autocorrelation. III. RESULT AND DISCUSSION Variable Data Testing of Taxpayer Attitude at the implementation of administrative sanctions (X 1 ) Having tested the validity and reliability of translated is Legally Avoid in paying income tax the questionnaire variable X 1, it can be seen in the table Item Statistics that whole grains have an answer statement averaging around 3 or when means tax payers avoid the boundary -limit allowed and does not violate the law. As can be seen in Table 3. Table 3 Item Statistics Mean Std. Deviation N item item item item item item item item item item Corrected Item-Total Correlation is the correlation between the item score with total score of items that can be used to test the validity of the instrument. Correlation score of item 1 to the total score was 0.507, the correlation score point 2 with a total score is and so on. Furthermore, to determine whether or not the statement is valid should be compared with the r table. With n = 30, 5% error level obtained table r = For the 1% error level obtained table r = Decision-making If r count is a positive and r count > r table, then the item is valid If r count is negative or r count < r table, then the item is not valid. 61

8 R count can be seen in column Corrected Item- Total Correlation contained in Table 4, because the r count is positive and r is greater than the r table it can be concluded that all the statements of the variables X 1 are valid. Scale Mean if Item Deleted Table 4 Item-Total Statistics Scale Variance if Item Deleted Corrected Item- Total Correlation Cronbach's Alpha if Item Deleted item item item item item item item item item item After all the statement is valid, then the next test is to test the realibility of questionnaire. Way of decision-making : If r Alpha is positive and Alpha> r table, then reliable. If r Alpha is negative or r Alpha <r table, it is not reliable. R Alpha (Cronbach alpha) can be seen in the table that is score Reliability Statistics is 0863, while we have r tables previous search is equal to for a significance level of 5% and for the significant level of 1% Cronbach's alpha is a measure of reliability, especially lower limit of reliability acceptable in the survey. Mathematically, reliability is defined as the proportion of respondents that heterogeneity will generate different responses from the respondents. Response from respondent answers will vary because each has a different opinion, not because questionnaires were confusing and multiinterpretation. Based on Table 5 Reliability Statistics, can be seen that r Alpha > r table, means that the X1 questionnaires are reliable. Once the questionnaire has been declared valid and reliable, it has been worth the questionnaire distributed to the respondents to conduct research. Table 5 Reliability Statistics Cronbach's Alpha N of Items

9 Testing variable data of Understanding Income Tax Act (X 2) Based on the results of testing the validity grains have an answer statement averaging around and reliability of the questionnaire variables X 2, it 3 or when translated is quite understand Income can be seen in the table Item Statistics that whole Tax Act. As can be seen in Table 6. To test the validity of the instrument variable X 2 done by comparing the r Scale Mean if Item Deleted Table 6 Item Statistics Mean Std. Deviation N item item item item item item item item item item Table 7 Item-Total Statistics Scale Variance if Item Deleted count in table 7 contained in the column Corrected Item-Total Correlation with r table. Corrected Item- Total Correlation Cronbach's Alpha if Item Deleted item item item item item item item item item item After than, apparently r count > r table, so it can be concluded that all the statement in the questionnaire variables X 2 are valid. After all the statement is valid, then the next test is to test the reliability of questionnaire the variable X 2. Based on Table 8 Reliability Statistics, can be seen that r 63

10 Alpha> r table, then the questionnaires X 2 are reliable. Once the questionnaire has been declared valid and reliable, it has been worth the questionnaire distributed to the respondents to conduct research. Table 8 Reliability Statistics Cronbach's Alpha N of Items Testing Data of Variabel Taxpayer s Compliance (Y) Based on the results of testing the validity grains have an answer statement averaging around and reliability of the questionnaire variable Y, can 3 or when translated is reasonable wee behaved be seen in Table 9. Item Statistics, that whole in pay income tax. Table 9 Item Statistics Mean Std. Deviation N item item item item item item item item item item To test the validity of the instrument variable Y done by comparing the r count in Table 10 contained in the column Corrected Item-Total Correlation with r table. 64

11 Scale Mean if Item Deleted Table 10 Item-Total Statistics Scale Variance if Item Deleted Corrected Item- Total Correlation Cronbach's Alpha if Item Deleted item item item item item item item item item item After than, apparently r count r> r table, so it can be concluded that all the statement in the questionnaire variable Y is valid. After all the statement is valid, then the next test is to test the reliability of questionnaire the Y variables. Based on Table 11 Reliability Statistics, can be seen that r Alpha > r table, then the Y questionnaires are reliable. Once the questionnaire has been declared valid and reliable, it has been worth the questionnaire distributed to the respondents to conduct research. Table 11 Reliability Statistics Cronbach's Alpha N of Items Testing of Normality Data Data normality test to know that the data were normally distributed or not. Normality test data using the standardized residual histogram and PP plot of standardized residuals. If the data are normally distributed histogram otherwise normal, and if the PP plot forming a diagonal line then the data is normal. In figure 2 and 3 can be standardized residual histogram and PP plot the standardized residuals. 65

12 Figure 2. Histogram Standardized Residual Figure 3. PP plot Standardized Residual Based on Figure 2 and 3 it can be seen that the standardized residual histogram and PP plot shows the pattern of standardized residuals The model is used to analyze the multiple linear regression model, as the authors intend to predict how state (rise and fall) dependent However, regression analysis is not always smooth analysis can be used, there are normally distributed data. Based on this it was concluded that the data were normally distributed Testing of Regression Problems variable (criterion), when two independent variables as predictors manipulated (hung upside down in value). The model equation is as follows: some serious problems encountered in regression analysis, so the authors tested the assumptions made on the methodology to determine these 66

13 assumptions are met or not. Tests performed are as follows: Multikoleniaritas Multikoleniaritas is a condition in which the independent variables in the regression equation had a correlation (relationship) are close to each other. The characteristics of the multikoleniaritas i.e if the regression equation has the value of R 2 is high or very high, F count is also high but lower t- counted. The existence multikoleniaritas causing the problem: the regression coefficient is positive in the simple regression could turn negative in multiple regression or otherwise, fluctuations in the value of the estimated regression coefficients are very large, if the independent variables correlated with each other, these variables explain the same variance in estimating the dependent variable. Thus, the addition of independent variables does not affect anything. Regression free multikoleniaritas characterized by: VIF values ranged from number 1 Tolerance values ranged from number 1 In table 12, we can see the value of VIF for both independent variables ranged from number 1, which is 1,024 respectively. as well as the tolerance value of each variable rate ranging from 1, which is Table 12. Collinerity Statistics Model 1 (Constant) The Attitude of Taxpayer on implementation administrative sanction Understanding of Income tax laws It can be concluded that the regression model is not impaired multikoleniaritas Collinearity Statistics Tolerance VIF a. dependent variable : Taxpayer s compliance disturbances can make the results of the analysis biased and incorrect statistical test and confidence Heteroscedasticity Heteroscedasticity occurs because of changes in circumstances that are not reflected in the specification of the regression model. For example, changes in the economic structure and government policy which may result in changes in the level of accuracy of the data. In other words, heteroscedasticity occurs if the residuals do not have constant variance. Heteroscedasticity intervals for the parameter estimates are also less precise. Detection of the symptoms of heteroscedasticity is to look at the pattern of the scatter diagram, namely the difference between the predicted value of Y observations. If the scatter diagram that there are certain patterns to form a regular, impaired regression heteroscedasticity 67

14 If the scatter diagram does not establish a pattern or random, regression is not impaired heteroscedasticity Figure 4. Scatterplot Standardized Predicticted Value Scatter diagram above was not a pattern or random. Thus, it can be said that the regression is not impaired heteroscedasticity Autocorrelation Autocorrelation is the main cause of the error specification, such as neglect of an important variable or incorrect functional form. If the residuals in the regression equation contains autocorrelation, then use the least squares method raises several issues, namely: the standard error estimates assess a lower error variability, confidence intervals and testing using t and F distributions can no longer be applied appropriately, the standard error of regression coefficients assessing variability of the regression coefficients are underestimated. The approach used to test for autocorrelation writer or not is the Durbin Watson test, with the provision of decision-making: If DW > upper limit (d U), then there is no autocorrelation If DW < lower limit (d L), then the autocorrelation If d L <DW <D U can not or are not known to occur autocorrelation In Table 13, note that the value of Durbin Watson = 2.174, with DW tables at = 0.05, n = 30, k = 2, then d L = 1,284 and d U = 1,567. Accordingly, because DW count> d U it does not happen autocorrelation interference. Model R R Square Table 13 Model Summary b Adjusted R Square Std. Error of the Estimate Durbin-Watson a a. Predictors: (Constant), Understanding of Income tax laws, Taxpayer s attitude on Implementation of Administrative Sanction b. Dependent Variable: Taxpayer s Compliance Accordance with the rules in the multiple regression analysis as suggested by Gujarati (2009:203), that a regression equation should have free data of multikoleniaritas, heteroscedasticity and autocorrelation in order to obtain a good 68

15 regression equation and not biased. From the test results it can be seen that the classical assumption of independent variables free from interfence Discussion of Result of Statistical Analysis After analysis using multiple linear regression method, the obtained regression equation Where: multikoleniaritas, heteroscedasticity and autocorrelation. Thus, all of the variable X can be used in the regression analysis. Y = Taxpayer Compliance X 1 = Taxpayer Attitude at The implementation of Administrative Sanctions X 2 = Understanding Income Tax Act Model Table 14 Coefficients a Unstandar B Std. Error Standardized Coefficients 1 (Constant) Taxpayer s attitude on Implementation of Administrative sanction Understanding of Income tax laws a. Dependent Variable:Taxpayer s Compliance Beta T Sig Discussion of Hypothesis 1 Regression analysis was conducted to determine the relationship between the independent variable (X) with the dependent variable (Y). Hypotheses used in this study is H 0 :βx 1 βx 2 0, the independent variable is less than or equal to zero H a :βx 1 > βx 2 >0, the independent variable is greater than zero Based on Table 14, it is known that the independent variable the attitude of taxpayer on the implementation of administrative sanctions have regression coefficients with positive signs for This shows that the direction of correlation influences the attitude of the taxpayer on the implementation of administrative penalties is positive., Meaning that if the enhanced administrative sanctions taxpayers will respond by increasing compliance in paying income tax. Independent variable understanding of the income tax law has a regression coefficient of with a positive sign. This shows that the direction of correlation influences the understanding of income tax law is positive, meaning that if the taxpayer understanding of the income tax law increased the tax compliance will increase because taxpayers know the correct regarding general provisions and procedures of taxation so as to fill tax form and reported correctly and on time. In Table 13, note that the correlation coefficient (R) = 0.903, close to the value 1. It means that the relationship between the independent variable (X 1.2) and dependent (Y) very closely. Correlation between the dependent and independent variables are positive, meaning: if the value of X increases, the increase will be responded with Y. This suggests that, if the value of the taxpayer s attitudes on the implementation of the administrative sanction and the value of understanding of the income tax law to increase the tax compliance will increase. Thus, one hypothesis is proven true. Therefore it can be concluded that H 0 = rejected. 69

16 Discussion of Hypothesis 2 To find out how much the independent variables affect the dependent variables simultaneously and partial and independent variables to determine which one is more significant in influencing the dependent variable, the authors test the hypothesis. Hypotheses used in this study is H 0: β 1 β 2, variable X 1 significantly less or equal to the variable X 2 H a :β 1 >β 2, variable X 1 is more significant than the variable X 2 Statistical F test is performed to determine that all the independent variables were entered into the model have an influence simultaneously on the dependent variable. If the probability value < significance level (α = 0.05), then the model can be accepted and vice versa if the probability value > significance level (α = 0.05), then the model is rejected. Table 15 shown that the value of the probability is (Sig = 0.000) <0.05 then the model equation used, acceptable, means that the variable attitude taxpayer on the implementation of administrative sanctions and understanding of tax laws simultaneously affect the compliance of corporate taxpayers. Table 15 ANOVA b Model Sum of Squares Df Mean Square F Sig. 1 Regression a Residual Total a. Predictors: (Constant), Understanding of income tax act, Taxpayer s attitude on the implementation of administrative sanction b. Dependent Variable: Taxpayer s Compliance Statistical t-test performed to prove that the partial independent variables (X 1) has a significant effect on the dependent variable (Y). Based on Table 14, note that the probability value for variable the taxpayer s attitude on the implementation of administrative penalties on corporate taxpayer s compliance is a partial (Sig = 0.000) <significance level (α = 0.05), meaning that the model can be accepted in partial taxpayer s attitudes on implementation of administrative sanction effect on corporate taxpayer s compliance. Probability value for the variable understanding of income tax laws on corporate taxpayer s compliance is a partial (Sig = 0.003) <significance level (α = 0.05), means that the model is acceptable, then in partial understanding of the income tax laws affect compliance corporate taxpayers. Therefore, it can be concluded that the variables taxpayer s attitude on the implementation of administrative penalty is more significant than the variable understanding of income tax laws in influencing corporate taxpayer s compliance. So hypothesis 2 proven true, therefore it can be concluded that H 0 = rejected. Influence of the taxpayer s attitude on implementation of Administrative Sanction and Understanding of Income Tax Act Against Corporate Taxpayer s Compliance Allingham and Sandmo (1972) suggest that entire income tax, penalties should be set higher underreporting taxpayers will perform if expected then the taxpayer will comply because if detected value of administrative penalties is still lower than by the tax officers to pay administrative penalties the expected value of income that was not more greater than the income that is not reported. reported, so as to make it mandatory to report the Moreover, Alm & McKee (1998) also suggested 70

17 that the increased penalties would increase tax compliance for substantially increasing the level of sanctions to eliminate evasion. Likewise Andreoni, Erard and Feinstein (1998) says that there is a significant relationship between the level of financial penalties and tax compliance, so as to increase tax compliance should be enforced severe penalties for any noncompliance. The study of authors found that the more positive attitude of the taxpayer on the implementation of administrative penalties also increase taxpayer compliance. The results support the research Andreoni, Erard and Feinstein (1998), Alm & McKee (1998), Alm, Bahl and Murray (1990), Jatmiko (2006), Allingham and Sandmo (1972), Dubin, Graetz and Wilde ( 1987), which also shows that the attitude of the taxpayer on the implementation of administrative penalties in a positive effect on tax compliance. Andreoni, Erard and Feinstein (1998), stated that the understanding of the tax laws had significant relationship to tax compliance, because if taxpayers understand the tax laws that taxpayers can comply with their tax obligations correctly, so that tax laws should be made as simple as possible in order not to complicate the taxpayer in the study. In the opinion Krause cited Santoso (2008) that the taxpayer's knowledge or understanding of the tax laws may affect whether or not abiding taxpayers. Krause opinion is in line with the OECD (2001) which states that understanding the taxpayer will determine the level of tax compliance. The results Ansori (2005) states that the taxpayer understanding of the tax law positively correlated to income tax savings, because the savings tax is legal, tax evasion through the course should understand well about the provisions of tax laws and changes and development. Results of the study authors found that a higher understanding of the taxpayer's income tax laws, it is also increasing taxpayer compliance. IV. CONCLUSSION Based on the description that has been said, it can be concluded that the attitude of the taxpayer on the implementation of administrative sanctions have a positive and significant impact on corporate taxpayer s compliance. This suggests that, if the implementation of enhanced Thus, the results are consistent with the results of the research the author conducted by Andreoni, Erard and Feinstein (1998), Ansori (2005), Irmadariyani (1999) and Reinganum & Wilde (1988) which shows that the understanding of the income tax law affects on tax compliance. Based on attribution theory proposed by Kelley (1973), it can be concluded that most corporate taxpayers pay income tax caused by external factors, namely the administrative penalty by 62%, and tax compliance is caused by internal factors, namely taxpayer s understanding of the income tax laws by 32%. Testing of Feasibility Model Based on Table 15, note that the probability value of F count is (Sig = 0.000) <0.05 significance level, it can be concluded that the model equation used was appropriate. The coefficient of determination (R 2) indicates the percentage of variation in the dependent variable that can be explained by the variation of the independent variable. R 2 value is small, means the ability of the independent variables in explaining the variation in the dependent variable is very limited. Value close to one, means that the independent variables provide almost all the information needed to predict the variation in the dependent variable. In Table 13, note that the coefficient of determination (R 2) = However, in this study the authors use the adjusted R square is the value of R 2 is adjusted so that the picture quality is closer assessment of the model in the population. From Table 13, note that the value of adjusted R square = 0.801, meaning that 80.1% compliance corporate taxpayers affected by attitudinal factors taxpayer on the implementation of administrative sanctions and understanding of tax laws, while 19.9% (100% %) affected by other factors that are not described in this regression equation, such as the service provided by the tax authorities to taxpayers. administrative sanctions, the corporate taxpayers will respond by increasing compliance in paying income tax. Understanding of the income tax law is a positive and significant effect on corporate taxpayr s compliance. This means that, if the 71

18 taxpayer understanding of the income tax laws improved, it will increase taxpayer s compliance because taxpayers know the correct regarding general provisions and procedures of taxation so V. ACKNOWLEDGMENT as to fill up, calculating and reporting tax form with true, complete and timely. The authors wish to thank the reviewers who have reviewed this article. REFERENCES Allingham, Michael.G & Sandmo, Agnar, 1972, Income Tax Evasion : A Theoritical Analysis, Journal of Public Economics, 1, Alm James, Bahl Roy& N Murray, Mathew, 1990, Tax Structure and Tax Compliance, The Review of Economics and Statistics, 72 (4), Alm, James and McKee, Michael, 1998, Extending the Lessons of Laboratory Experiments on Tax Compliance to Managerial and decision Economics, John Wiley & Sons, 19 (4/5), Andreoni James, Erard Brian & Feinstein Jonathan, 1998, Tax Compliance, Journal of Economic Literature, 36 (2), Ansori, Surya, 2005, Dampak Pemahaman Wajib Pajak Badan Mengenai Akuntansi Keuangan dan Undang-Undang Perpajakan Terhadap penghematan Pajak penghasilan, Jurnal Riset Akuntansi, 7 (1), Dubin, A, Jeffrey, Graetz, Michael J. & Wilde, Louis L 1987, Are We a Nation of Tax Cheaters? New Econometric Evidence on Tax Compliance, The American Economic Review, 77 (2), Erard, Brian & Feinstein, Jonathan S 1994, Honesty and Evasion in The Tax Compliance Game, The RAND Journal of Economics, 25 (1), 1-19 Feildstein, Martin 1999, Tax Avoidance and The Deadweight Loss of The Income Tax, The RAND Journal of Economics, 19(1), Gujarati, N. Damodar 2009, Basic Econometrics, 5 th edition, New York : The McGraw Hill Companies Irmadariyani, Ririn, 1999, Pengaruh Pemahaman Akuntansi Keuangan dan Undang- Undang Perpajakan Terhadap Kepatuhan Wajib Pajak Badan Dalam Membayar Pajak Penghasilan, Bandung : Universitas Padjadjaran Jatmiko, Nugraha, Agus, 2006, Pengaruh Sikap Wajib Pajak Pada Pelaksanaan Sanksi Denda, Pelayanan Fiskus dan Kesadaran Perpajakan Terhadap Kepatuhan Wajib Pajak Orang Pribadi, Semarang : Universitas Diponegoro Nazir, 2009, Metode Penelitian, Bogor : Ghalia Indonesia. Organization For Economic and Corporation Development (OECD), 2001, Compliance Measurement, Practice Note :1-23 Reinganum Jennifer F & Wilde Louis L, 1988, A Note on Enforcement Uncertainty and Taxpayer Compliance, The Quarterly Journal of Economis, 103 (4), Riduwan & Kuncoro, Achmad, Engkos, 2009, Cara Menggunakan & Memaknai Analisis Jalur (Path Analysis), Bandung : Alfabeta Santoso, Wahyu, 2008, Analisis Risiko Ketidakpatuhan Wajib Pajak Sebagai Dasar peningkatan Kepatuhan Wajib Pajak, Jurnal keuangan Publik, 5 (1), Santrock, W, John, 2003, Psychology, 7 th edition, International Edition, New York: McGraw Hill 72

19 Singarimbun, Masri & Effendi, Sofian, 2006, Metodologi Penelitian Survai, Edisi Revisi, Jakarta : LP3ES Sparkman,Jr, M, Richard & Locander, B, William, 1980, Attribution Theory and Advertising Effectiveness, Journal of Consumer Research, 7(2), Sugiyono, 2009, Statistik Penelitian, Bandung : CV. Alfabeta. Tarjo & Kusumawati, Indra, 2006, Analisis Wajib Pajak Terhadap Pelaksanaan Self Assessment System, JAAI, 10(1), Undang-Undang No.36 Tahun 2008 Tentang Pajak Penghasilan www. pajak. go.id accessed on dated December, 20, 2010 General Statement : Attachment 1 : Questionnaire Business Type : BUMS/BUMN (*Pick up as required) Special Statement : Instructions for Filing : 1. Please choose the answer that you think Mr/Mrs appropriate, by way of give check the mark ( ) in the field provided. 2. Description : 1. I had never done 2. I ve done 3. Sometimes I do 4. I often do 5. I always do 3. Example : No Item Statement Choice I comply with the tax laws, even when I see others break Please check the tick ( ) in the fields according to the choice of Mr/Ms Description : 1: I ve never done 2: I ve done 3: I sometimes do 4: I often do 5: I always do No Item Statement Choice SPT late because I do not necessarily obtain financial penalties from the tax man 2 I check the completeness and correctness oftaxreturns to report to the tax office before being exposed to financial penalties 3 When there is an error in filing tax return, I did not fix tax return because rarely inspected by the tax man 4 I pay taxes on installment due interest pinalties should I pay light enough 5 I pay annual and period taxes on time in order not to be interest penalized by the taxman 6 I calculate taxes with caution so as not to be sanctioned interest 73

20 7 Although sanctions increase of 50% given the revelations untruth after over 2 years of tax returns. It did not cause me to be more careful in filling SPT 8 Sanctions increase of 100% is imposed on tax deficiencies on SKPKBT makes me feel scared violation 9 Although many people violations of taxation because do not necessarily administrative sanction, I do not want to take the risk of administrative penalties 10 The more often given administrative sanctions by the tax man, the more compelled I am to learn the rules and apply sanctions for violations of taxation Description: 1: It is not true 2: Not True 3: Do not Know 4: True 5: Very True No Item Statement Choice Taxpayer code entity in an annual tax return is 1771 and 1771S tax returns 12 Income tax rate for corporate taxpayers by 25% 13 Donations including taxable income 14 Subject to income tax when business activities have benefited 15 Tax payments can only be made at the office of the tax 16 SPT is a letter by the time the taxpayer is used to report the calculation and / or payment of tax payable in a tax year 17 Interest income in the form of deposits and other savings may be subject to final tax 18 Insurance premiums including taxable income 19 Share of profits received or accrued by a member of a limited partnership whose capital is not divided into shares, fellowship associations, firms and joint venture investments, including holders of units of collective investment contracts not including taxable 20 To calculate the amortization of intangible assets that the group has a useful life of 8 years, the rate is based on the declining balance amortization amounted to 25% Description: 1: Do not ever do 2: Once I did 3: I sometimes do 4: Often I do 5: I always do No Item Statement Choice I register with the Tax Office when my income exceeded exemption (PTKP) 22 I reported my efforts in order confirmed as a Taxable Person (PKP) 23 If there is a mistake in filing tax return, I filed a written application and fill in the new SPT correction 24 In completing the annual tax return, once the correction by the tax man 25 My actions towards the correction is filling SPT re-fit the data I have 26 I reported to the tax office when I know there is a difference between the actual tax payable by the tax credit (less pay) 74

21 27 I pay taxes after the due date 28 I reported SPT accordance with the specified time 29 I provide data relating to taxation when requested by the tax man 30 I just reported one type of business alone although I have other types of business Signature filler (the respondent) Thank you for your willingness to help me in filling out this questionnaire. Yours sincerely Yanah 75