Tax and levy burden of wages in the Czech Republic

AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH TAX AND LEVY BURDEN OF WAGES IN THE CZECH REPUBLIC a MICHAL KRAJŇÁK VSB-Technical University of Ostrava, Faculty of Economics, Department of Accounting and Taxes, Sokolská třída 33, 702 00 Ostrava email: a Michal.krajna[email protected] This work was created in connection with scientific research project of the VSBTechnical University Ostrava no. SGS 2022/1. Abstract: The article deals with evaluating the influence of the tax and levy burden of wages on employment and tax revenue. The high tax and levy burden on wages is not favourable for the economy. The analysis results show that the increase in the tax and levy burden of wages has a negative effect on employment. When evaluating the effect of the tax and levy burden on tax revenue, the effect is different from the scope of used deductions. However, the long-term problem of the Czech Republic in comparison with other countries is not the tax burden on labour, but the related levy burden, which increases the price of labour and thus has a negative effect on the labour market from its transnational perspective. The analysis confirms that the Czech Republic remains among the countries with a high levy burden. To achieve the paper's objectives, methods of description, comparison, analysis, synthesis, and regression and correlation analyses are used. The added value and uniqueness of this article is, among other things, the fact that these indicators – effective, respectively, implicit tax rates on labour are calculated depending on the distribution of wages. Keywords: Effective Tax Rate; Nominal Tax Rate, Personal Income Tax; Implicit Tax Rate, Tax Reform, Social Security Contributions, Wage. 1 Introduction The article deals with analysing the tax and levy burden of income from dependent activity in the Czech Republic in the period 2003-2020 in several areas. First of all, it analyses how this tax and levy burden affects employment. The next part of the research analyses the impact of this tax and levy burden on the share of total tax revenue of the Czech Republic. One of the problems reducing the competitiveness of the Czech labour market is the high levy burden (Teplická and Daubner, 2013). Whether and how this levy and tax burden in the Czech Republic develops and changes in the mentioned period is the central topic of the last part of this research study. The article's main objective is to analyse how effective tax rate and implicit tax rate on labour affect employment and whether implicit tax rate on labour affects the share of personal income tax to the total tax revenue. Another goal is to evaluate the development of effective tax rates and implicit tax rates on labour and determine whether there is dependence between the development of the values of these indicators. The tax burden is represented by personal income tax, the levy burden by social security contributions, including public health insurance and social security premiums, and the contribution to the state employment policy. The nominal income tax rate has been linear in the Czech Republic since 2008. However, it does not say much about the actual tax burden, and it is more appropriate to use the effective tax rate (hereinafter ETR) to assess the tax burden. The indicator of the implicit tax rate on labour (hereinafter ITRL) is then used to express the total tax and levy burden of labour. Despite one nominal rate for social security contributions (hereinafter SSC), and since 2008 also for personal income tax, the tax burden is not the same for all taxpayers. The research question of this article is whether this burden affects employment and the share of personal income tax from dependent activity to the total tax revenue in the Czech Republic. The mentioned ITRL and ETR indicators are used to analyse this research question. The uniqueness and added value of this article are that these rates (ETR and ITRL The article's introduction is followed by an outline of the theoretical background with a focus on the personal income tax. The following section is focused on the characteristics of the methods (regression and correlation analysis) and input data used. A linear regression model is used to analyse selected dependences. Further, the analysis is the main part of the text. The last part of the paper summarises the article's results and describes the limits of the analysis. ) are calculated according to the distribution of wages in the Czech Republic as weighted averages. Another uniqueness of this text is the scope of the research period of 18 years, as studies analysing these factors in the conditions of the Czech Republic in relation to the distribution of wages and such a long period have not yet been conducted. 2 Literature review The tax and levy burden of wages is the subject of research in the Czech Republic and abroad. One of the factors influencing the tax and levy burden on labour is the level of employment. According to Dalenberg and Partridge (1995) the high tax burden on labour negatively influences employment. Similar findings that a high tax burden does not positively influence employment were found by Mark et al., (2000); Kosi and Bojnec (2006). ITRL or ETR are widely used to express the tax burden on labour. More about these indicators, e.g. Mankiw et al., (2009); Glday and Madl (2018). Grace (2018) states that the probability that a company will employ labour is 1.18 percent higher when current tax rates increase by one percentage point. Other studies (Cutler et al., 2018; Burda and Weder, 2016) state that personal income tax on employment cannot be determined unambiguously and depends on the sensitivity to the income tax rate. For this reason, employment may rise despite rising tax rates. There is a trend towards higher SSC and lower tax rates on personal income (Michaelis and Birk, 2006). In addition to personal income tax, payments for SSC are a significant factor influencing the amount of personnel costs (Teppererová, 2019; Prammer, 2019; Goudswaard and Caminada, 2015). The study Bauer and Riphahn (2002) analysing payroll and employment taxes in Germany, states that the employment rate is not negatively affected by the personal income tax but by SSC. According to the results of Bronchi and Burns (2001), the tax system of the Czech Republic is recommended by lowering SSC and increasing the reliance system on the personal income-tax system. Adam et al. (2019) mention that reducing the levy burden will positively affect employment. High personnel costs reduce the performance and competitiveness of companies. More about the performance of companies Knápková et al., (2014); Belas et al., (2020). The amount of the tax burden affects employment and the state´s total tax revenue. In the Czech Republic, personal income tax has significantly shared the state's total tax revenues, approximately 20% of the total tax revenue. Throughout the existence of income tax, there has been no significant volatility in this share, as in the USA (Garrett, 2009). Another important direct income tax is also corporate income tax (Konečná and Andrejovská, 2020; Moravec et al., 2019). On the contrary, Chernick and Reimers (2019) claim that indirect taxes show a more stable share of tax revenue than direct taxes. According to Yilmazkuday (2017), the nonrecommendation to increase income tax burden from dependent activities also follows. Rather, it is recommended to increase indirect or property taxes. More about property taxes e.g. Široký et al., (2015). Tax revenue is significantly affected by the constructional elements of taxes, so each tax reform changes the share of tax to the total tax revenue and the total amount of tax revenue (Mahdavi, 2008; Sanz-Sanz, 2016). The Laffer curve defines the theoretical relationship between tax rates and tax revenues. According to Lin and Jia (2019) in China, it is recommended to set the tax rate at 35% to maximize tax revenue. In determining the tax burden, the tax rate should be taken into account and the amount of the taxpayers' income or social status (Jordaan and Schoeman, 2015). - 112 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH While the personal income tax burden is generally lower for families with children, mandatory payments on the SSC often do not take these factors during calculating these payments (Alvarez-Martinez and Polo, 2014). The already mentioned problem of the Czech Republic is the persistently high levy burden of labour (Kramer et al., 2016; Lyková, 2015). The overview of studies shows that studies evaluating the impact of ETR, resp. ITRL on employment, resp. tax revenue, as well as the analysis between the development of ETR and ITRL have not yet been performed under the conditions of the Czech Republic. This demonstrates the certain uniqueness and added value of this article. 3 Methodology A linear regression function generally determined (1) is used to model the dependence of the share of personal income tax from dependent activity to the total tax revenue according to the value of effective and implicit tax rates (1), 𝑌= 𝛽0+𝛽1𝑋1+𝛽2𝑋2+𝛽3𝑋3+𝛽4𝑋4+ 𝛽5𝑋5+ 𝛽6𝑋6+𝜀, (1) where Y is the dependent variable expressing the share of tax revenue of the personal income tax from a dependent activity to the total tax revenue of the Czech Republic, the values X1 to X6 are the average values of the indicator ITRL, resp. ETR calculated as a weighted average according to the empirical distribution of wages in the relevant year. Specifically, X1 expresses the ITRL for the case where only the deduction per taxpayer is applied, X2 the situation where, in addition to the deduction per taxpayer and the deduction per child are also used to reduce the tax liability, X3 when the deduction per taxpayer and deduction for two children are used. The structure of indicators for ETR is similar, specifically X4 expresses ETR if only the deduction per taxpayer is applied, X5 cases where the deduction for taxpayer and for one child and X6 when the deduction for taxpayer and for two children are used. ETR is generally determined by (2), ITRL according to (3), 𝐸𝑇𝑅 = 𝑇 𝑌, (2) 𝐼𝑇𝑅𝐿= 𝑇+ 𝑆𝑆𝐶𝐸+𝑆𝑆𝐶𝐼 𝑌+ 𝑆𝑆𝐶𝐸, (3) where T is the tax liability, Y the taxpayer's income, SSCE is the social security contributions paid by the employer, SSCI is the social security contributions paid by the employee. According to data from income tax returns by individuals are deductions per taxpayer and children the most used deductions, specifically for the year 2020 the relief for the taxpayer was applied in 99% of tax returns, a tax credit for children in approximately 30% of tax returns (Financial Administration, 2022a). Compared to other tax deductions, these two are used by taxpayers for tax optimization the most often. A limiting aspect of this model may be the fact that non-taxable parts of the tax base are not taken into analyses, which do not affect SSC, but influence the tax base and the taxpayer's tax liability. Tax statistics of the Ministry of Finance of the Czech Republic (Financial Administration, 2022b) show that the most used nontaxable parts of the tax base for 2020 were deductions for private life insurance and amount of trade union contributions (approximately 20% of tax returns). However, no information is available on the average applied values of these non-taxable parts. For this reason, it is abstracted from non-taxable parts of the tax base. On the contrary, the uniqueness and added value of the study is the fact that effective, resp. implicit tax rates in the regression model are calculated as weighted values according to the percentage of employees in the bands of gross wages. In addition to the analysis of factors influencing the share of personal income tax to the total tax revenue another such indicator is the motivation to work, which is expressed by the employment rate (Rick et al., 2018). According to Reiss and Schuster (2020), this indicator significantly affects ITRL 𝑌= 𝛽0+𝛽1𝑋1+𝛽2𝑋2+𝛽3𝑋3+𝜀, (4) . Based on the above, relation (4) is formulated, where Y is the employment rate, X1 to X3 express the average values of the ITRL (X1 for the case if the deduction of only the taxpayer is applied, X2 per taxpayer and one child, X3 per taxpayer and two children). As follows from the part dealing with the theoretical background, one of the factors weakening the competitiveness of the Czech tax system isn´t the high tax but the levy burden. Regression lines will be constructed to examine whether the spread between the tax and levy burden decreases or increases. The angle between the regression lines is formed by two selection linear regression lines expressing the dependence between the characters y1 and y2 (Marill, 2004). Y1 is a variable expressing the tax burden on wages (ETR), y2 is a variable expressing the tax and levy burden on wages (ITRL ). The angle 𝑐𝑜𝑡𝑔 𝜑 is formed according to Toka et al., (2019) by (5), 𝑐𝑜𝑡𝑔 𝜑= �𝑟𝑥𝑦� 1−𝑟𝑥𝑦 2�𝑠𝑦 𝑠𝑥+𝑠𝑥 𝑠𝑦� , (5) where rxy is the correlation coefficient, r2xy is the coefficient of determination, sx is the variance of the values of the character x, sy the variance of the values of the character y. According to the aim of the paper, the following research hypotheses were formulated:  as the tax burden increases, the tax revenue increases,  the tax and levy burden of labour negatively affects employment,  the difference between the tax and levy burden of work doesn´t change in the Czech Republic. For analysis of the interdependencies between the selected indicators, the following resources have been applied:  tax revenue of the personal income tax from a dependent activity in the Czech Republic from 2003 to 2020 were sourced from Financial Administration (Financial Administration, 2022c),  employment rate and shares of employees in the bands of gross monthly wages were sourced from the Czech Statistical Office (Czech Statistical Office, 2022a and Czech Statistical Office, 2022b),  for calculation of ETR and ITRL Act. No. 586/1992 Coll. on Income Taxes, Act. No. 48/1997 Coll. on Public Health Insurance and Ac. No. 589/1992 Coll. on Premiums for Social Security and Contribution to the State Policy of Employment. About the availability of data, the period 2003–2020 is analysed. The year 2020 is the last year for which data are available. The shares of employees in the gross wage bands have been published in the database of the National Statistical Office (Czech Statistical Office, 2022a) since 2003. 4 Results Before the analysis of the dependence of the employment rate, resp. the share of personal income tax to the total tax revenue in relation with effective or implicit tax rates, the basic characteristics of each variable are performed using the tools of descriptive statistics. 4.1 Characteristic of input data The weighted average effective, resp. implicit tax rates are graphically captured as one of the analysis's input data in Figures 1a and 1b. These values are calculated according to the distribution of wages in the Czech Republic in the period 20032020 for each analysed year in cases where only the basic deduction per taxpayer (ETR0, ITRL0), deduction per taxpayer - 113 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH and one child (ETR1 and ITRL1) and deduction per taxpayer and two children (ETR2, ITRL2 ) are applied. Figure 1a. Effective Tax Rate (in %) Source: own research Figure 1b. Implicit Tax Rate on Labour (in %) Source: own research In addition to the weighted average of ETR and ITRL , the input database for the analysis is also formed by the values of the employment rate in the Czech Republic (EMP) and the share of personal income tax from the dependent activity to the total tax revenue (TR). For each variable, the minimum and maximum values, the mean and the median are indicated in Table 1. Tab. 1: Overview of the Data Used in the Regression Analysis Variable Minimum Maximum Mean Median Y - TR 18.98 24.08 21.45 21.27 Y - EMP 64.23 76.90 68.62 66.56 X1 - ITR 41.12 L0 44.20 42.59 42.80 X 2 - ITR 37,62 L1 42,33 39.63 39.49 X3 - ITR 33.69 L2 40.76 36.35 35.43 X4 - ETR 10.09 0 13.29 11.39 11.17 X 5 - ETR 5.12 1 9.64 7.39 7.20 X 6 - ETR -0.14 2 7.52 2.98 2.47 Source: own research 4.2 Correlation analysis As follows from the development of ETR and ITRL, the tax burden does not follow the same trend and differs according to the applied deductions. In general, it can be said that taxpayers without the application other deductions than for taxpayer increase the tax burden on labour, the opposite trend can be observed for taxpayers with children. In some cases, the ETR is even in negative values. The share of personal income tax to the total tax revenue is not wholly constant it clearly does not show an increasing or decreasing trend. Is the change in the share of tax revenue related to the development of ETR or ITRL values? Does the level of employment affect the tax burden on labour? Does SSC in the Czech Republic significantly increase the tax burden on labour? The following analysis attempts to answer these questions. Tab. 2: Correlation Analysis. Source: own research TR EMP ITR ITR L0 ITR L1 ETR L2 ETR 0 ETR 1 2 TR 1 EMP -0.034 1.000 ITRL 0.559 0 -0.170 1.000 ITRL 0.546 1 -0.335 0.970 1.000 ITRL 0.470 2 -0.524 0.889 0.970 1.000 ETR 0.759 0 0.425 0.713 0.571 0.380 1.000 ETR 0.737 1 -0.154 0.952 0.945 0.864 0.778 1.000 ETR 0.583 2 -0.491 0.894 0.970 0.988 0.466 0.911 1 Source: own research -2 0 2 4 6 8 10 12 14 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ETR in % Year ETR0 ETR1 ETR2 0 5 10 15 20 25 30 35 40 45 50 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ITRL in % Year ITRL0 ITRL1 ITRL2 - 114 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH Table 2 presents the results of the correlation analysis. According to the results, a statistically significant positive correlation was indicated between ITRL2 and ITRL1, between ETR1 and ETR2, between ITRL2 and ETR2 or between ITRL1 and ETR2. In general, the dependence between ITRL and ETR is positive. If the tax burden increases for one group of taxpayers, the tax burden usually increases for the others. However, some moderately positive correlations, such as between ETR0 and ETR2, suggest that this tax burden may not increase equally. In this case, the reason is the already mentioned increase in the amount of tax credits for children. A negative correlation was found only between the employment rate indicator and the ITRL or ETR indicators. A more detailed evaluation of the effective tax rate, resp. the implicit tax rates on tax revenue (Model A) or the employment rate (Model B) are provided in the regression analysis in Table 3. 4.3 Regression analysis The results of Table 3 show that both models have a high determination index R2. In the case of model A and B this means that more than 95% of the variance is due to the character XN and less than 5% due to random deviations. There are also no autocorrelation according to the results of the Durbin-Watson test (more about the Durbin-Watson test Ali, 1987; Turner, 2019). According to the outcomes of the F-test, both regression models are statistically significant at 5% level of significance. This level of significance is recommended according to Yalcinkaya et al., (2017). Tab. 3: Regression Analysis Model A Model B Variable X 1 , X 2 , X 3 , X 4 , X 5 , X X 6 1, X2, X3 Signif. Coefficient Signif. Coefficient X1 - ITR 0.017 L0 -4.713 0.689 -0.876 X2 - ITR 0.096 L1 2.865 0.008 7.396 X3 - ITR X L2 X 0.000 -5.016 X4 - ETR 0.006 0 3.841 - - X5 - ETR 0.076 1 -2.125 - - X6 - ETR 0.035 2 0.649 - - Constant 0.000 78.582 0.009 -5.379 Observation 18 18 R0.957 2 0.952 Signif. F 0.000 0.000 DW 2.304 2.144 In the case of model A, which is generally formalized by (1), the regression coefficient X2 and X5 are statistically insignificant. The same situation is with regression coefficient X3 , which is excluded from the model. After removing insignificance variables, model A has a form (6), Y = -4.713 X1 + 3.481 X4 + 0.649 X6 + 78.582 (6) The negative parameter X1 of the model in equation (6) shows that the employment rate would increase if the labour tax burden decreased. On the contrary, the positive parameters for X4 and X6 , which express ETR, mean that even with the growth of the tax burden, an increase in the employment rate can be expected. The extent to which the tax burden on labour affects the share of personal income tax from dependent activity to the total tax revenue is modelled in Model B in Table 3. For Model B, the explanatory variable X1 is insignificant. The final form of the model is shown by (7). Y = 7.396 X2 – 5.016 X3 – 5.379 (7) This equation illustrates that with a higher tax burden represented by ITRL2 it is possible that the personal income tax from dependent activities also has a higher share of tax revenue. However, as the tax burden on taxpayers decreases with applying the deduction for children, parameter X3 is negative. Using the angle between the regression linear lines, generally determined (5), the dependence between the development of ETR and ITRL from 2003 to 2020 is examined in cases where the employee applies a deduction per taxpayer (Figure 2a), per taxpayer and one child (Figure 2b) and taxpayer and two children (Figure 2c). As in the previous analysis, this study's uniqueness and added value are that the ETR and ITRL are weighted averages calculated for the wage distribution. The value of the cotg φ angle enclosing the linear regression lines is shown in Table 4. Figure 2a. Situation S 0 Source: own research Figure 2b. Situation S 1 Source: own research Figure 2c. Situation S 2 Source: own research 0 10 20 30 40 50 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ETR, ITRL in % Year ETR0 ITRL0 0 10 20 30 40 50 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ETR, ITRL in % Year ETR1 ITRL1 -5 0 5 10 15 20 25 30 35 40 45 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 ETR, ITRL in % Year ETR2 ITRL2 - 115 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH Tab. 4: Angle of Linear Regression Lines Situation Cotg S cotg φ0= 0.712 1−0.507 ∙�0.87 1.05 +1.05 0.87� = 18° 79´ 0 S cotg φ1= 0.945 1−0.893 ∙�1.60 1.75 +1.75 1.60� = 3° 22´ 1 S cotg φ2= 0.988 1−0.976 ∙�2.70 2.66 +2.66 2.70� = 0° 69´ 2 5 Discussion Figure 1 shows that the tax burden decreases with increasing deductions. Compared to the first analysed year (2003), the ETR decreased in cases where the deduction for children is applied due to the frequent valorisation of the values of this tax credit. On the contrary, the taxpayer's relief has not been increased, and so with the gradual increase in wages in the national economy, the tax burden is also growing. The trend of a gradual increase in the tax burden on labour since 2009 is also confirmed by the ITRL . The results of Michaelis and Birk (2006) confirm that compulsory SSC influences the most significant tax burden on labour. This trend continues although the rate of SSC paid by the employee has decreased from 8% of the assessment base since 2009 to 6.5%, the rate of SSC paid by the employer from 26% to 25% since 2009, resp. to 24.8% from 1 July 2019. Based on the results, it can be concluded that labour taxation impacts the employment rate. However, when analysing the effects and possible changes in the reduction or increase of the tax burden, it is also necessary to consider the declining stability of the legal environment, which affects both supply and demand for labour (Kotlán et al., 2019). Compared to labour markets in North America, these changes are less flexible (Nickell, 1997). The results also partially confirm the main findings of the study Kosi and Bojnec (2006), Goudswaard and Caminada (2015) or Tepperová (2019), which states that labour taxation has a negative impact on employment as a high tax burden increases labour costs. Results Figure 2 confirms Mahdavi (2008) that tax reform can significantly change the tax burden and thus the share on tax revenues. In all analysed cases, the tax burden decreased between 2007 and 2008. The results of model (6) confirm the conclusions of Hájek (2003). One of the ways to achieve a lower tax burden on labour is to transfer the tax burden from direct taxes to indirect taxes, which was partially done during the reform of public finances in the Czech Republic between 2007 and 2008. More about the analysis of selected indirect taxes e.g. Krzikallová and Střílková, (2016); David (2019). Negative parameter X3 in relation (7) follows that if the state goal is to increase the share of personal income tax from the dependent activity to the tax revenue, this increase can be achieved mainly through taxpayers who apply a deduction for one child or do not apply the deduction at all. Otherwise, it is more advantageous to recommend a change in the structural elements of the tax - for example, adjustments to the amounts of deductions for children or changes in the method of tax base construction. These changes subsequently affect tax progressivity, as Koskela and Schob (2009) stated. Another problem with changes in personal income taxation is long-run elasticities for wage tax (Havránek et al., 2016). In addition to personal income tax, ITRL also includes SSC paid by the employer and the employee. In the Czech Republic, the SSC paid by the employer enters the tax base and, according to Prammer (2019), increases personnel costs. This factor influences the share of personal income tax to the total tax revenue. Despite the nominal reduction in selected SSC rates, as shown by the ITRL in Figure 1, SSC represents the highest component of the tax and levy burden of wages. The closer the value of cotg φ is to zero, the higher dependence between the pairs of analysed values. In situation S0, where only the deduction for a taxpayer is applied, the degree of dependence between ETR and ITRL is the lowest. ITRL ranges from 40 to 45%. If the high levy burden increased, the difference between the ETR and ITRL values would decrease, which is not the case here. For this reason, the results of the Bronchi and Burns (2001) study on the high disparity between the tax and levy burden of labour are also confirmed. Therefore, a high levy burden on labour remains a problem. A slight decrease in the tax burden occurred only during 2005 and 2006 when the non-taxable part of the tax base was replaced by relief, and between 2007 and 2008, when the nominal progressive tax rate changed to a linear rate. The levy burden on wages decreased only between 2008 and 2009 with a decrease in the rate of social security premiums and contributions to the state employment policy paid by both the employee and the employer. In this case, there are partial identical conclusions of Adam et al. (2019) that reducing SSC has positive effects on labour costs. The rate of SSC was also changed during the year 2019, but due to positive economic growth and rising wages in the economy, this did not reduce the price of labour. This high levy burden can then lead employers to shift the burden of compliance with SSC back to employees in lower wages (Nielsen and Smyth, 2008). A higher dependence between ETR and ITRL exists when, in addition to the deduction per taxpayer, the deduction for children is also applied. This leads to a decrease in ETR values, leading to a reduction in ITRL. Even in this case, the SSC is no longer graded according to, for example, the number of children supported by the taxpayer, which could be a criterion for determining the amount or rate of SSC (Alvarez-Martinez and Polo, 2014). The main factor reducing the tax burden on labour, especially in the middle of the analysed period, was the increase in tax credits for children, resp. replacement of non-taxable parts of the tax base by a tax advantage. As in the situation of S0 , the average labour costs and the average effective tax rate have increased in recent years. Employers bear wage costs and related SSC in their costs (Tachibanaki and Yokoyama, 2008). 6 Conclusion The article aimed to evaluate whether ETR resp. ITRL affects employment and whether ITRL affects the share of personal income tax from the dependent activity to the total tax revenue. Another goal was to evaluate the development of ETR and ITRL rates. The analysis was performed under the conditions of the valid legislation of the Czech Republic in the period 2003–2020. The period for analysis is limited by the availability of data on the distribution of wages based on which the weighted average ETR and ITRL were calculated. The analysis of dependencies with weighted averages of these values is one of the added values of this study. The ETR and ITRL rates were calculated for the three most frequent situations that arise in the Czech Republic for the taxation of income from dependent activities, i.e. the taxpayer applies the deduction only to himself, to the taxpayer and one child or the taxpayer and two children. The analysis was abstracted from the use of non-taxable parts of tax bases, which may be a limiting factor in this study. The main results of this study include the finding that the tax and levy burden on wages has a negative impact on employment. If the connection of employment only with the effective tax rate were analysed, the partial opposite effect was found. The analysis examining the dependence between the share of personal income tax from the dependent activity to the tax revenue to ITRL shows that to evaluate the impact of the share of tax revenue, it is necessary to analyse whether and how many deductions are used the taxpayer to reduce tax liability. Especially in tax levies in connection with personal income tax, the tax burden is in some cases zero, resp. negative. This is confirmed by the analysis results examining whether the levy and tax burden on labour is decreasing. This highest tax burden decrease occurred in connection with the reform of public finances in 2008. However, the share of SSC to the total tax and levy burden changes only minimally and thus, the high levy burden of income from dependent activity remains a problem. - 116 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH This study has some limitations. First of all, is the fact that research was done under the condition of the Czech Republic. The analysis results can´t be generalized to all European or world countries. The second limitation is data availability, especially wage distribution from the year 2003. This is the reason why analysis started this year. Finally, the third limitation is that tax burden calculation is abstracted from non-taxable parts. The topic for further research in this area may be comparing these results with a similar analysis in another country, extending the period's length according to the availability of data, or considering changes in income taxation from dependent activity in future years. Literature: 1. Adam, S., Philips, D., & Roantree, B. (2019). 35 Years of Reforms: a Panel Analysis of the Incidence of and Employee and Employer Responses to Social Security Contributions in the UK. Journal of Public Economics, 171, 29–50. doi: 10.1016/j.jp ubeco.2018.05.010. 2. Ali, M. M. (1987). Durbin-Watson and Generalized DurbinWatson tests for Autocorrelations and Randomness. Journal of Business, Economic Statistic, 5(2), 195–203. 3. Alvarez-Martinez, M. T., & Polo, C. (2014). Reduce Employers' Social Security Contributions and Control Labor Fraud: Remedies for Spain's Ailing Economy? Economic System Research, 26(2), 141–154. doi: 10.1080/09535314.2014.897218. 4. Bauer, T., & Riphahn, R. T. (2002). Employment Effects of Payroll Taxes - an Empirical Test for Germany. Applied Economics, 34(7), 865–876. doi: 10.1080/00036840110058914. 5. Belas, J., Gauvurov, B., Cepel, M., & Kubak, M. (2020). Evaluation of Economic Potential of Business Environment Development by Comparing Sector Differences: Perspective of SMEs in the Czech Republic and Slovakia. Oeconomia Copernicana, 11(1), 135–159. doi: 10.24136/oc.2020.006. 6. Bronchi, C., & Burns, A. (2001). The Tax System in the Czech Republic. Finance a úvěr, 51(12), 618–638. 7. Burda, M., & Weder, M. (2016). Payroll Taxes, Social Insurance, and Business Cycles. Journal of the European Economic Association, 14(2), 438–467. http://dx.doi.org/1 0.1111/jeea.12145 8. Cutler, H., Shields, M., & Davies, S. (2018). Can State Tax Policy Increase Economic Activity and Reduce Inequality? Growth and Change, 49(1), 142–164. http://dx.doi.org/1 0.1111/grow.12216 9. Czech Statistical Office (2022a). Czech Republic in Numbers. Retrieved from http://www.czso.cz/csu/czso/ceska-republika-odroku-1989-v-cislech-aktualizovano-1552020 10. Czech Statistical Office (2022b). Struktura mezd zaměstnanců. Retrieved from http://www.czso.cz/csu/czso/struk tura-mezd-zamestnancu-2019 11. Dalenberg, D. R., & Partridge, M. O. (1995). The Effects of Taxes, Expenditures, and Public Infrastructure on MetropolitanArea Employment. Journal of Regional Science, 35(4), 617–640. http://dx.doi.org/10.1111/j.1467-9787.1995.tb01296.x 12. David, P. (2019). Basic Economic Gap Related to Smoking: Reconciling Tobacco Tax Receipts and Economic Costs of Smoking-attributable Diseases. Tobacco Control, 28(5), 558– 561. doi: 10.1136/tobaccocontrol-2018-054307. 13. Financial Administration. (2022a). Údaje z daňových přiznání. Retrieved from http://www.financnisprava.cz/cs/dan e/analyzy-a-statistiky/udaje-z-danovych-priznani 14. Financial Administration (2022b). Daňová statistika. Retrieved from http://www.financnisprava.cz/cs/dane/analyzy-astatistiky/danova-statistika 15. Financial Administration (2022c). Údaje z výběru daní. Retrieved from http://www.financnisprava.cz/cs/dane/analyzy-astatistiky/udaje-z-vyberu-dani 16. Garrett, T. (2009). Evaluating State Tax Revenue Variability: a Portfolio Approach. Applied Economics Letters, 16(3), 243– 246. http://dx.doi.org/10.1080/13504850601018403 17. Glday, A., & Madl, L. (2018). Tax Burden on Labour Income in the Visegrad Region. Public Finance QuarterlyHungary, 63(3), 376–401. 18. Goudswaard, K., & Caminada, K. (2015). Social Security Contributions: Economic and Public Finance Considerations. International Social Security Review, 68(4), 25–45. http://dx.d oi.org/10.1111/issr.12086 19. Grace, K. (2018). The Impact of Personal Income Tax Rates on the Employment Decisions of Small Businesses. Journal of Entrepreneurship and Public Policy, 7(1), 74–104. http://dx.doi.org/10.1108/JEPP-D-17-00030 20. Hájek, L. (2003). Tax Burden in OECD Countries and in the Czech Republic. Politická ekonomie, 51(5), 714–725. 21. Havránek, T., Irsová, Z., & Schwarz, J. (2016). Dynamic Elasticities of Tax Revenue: Evidence from the Czech Republic. Applied Economics, 48(60), 5866–5881. http://dx.doi.org/1 0.1080/00036846.2016.1186796 22. Chernick, H., & Reimers, C. (2019). Consumption Taxes, Income Taxes, and Revenue Sensitivity: States and the Great Recession. Public Finance Review, 47(2), 349–381. http://dx.doi.org/10.1177/1091142117741669 23. Jordaan, Y., & Schoeman, N. (2015). Measuring the Impact of Marginal Tax Rate Reform on the Revenue Base of South Africa Using a Microsimulation Tax Model. South African Journal of Economic and Management Sciences, 18(3), 380– 394. http://dx.doi.org/10.4102/sajems.v18i3.795 24. Knápková, A., Homolka, D., & Pavelková, D. (2014). Utilization of Balanced Scorecard and the Effect of Its Use on the Financial Performance of Companies in the Czech Republic. E+M Ekonomika a management, 17(2), 146–160. 25. Konečná V., & Andrejovská, A. (2020). Investor Decisionmaking in the Context of the Effective Corporate Taxation. E+M Ekonomika a management, 23(1), 111–120. doi: 10.15240/tu l/001/2020-1-008 26. Kosi, T., & Bojnec, S. (2006). The Impact of Labor Taxation on Job Creation and Unemployment. Ekonomický časopis, 54 (7), 652–667. 27. Koskela, I., & Schob, R. (2009). Is Tax Progression Good for Employment? Efficiency Wages and the Role of the Prereform Tax Structure. Finanzarchiv, 65(1), 51–72. http://dx.doi.org/10.1628/001522109X444206 28. Kramer, D., Cok, M., Cirman, A., & Verbic, M. (2016). Switching Personal Income Tax and Social Security Contributions between Slovenia and the Federation of Bosnia and Herzegovina. Economic Research-Ekonomska Istrazivanja, 29(1), 682–695. doi:10.1080/1331677X.2016.1193948. 29. Kotlán, I., Němec, D., & Machová, Z. (2019). Legal Uncertainty in Taxation and Its Impacts on Labour Supply in the Czech Republic. Politická ekonomie, 67(4), 371–384. http://dx.doi.org/10.18267/j.polek.1246 30. Krzikallová, K., & Střílková, R. (2016). Labour-Intensive Services and Changes in Value Added Tax Revenue. Journal of Competitiveness, 8(1), 5–18. http://dx.doi.org/10.7441/joc.2 016.01.01 31. Lin, B., & Jia, Z. (2019). Tax rate, Government Revenue and Economic Performance: A Perspective of Laffer Curve. China Economic Review, 56. http://dx.doi.org/10.1016/j.chieco.20 19.101307 32. Lyková, L. N. (2015). Tax Burden in the Russian Federation: A Comparative Analysis Point of View. Journal of Tax Reform, 1(1), 76–89. http://dx.doi.org/10.15826/jtr.2015.1.1.005 33. Mahdavi, S. (2008). The Level and Composition of Tax Revenue in Developing Countries: Evidence from Unbalanced Panel Data. International Review of Economics & Finance, 17(4), 607–617. http://dx.doi.org/10.1016/j.iref.2008.01.001 34. Mankiw, N. G., Weinzlerl, M., & Yagan, D. (2009). Optimal Taxation in Theory and Practice. Journal of Economic Perspectives, 23(4), 147–174. http://dx.doi.org/10.1257/jep.2 3.4.147 35. Mark, S. T., McGuire, T. J., & Papke, L. E. (2000). The Influence of Taxes on Employment and Population Growth: Evidence from the Washington, DC Metropolitan Area. National Tax Journal, 53(1), 105–123. http://dx.doi.org/10.17310/ntj.20 00.1.06 36. Marill, K. (2004). Advanced Statistics: Linear Regression, Part I: Simple Linear Regression. Academic Emergency Medicine, 11(1), 87–93. http://dx.doi.org/10.1197/S1069-6563(0 3)00600-6 - 117 - AD ALTA JOURNAL OF INTERDISCIPLINARY RESEARCH 37. Michaelis, J., & Birk, A. (2006). Employmentand Growth Effects of Tax Reforms. Economic Modelling, 23(6), 909–925. http://dx.doi.org/10.1016/j.econmod.2006.04.005 38. Moravec, L., Rohan, J., & Hinke, J. (2019). Estimation of International Tax Planning Impact on Corporate Tax Gap in the Czech Republic. E+M Ekonomika a management, 22(1), 157– 171. doi: 10.15240/tul/001/2019-1-011 39. Nickell, S. (1997). Unemployment and Labor Market Rigidities: Europe versus North America. Journal of Economic Perspectives, 11(3), 55–74. http://dx.doi.org/ 10.1257/jep.11.3 .55 40. Nielsen, I., & Smyth, R. (2008). Who Bears the Burden of Employer Compliance with Social Security Contributions? Evidence from Chinese Firm Level Data. China Economic Review, 19(2), 230–244. http://dx.doi.org/10.1016/j.chieco.2 007.06.002 41. Prammer, D. (2019). How does Population Ageing Impact on Personal Income Taxes and Social Security Contributions? Journal of the Economics of Ageing, 14. http://dx.doi.org/1 0.1016/j.jeoa.2018.12.005 42. Reiss, L., & Schuster, P. (2020). Explaining the Evolution of the Austrian Implicit Tax Rate on Labor from 1976 to 2016. Empirica, 47(2), 303–341. http://dx.doi.org/10.1007/s10663-01 8-9419-5 43. Rick, S., Paolacci, G., & Burson, K. (2018). Income Tax and Motivation to Work. Journal of Behavioral Decision Making, 31(5), 619–631. http://dx.doi.org/ 10.1002/bdm.2078 44. Sanz-Sanz, J. F. (2016). Revenue-maximizing Tax Rates in Personal Income Taxation in the Presence of Consumption Taxes: a Note. Applied Economics Letters, 23(8), 571–575. http://dx.doi.org/10.1080/13504851.2015.1088136 45. Široký, J., Krajčová, J., & Hakalová, J. (2015). The Taxation of Agricultural Land with the Use of Multi-criteria Analysis. Agricultural Economics-Zemedelska Ekonomika, 62(5), 197– 204. http://dx.doi.org/10.17221/183/2015-AGRICECON 46. Tachibanaki, T., & Yokoyama, Y. (2008). The Estimation of the Incidence of Employer Contributions to Social Security in Japan. Japanese Economic Review, 59(1), 75–83. http://dx.do i.org/10.1111/j.1468-5876.2007.00380.x 47. Teplická, K., & Daubner, M. (2013). Study of the Relation between Tax Competition and Tax Burden in European Union Member States. Ekonomický časopis, 61(2), 172–186. 48. Tepperová, J. (2019). Personal Income Tax and Social Security Coordination in Cross-Border Employment - a Case Study of the Czech Republic and Denmark. European Journal of Social Security, 21(1), 23–41. http://dx.doi.org/10.1177/1388 262719833766 49. Toka, O., Cetin, M., & Arslan, O. (2019). Robust Estimation in Restricted Linear Regression. Communications in StatisticsSimulation and Computation. http://dx.doi.org/10.1080/036109 18.2019.1661475 50. Turner, P. (2019). Critical Values for the Durbin-Watson test in Large Samples. Applied Economic Letters, http://dx.doi.or g/10.1080/13504851.2019.1691711 51. Yalcinkaya, A., Cankaya, M. N., Attindag, O., & Tuac, Y. (2017). Investigation for the Robustness of Significance Level When the Normality Assumption in Hypothesis Tests is Violated. Gazi Univerzity Journal of Science, 30(2), 247–259. 52. Yilmazkuday, H. (2017). Individual Tax Rates and Regional Tax Revenues: a Cross-state Analysis. Regional Studies, 51(5), 701–711. http://dx.doi.org/10.1080/00343404.2015.1119266 Primary Paper Section: A Secondary Paper Section: AD, AH - 118 -