| Popis: |
The main purpose of present study is to investigate the relationship between macroeconomic variables and tax evasion using nonlinear approaches. . First of all, it was used Markov Switching Vector Autoregression method, statistics and information from 1969 to 2019 to find Kuznets’s hypothesis and nonlinear relationship between macro-economic variables and tax evasion, annually. Then, after confirming the existence of a nonlinear relationship, the Threshold Vector Autoregression was also used for further investigation. According to the results of the markov-switching, the existence of the Kuznets's law between per capita GDP and tax evasion is confirmed. Therefore, in order to reduce tax evasion, it is necessary to reduce the share of oil in GDP and increase the share of productivity in production. For variables of direct burden tax, public education expenditure, Government expenditure omitting education expenditure, Gini coefficient and currency outside the bank, Kuznets law is rejected. However, the existence of a nonlinear relationship is confirmed. Also, tax evasion, GDP per capita, direct tax burden and currency outside the bank was used for Threshold Vector Autoregression model. Tax evasion was selected as threshold variable and The threshold value of the tax evasion was specified 21906/1 Milliard Rials.The results of the Threshold Vector Autoregression model show that the impulses of per capita GDP and currency outside the bank have the greatest effect on increasing tax evasion. The results of the Threshold Vector Autoregression model show that the impulses of per capita GDP and currency outside the bank have the greatest effect on increasing tax evasion. |
