Determining the Future Direction and Amount of Tax Revenue in Indonesia Using an Error Correction Model (ECM)

Abstract

The characteristics of how tax revenues, as well as the macroeconomic factors that are expected to influence and determine tax revenues, develop have become the basic macroeconomic assumptions by the government in compiling various components of the posture of the state revenue and expenditure budget to maintain and increase the expected economic growth. To understand the evolution of tax revenues and the impacts of these variables, it is important to investigate the relationship between tax revenues and specific macroeconomic variables. This study aims to identify the effect of macroeconomic variables on Indonesian tax revenues in the long and short term using multiple regression analysis in an error correction model (ECM). The macroeconomic conditions and variables used in this study are gross domestic product (GDP), inflation, interest rates, and exchange rates. This study employed the time-series data from 2000–2019. The ECM method was conducted in the following stages: stationarity test, cointegration test, and ECM regression test. The ECM model is declared valid, if the cointegrated variables are supported by a significant and negative ECT coefficient value. The statistical analysis, i.e. t-statistic, the F-statistic, and the coefficient of determination, were used to evaluate the significance of the ECM model. The results showed that among the four of macroeconomic variables used in this study, GDP has the highest significant effect on the tax revenue. GDP has a highly positive and considerable impact on tax revenue. The increasing of GDP growth is in line with the realization of tax revenue. A strong positive long-term (t-statistic = 13.94075^*, P-value = 0.0000) and short-term (t-statistic = 5.515026^*, P-value = 0.0001) association is also observed between inflation and tax revenue. Inflation has a favorable and considerable impact on tax income either in long (t-statistic = 2.298586^**, P-value= 0.0363) or short term (t-statistic = 2.515695^**, P-value = 0.0258). On the other hand, Bank Indonesia interest rate has a negative insignificant effect (t-statistic = −1.542970ᵈ, P-value = 0.1437) in the long run on tax revenue. However, it has a negative significant effect (t-statistic = −2.699231^**, P-value = 0.0182) in short-term tax revenue. A rise in interest rates results in the lowered tax revenues due to the reduction of public consumption patterns, and a decrease in public consumption. The exchange rate has a negative negligible connection in long term (t-statistic = −1.045768ᵈ, P-value = 0.3122), as well as in short term (t-statistic = 1.250076ᵈ, P-value = 0.2333) with tax revenue, meaning that tax revenue is not significantly affected by changes in exchange rate. In conclusion, total tax revenues and the four macroeconomic variables have a significant long- and short-term association according to the ECM analysis with an adjusted R² value of 0.985866 and 0.792880, respectively. Changes in future tax revenues are largely determined by the stability of macroeconomic variables. Therefore, it is highly recommended that the government maintain the stability of macroeconomic variables, especially GDP, because GDP has a dominant influence on tax revenues in the long and short term so that the target of tax revenues can be achieved. The ECM analysis applied in this study can explain the effect of changes in macroeconomic variables on the tax revenue in the short and long term; so that the future direction and amount of tax revenue can be determined in preparing the State Budget and projecting the expected level of economic activity growth in a nation.