Tax Revenue Forecasting: an Empirical Evidence Using Autoregressive-Distributed Lag (ARDL)

Abstract

It is essential for the Directorate General of Taxes (DGT) to forecast tax revenue fast, accurately, and consistently, especially with high volatility in the current macroeconomics condition. This article tries to build an empirical forecasting model capturing global and local macroeconomic conditions as variables. The researcher found that most macroeconomic variables can be utilized to predict tax revenue in the short run. However, these variables are less likely to project tax revenue in the future for a particular type of tax revenue. Given these results, DGT, as a tax authority in Indonesia, may apply this approach to forecast tax revenue, both in the short run and the long run.