Tax Burden and GDP: Evidence from Frequency Doman Approach for the USA

We employed Breitung and Candelon's (2006) frequency domain approach to investigate the short-and long-run Granger-causality from different tax burden to GDP in the USA for the period 1947:1 –2009:3. The frequency domain analysis shows that current receipts, personal current tax, taxes on production and imports and taxes on corporate income do not Granger-cause GDP, both at the short and high frequency level; however, current tax receipts Granger-cause GDP in the frequency range of (0.9,1.9), corresponding to the cycle of to 3 months to 7 months. These results suggest that when the USA looks forward to rebalancing her GDP, by means of taxation, it is preferable to reconsider the tax structure with a focus on current tax receipts. This is so because by changing the structure of current tax receipts, the USA will be able to earn more revenue, even in the initial stage. However, if the USA decides to increase welfare, with the stability and sustainability of GDP, the policy makers are advised to readjust the tax burden by infusing the changes of the current receipts, personal current tax, taxes on production and imports and taxes on corporate income.