Abstract

This study aims to investigate the tax buoyancy hypothesis in an asymmetrical framework. Increasing levels of gross domestic product (GDP) cause higher levels of tax revenue for economies. However, the magnitude of this impact varies from country to country, or year by year. Some countries transfer more from GDP to tax revenue, while others transfer less. From this perspective, in the present study, the tax buoyancy is estimated for the Turkish economy over the 2006Q1-2022Q3 period by considering the asymmetrical effects. The results obtained from the nonlinear autoregressive distributed lag model suggest that tax buoyancy is valid for the personal income tax while for tax on goods and services, and for corporate income tax there is no evidence favoring tax buoyancy. In other saying, the discretionary changes are effective only for personal income tax. Moreover, the positive and negative shocks affect tax revenues at different rates, and in some cases, even insignificant impacts were observed for the negative shocks. The findings may simply guide policymakers in two ways; first, personal income tax is the only tax type causing to tax buoyancy; second, the positive and negative GDP shocks affect tax buoyancy at different rates.