Estimating the Effect of Smoking and Alcohol Consumption on Health Expenditures by Spectral Causality Approach: Evidence From the Turkish Case

Abstract

The main purpose of this study is to evaluate the effects of smoking and alcohol use on health expenditures in Turkey. For this purpose, the spectral Granger causality approach was used, and the data set was collected from the OECD stat website for the period 1990 and 2019. According to the first finding, there is a strong causality between smoking and alcohol use and health expenditure, specifically in the 0, 0.5 domains, and supports the existence of bidirectional causality. Secondly, while there was a unidirectional causality from alcohol to health, this causality was found to be stronger in the following years. The last finding surprisingly revealed that tobacco use was not the Granger cause of health expenditures, whereas reverse causality was only supported in the domains of 0 and 2.5.