Testing the validity of the BARS curve for Turkey

Abstract

This study has two main aims: to test the validity of the BARS curve in Turkey for the period 1974-2016 and to estimate the optimal government size for that period and compare it with the current situation. It uses the Autoregressive Distributed Lag bounds test and quadratic equation methods. The empirical findings of the study confirm the validity of the BARS curve by providing strong evidence for the existence of an inverted U-shaped long-run relationship between government size and economic growth. Unlike many previous studies that use a single proxy measure for government size, this study uses all available fundamental indicators and their sub-components. The empirical results show that for the period studied all proxy measures of government size exceed the optimal except for total central government budget expenditure and defence expenditure. Therefore, decreasing the size of the government, other than for these two indicators, will increase economic growth in the long run.