Energy, cilt.197, 2020 (SCI-Expanded)
The investigation of the stationarity property of energy consumption has recently been a challenge for researchers in energy economics and important for consistent policy proposals. This study uses a set of conventional unit root tests and a recently developed Fourier panel KPSS test to investigate the persistence of shocks in disaggregated energy consumption data in Turkey from 1970 to 2016. The empirical results are as follows. First, the results of the unit root tests with no structural breaks reveal that energy consumption is overwhelmingly nonstationary in regard to sectoral energy consumption, except for the government's consumption; additionally, panel statistics also indicate that energy consumption is nonstationary. Second, the results of the stationarity test that allows multiple structural breaks reveal that sectoral energy use is stationary, but the panel results are controversial. Third, the Fourier panel KPSS test shows that energy use is nonstationary at both the sectoral and the panel level. Overall, the empirical findings indicate that disaggregated energy use is nonstationary, and that the effects of possible shocks on energy consumption in Turkey are permanent. Therefore, stabilization policies on energy consumption are required to avoid a possible spillover effect of energy shocks on other macroeconomic variables.