Energy use, real output‐growth, FDI, energy‐intensity and CO2 emission; the case of Kazakhstan.

The study investigates asymmetries in energy use, real output growth, FDI, energy intensity and CO2 emission in Kazakhstan. The linear ARDL approach of Pesaran et al. (2001. Journal of Applied Econometrics, 16(3), 289–326) was adopted following Aslan (2014. Energy Sources, Part B: Economics, Planning, and Policy, 9, 25–31) to examine the causality between energy use and real output within a simple model. Subsequently, in line with Ndoricmpa (2017. Energy, 125, 543–551), the Shin et al. (2014. Festschrift in honour of Peter Schmidt. New York, NY: Springer Science & Business Media) non‐linear ARDL asymmetric specification was followed to analyse the dynamic asymmetries amongst the estimated variables—in an extended model. Annual time series data expressed in natural logarithm from 1992 to 2016 was utilized for the study. ARDL model established a long‐run relationship between energy use and real output in Kazakhstan. The modified Granger causality test confirms uni‐directional causality from real output to energy use—suggestive that energy conservation hypothesis is supported. In the estimated NARDL model, no empirical evidence exist yet‐in support of the 'pollution haven' hypothesis and energy‐related CO2 induced emission in Kazakhstan whereas asymmetry exists in energy intensity. [ABSTRACT FROM AUTHOR]