Long-run relation and short-run dynamics in energy consumption–output relationship: International evidence from country panels with different growth rates

The long-run relation and short-run dynamics between output and consumption of energy (electricity) are examined in panels of countries with different growth rates. Seventy nine countries over 1971–2011 are grouped into high-, low- and negative-growth categories based on exponential growth rate of their per capita output. Tests of cointegration suggest the existence of long-run relation between energy (electricity) consumption and output in high- and low-growth panels but its absence in the panel with negative growth. Accounting for cross-country dependency strengthens the findings. Estimates of long-run elasticity of output with respect to energy (electricity) are significant in panels with positive growth rates. The common correlated effect mean-group estimators of the error-correction model suggest (1) long-run bidirectional causality between output and energy (electricity) in all three groups of countries, (2) short-run bidirectional causality in output-energy relation for the full sample as well as in the low-growth category; and (3) unidirectional causality from output to energy in the negative-growth category. The finding of long-run bidirectional causality is robust to inclusion of carbon emission, urbanization, exports, and foreign direct investment as control variables.