In re-examining the nexus between energy consumption and economic growth through the predictability framework, we adopt a panel data predictive regression model to examine the possibility of growth, conservative, feedback, or neutrality hypotheses for 135 countries. A predictive regression model is fitted to panels of countries on the basis of location and level of economic development. Findings suggest strong support for the neutrality hypothesis. A developing economy panel (90 countries) favours the conservative hypothesis, although a panel of 32 lower middle-income countries suggests that energy consumption per capita predicts real GDP per capita. These forecasts could provide future policy directions.