The present study examines the long-run relationship between energy consumption and greenhouse gas emission for different groups of countries comprising lower middle income, upper middle income, and heavily indebted countries, East Asia and Pacific, East Europe and Central Asia, Latin America and Caribbean, Middle East and North Africa, South Asia, Sub-Saharan Africa and for aggregate data of the world. The data has been analyzed by using various econometric techniques, specifically the Johnson cointegration, modified version of Granger causality and variance decomposition analysis from the period of 1975 to 2011. The results confirm that there is a long-run relationship between greenhouse gas emissions (i.e. agricultural methane emission, agricultural nitrous oxide emission and carbon dioxide emission) and energy consumption. The results of Granger causality indicate that energy consumption Granger causes greenhouse gas emission but not vice versa. The important finding is that energy consumption Granger causes GDP per unit energy use, which confirms the energy led growth hypothesis in the world. However, the vice versa relationship does not hold. The results imply that a policy to cut energy consumption tends to diminish greenhouse gas emission though affecting GDP of countries negatively.