Purpose The purpose of this paper is to focus on global energy consumption using the economic growth nexus, the prevalent energy hypothesis at a global level and the impact of the main historical events assessed for the period from 1965 to 2015. Design/methodology/approach Given the confirmed presence of endogeneity and cointegration between energy consumption and economic growth, a vector error correction with structural dummies model was used. Furthermore, the impulse-response functions and variance decomposition were computed to evaluate the variables’ dynamics. Findings Bi-directional causality running from energy consumption to economic growth was found, both in the short and long-run, supporting the feedback hypothesis. It is proved that the 2008 crisis impacted on the global energy–growth nexus. Furthermore, there is evidence of the impact of the 1990s oil price shock on the nexus. Innovations in energy consumption have a positive impact on economic growth; however, this impact tends to be null in the long run. Practical implications The results suggest that at a global level, any energy policy should be carefully designed in order not to hamper economic growth. Countries should not remain indifferent to the policies that other countries might follow. Very few historical crises impacted on the global energy–growth nexus. Originality/value This paper offers a different approach to the study of the energy–growth nexus. The energy–growth nexus is analysed in the major macroeconomic aggregate. Global variables reveal their relevance as a benchmark in the energy–growth nexus. Furthermore, this paper arrives at some conclusions about how historical crises impact on global relationships.