Is the causal nexus of energy utilization and economic growth asymmetric in the US?

► Asymmetric causality tests are implemented to investigate the causal interaction between energy and output in the US. ► This paper is the first attempt to separate the effect of positive shocks from negative ones. ► Bootstrap simulations are used to generate new critical values. ► The results show that positive and negative shocks have different causal impacts. ► The underlying parameters are estimated and the implication of the empirical results is discussed. This paper re-examines the causal nexus of energy utilization and GDP per capita in the US. The novelty of the paper is to allow for asymmetry in causality by using a new test introduced by Hatemi-J (forthcoming). A bootstrap procedure is used with leveraged corrections that perform more accurately when the statistical assumptions for validity of asymptotic distributions are not fulfilled. This is especially the case for sample sizes as in the current paper. The estimation results reveal strongly that a negative energy consumption shock will cause a negative shock in the output per capita. That is, if the energy utilization per capita decreases then the output per capita will also decrease. Surprisingly, such a causal impact for positive shocks is not found. These empirical results might indicate that there is an optimal quantity of energy in the US that needs to be consumed as otherwise the economic growth will suffer. However, the consumption of energy beyond that optimal quantity will not necessarily result in an enhanced rate of economic growth.