Analyzing structural change and causality between energy consumption and Bitcoin’s activity

Abstract

This study purposes to examine a Granger causality, in the presence of the structural change, between Bitcoin’s activity and fossil and renewable energy consumptions. To achieve this goal, the two-regime Markov Switching Granger Causality test-based Vector Autoregressive is employed. The empirical results show that there is a significant structural change in the causality between both energy consumption and Bitcoin’s activity as the lower Akaike, and Bayesian information criterion are revealed. Regarding the causality results, our findings reveal a significant bidirectional causality between renewable energy consumption and Bitcoin’s activity in both high and low energy consumption regimes. However, in terms of fossil consumption and Bitcoin’s activity nexus, it is solely strong evidence of bidirectional causality between them in a low consumption regime. Our results indicate a strong predictive power of Bitcoin’s activity in energy consumptions in both peak and non-peak periods, and it is also highlighted that Bitcoin’s activity could not be neglected in preparing the energy policy.