Renewable energy effect on economy and environment: The case of G7 countries through novel bootstrap rolling window approach

One mystifying outcome in the literature on renewable-economy-environmental causality is the inconsistency of outcomes specifically across various sample sizes, model specifications, and periods. Considering such difficulties, the study applies Bootstrap Rolling Window Granger Causality (BRWGC) test with fixed-size rolling subsamples to evaluate connections. The data used incorporates REC, EG, and CO2 emissions for G-7 nations between 1970 and 2021. Using the full-sample, it is observed that there is predictive power from REC to EG only in the USA and UK while there is predictive power from REC to CO2 emissions in the USA and Italy. However, fullsample outcomes are unreliable because models do have not parameter constancy based on parameter instability tests. Similar to full-sample results, BRWGC estimation results do not present evidence of a consistent relationship from REC to EG and CO2 emissions. However, it is discovered that causal interrelationships exist between the series in a number of the sub-samples. Additional evidence that the results are not statistical artifacts but rather reflect actual economic shifts comes from the fact that these sub-sample times coincide with important economic events. The findings of this study complement earlier research and provide a rationale for divergent outcomes.

Automated summary

The literature on renewable-economy-environmental causality shows inconsistent outcomes across sample sizes, model specifications, and periods. This study addresses this issue by using Bootstrap Rolling Window Granger Causality (BRWGC) test and finds causal interrelationships in sub-samples. The findings complement earlier research and provide a rationale for divergent outcomes.