Shaping energy supply diversity through green innovation in the face of global uncertainty in G20 economies

In a constantly evolving environment and the quest for energy transition, green innovation can be seen as an essential lever for boosting energy mix diversification and therefore energy security. Despite its importance, few studies have focused on the nexus between green innovation and energy diversity, especially in a context of uncertainty. This research bridges this gap by investigating, alongside the direct repercussions of green innovation on energy supply diversity, its non-linear effects depending on the level of global uncertainty. In this regard, appropriate techniques were applied, including bi-wavelet coherence, partial wavelet coherence, kernel regularized least squares, and panel threshold regression methods, using data from emerging and developed G20 economies and covering the period 1997–2022. The findings reveal a significant role of green innovation in boosting energy mix diversity across several developed G20 economies, such as France, Germany, Japan, and South Korea, particularly at high frequencies. However, with the exception of China and India, emerging economies within the G20, these effects are insignificant in terms of energy portfolio diversification. Furthermore, the results of kernel-based machine learning and panel threshold regression reveal a positive influence of global uncertainty on energy supply diversity within the G20, but only when uncertainty exceeds a threshold of 0.030. The outcomes of this study provide significant insights for policymakers regarding the role of green innovation in building resilience in a situation of ongoing and multidimensional shocks, and in different schemes given that the approach adopted differentiates between emerging and developed countries.