Climate risk and carbon emissions: Examining their impact on key energy markets through asymmetric spillovers

Abstract

The global energy sector faces crucial challenges in the wake of climate change and the escalating emissions of greenhouse gases (GHGs). Recent months have witnessed substantial fluctuations in energy prices, attributed to various factors, including surging demand from emerging economies, geopolitical tensions in regions like Russia and Iran, and progressive efforts to phase out fossil fuels. The alarming pace of climate change is leading to a rise in global temperatures, intensifying severe weather phenomena like droughts and hurricanes. In this context, this empirical research article aims to explore the asymmetric spillovers of climate risk and carbon emissions uncertainty on key energy markets, namely natural gas, electricity, coal, oil, and diesel. Employing the time- and frequency-domain methodologies introduced by Diebold and Yilmaz (2014) and Baruník and Kˇrehlík (2018) (BK), we conduct an in-depth analysis of the monthly and daily frequency of climate policy uncertainty (CPU) and carbon emissions (CO), respectively. These measurements allow us to assess the time-varying spillover effects, directional spillovers, net directional spillovers, and connectedness network across short-term (1–3 months), medium-term (3–6 months), and long-term horizons (6 months-inf). Our research outcomes reveal that, based on the BK framework, the overall connectedness for CO and energy prices is most pronounced at the 3–6 months horizon, with a magnitude of 39.01%. For the CPU-based model, the overall connectedness exhibits an increasing trend over time, ranging from 33.84% for 1 month to 47.91% for 6 months and beyond. Furthermore, our investigation indicates that CPU predominantly serves as a net transmitter to energy prices for 1–3 months, whereas the net transmission of CO varies across different time frames. Specifically, CO acts as a net transmission for natural gas (NG) for 1 month and 1–3 months, while for 3–6 months, CO emerges as a net transmitter for WTI crude (WTI) and Brent Oil (BRENT). In addition, the results of the frequency-based Granger causality analysis demonstrate that CPU Granger causes NG, WTI, and BRENT for all frequencies, while CO Granger causes NG, WTI, BRENT, and ULS diesel (ULSD) across all frequencies. The implications of our findings are particularly significant for policymakers and energy importers, urging them to differentiate their short- and long-term strategies and procurement contracts. In the long run, policymakers should be cognizant of the influence of climate change and emissions on energy demand, while formulating prudent policies and sustainability initiatives.