Climate Change Concerns and Financial Market Returns: Distributional Predictability and Local State Dependence Across Technology, Clean Energy and Carbon‐Intensive Assets

ABSTRACT

This study examines how climate change concerns are associated with technology‐oriented, clean‐energy and carbon‐intensive energy returns using daily data from 23 April 2018 to 30 June 2025. The empirical framework combines quantile Granger non‐causality testing, quantile‐on‐quantile regression (QQR), pointwise moving‐block‐bootstrap inference and an aggregation‐based QR–QQR benchmark. The Granger results reveal limited Overall predictability, with only the CCS‐to‐SOG direction significant in the omnibus test, whereas technology‐oriented and clean‐energy assets display localized predictive content at selected quantiles, mainly in upper‐return states. QQR estimates show substantial local heterogeneity, but bootstrap confidence bounds indicate that statistically supported associations are concentrated in a limited number of asset‐specific quantile combinations, with ICLN showing the strongest localized support. The QR–QQR benchmark further reveals particularly close broad‐profile correspondence for carbon‐intensive energy assets. Overall, the financial relevance of climate change concerns is better characterized as selective, localized and asset dependent than as pervasive across markets or distributional states.