This study examines the impact of climate policy uncertainty (CPU) on green bond returns (GB) under different time horizons and market conditions. For this purpose, monthly data from the news-based CPU index developed by Gavriilidis (2021) and the S&P Green Bond Index for the period from June 2015 to June 2025 are employed. The analysis is conducted using wavelet-based quantile-on-quantile regression (WQQR) and wavelet-based quantile-on-quantile Granger causality (WQQGC), which capture non-linear, asymmetric, and distribution-dependent dynamics. The findings reveal that the impact of CPU on green bond returns varies across time horizons and quantile levels. In the short term, increases in CPU are found to suppress green bond returns in low-return regimes, while exerting a positive effect in high-return regimes. In the medium term, this relationship reverses, and the CPU effect gradually evolves from negative to positive. In the long term, CPU is found to exert a statistically significant and positive effect on green bond returns across all quantile levels. Furthermore, the causality analysis indicates that the effects of CPU on the green bond market exhibit a heterogeneous, market-condition-dependent structure across both time-frequency and quantile dimensions.
Aslan Aydoğdu (Wed,) studied this question.