Abstract
This study examines whether physical climate risk is associated with stronger cross-market transmission across insurance, reinsurance, sovereign bonds, real estate, agricultural commodities, water utilities, and climate-adaptation equities under stressed market conditions. Using daily data from January 2020 to March 2026, the analysis applies Asymmetric Slope CAViaR models and a Quantile Vector Autoregression connectedness framework at the 5%, 50%, and 95% quantiles, with moving-block bootstrap inference and an independence-based placebo benchmark. Total connectedness remains low across market states, at 3.40% in the lower tail, 1.53% at the median, and 3.04% in the upper tail. The lower-tail increase relative to the median is not statistically significant, no market displays a statistically significant net transmitter or receiver position, and physical climate risk does not account for a significantly larger share of climate-vulnerable sovereign-bond variance under downside conditions. Rolling estimates also show no significant upward trend in lower-tail connectedness. Robustness analysis demonstrates that covariance specification can materially affect tail estimates. These findings suggest that daily cross-market climate-risk transmission is limited and that connectedness measures should be interpreted with appropriate benchmarking, formal inference, and careful covariance specification, particularly when considered for supervisory or risk-monitoring applications.