Conference Agenda

Constructed primarily to defend coastal economies from severe storms and sea surges, seawalls have served as an effective protective measure across the world for thousands of years. In the past decade, as climate change has intensified projections of stronger, more frequent cyclones as well as significantly rising sea levels, seawalls have become even more critical for sustaining coastal economies. Many countries are already planning expansions of seawall infrastructure in preparation for future sea level rise. This study examines the current role of seawalls in protecting China's coastal economies from typhoon impacts during 2015-2024. Our results have shown that increasing the share of coastlines protected by seawalls has significantly reduced the negative impacts of typhoons, high waves, and rising sea levels on coastal economies, as reflected by changes in nightlight intensity. These impacts mostly arise from reduced disruptions to industrial, residential, and transportation activities, and are concentrated in districts that are not directly coastal but are the most economically active within each coastal city, implying that the defensive effects of seawalls spillover beyond directly protected coastal districts. We also find that seawall maintenance significantly improves protective effectiveness, while co-location with offshore mangroves can reduce the protective benefits. Overall, our findings highlight the importance of seawalls in strengthening coastal resilience to extreme climate events and underscore their relevance under future climate change.

Understanding heterogeneity in the climate-economy relationship is crucial for gaining insight into potential adaptation mechanisms. We apply causal forests to panel data to identify variables that drive treatment effect heterogeneity in climatic conditions, focusing on temperature impacts on economic output. We systematically evaluate more than 260 socio-economic and geographic characteristics, measured as long-run averages, that could moderate the relationship and lead to heterogeneous vulnerability. To select relevant characteristics beyond baseline climate, we use three novel panel implementations of causal forests and employ the least absolute selection and shrinkage operator (Lasso) as a robustness check. Mean years of education emerges as a key determinant of heterogeneity: countries with lower years of education are the most vulnerable to temperature changes. We also find that the percentage of the cropland area equipped for irrigation, the mean share of the country area that is covered by mosaic natural vegetation, and the mean terrain ruggedness index, which measures elevation changes, are relevant factors in explaining cross-country differences in vulnerability. Incorporating the selected variables into a fixed-effects panel regression framework highlights that some countries, particularly in Africa, have been substantially more vulnerable to temperature changes than reported by past studies. These moderators help inform which cross-country characteristics could reflect adaptation.