《保险研究》20260403-《气候变化下中国洪水经济损失评估及其保险风险量化》（罗浩、苏少军、汪倩）

［摘   要］全球气候变化导致洪水灾害频发,对中国社会经济和保险体系构成严峻挑战。本研究构建涵盖气候情景、洪水模拟、经济损失和保险风险的综合评估框架,利用ISIMIP2b多模式气候情景驱动CaMa-Flood模型,在县域尺度结合SSP2社会经济路径数据与深度-损失函数,评估中国未来中期（2024-2059年）和未来后期（2060-2095年）的洪水直接经济损失,并在省级尺度测算纯风险损失率与巨灾风险附加费率。研究发现,中国未来洪水经济损失率呈上升趋势,高排放情景（RCP8.5）下全国平均损失率未来中期达0.10%、未来后期超0.17%,增幅远超低排放情景（RCP2.6）的0.05%和0.06%；空间上呈现出在既有洪灾频发区域进一步增长或扩张的趋势。保险风险评估显示,高排放情景下多数省份纯风险损失率显著升高,巨灾风险附加费率持续攀升,华东、华中、华南及西南地区成为未来高风险区。本研究给出了未来不同气候情景下保险风险指标的定量评估结果及其空间分布,并探索了在缺乏理赔数据条件下利用模拟损失数据开展保险定价的方法,从而增强了评估的透明度和可重复性。研究成果可为保险机构实施精细化定价、优化产品设计和资本管理提供科学支持,也为政府制定区域差异化防洪政策和推进气候适应战略提供量化依据。

［关键词］气候变化；洪水风险；多模式集合；保险风险量化

［基金项目］中国保险学会2024年度研究课题“气候风险传导机制及其对保险公司的影响研究——以洪水风险为例”(ISCKT2024-N-1-29)；上海市超级博士后激励项目(2023095)。

［作者简介］罗浩,上海立信会计金融学院保险学院讲师、中国太平洋保险(集团)股份有限公司博士后科研工作站博士后,研究方向:气候风险管理；苏少军,中国太平洋保险(集团)股份有限公司董事会秘书、高级工程师,博士,研究方向:保险公司战略管理；汪倩,中国太平洋保险(集团)股份有限公司战略研究中心(ESG办公室)副主任,研究方向:绿色金融。

Assessing Economic Losses from Floods in China under Climate Change and Quantifying Insurance Risks

LUO Hao,SU Shao-jun,WANG Qian

Abstract：Global climate change has led to increasingly frequent flood disasters,posing severe challenges to China’s socio-economic systems and insurance industry.This study develops an integrated assessment framework encompassing climate scenarios,flood simulation,economic losses,and insurance risks.Using multi-model climate scenarios from ISIMIP2b to drive the CaMa-Flood model,and combining county-level results with SSP2 socio-economic pathway data and depth-damage functions,we estimate direct flood economic losses in China for the future mid-term(2024-2059) and long-term(2060-2095).At the provincial level,we further calculate pure risk loss rates and catastrophe risk loading factors.The results show that China’s future flood economic loss rates exhibit an overall upward trend,with substantial differences across emission scenarios.Under the high-emission scenario(RCP8.5),the national average loss rate reaches about 0.10% in the mid-term and exceeds 0.17% in the long-term,far exceeding the corresponding values of 0.05% and 0.06% under the low-emission scenario(RCP2.6).Spatially,flood risk tends to intensify or expand in regions already prone to frequent flood disasters.Insurance risk assessments further indicate that,under high emissions,most provinces experience significant increases in pure risk loss rates and sustained rises in catastrophe risk loadings,with Eastern,Central,Southern,and Southwestern China emerging as major high-risk regions.This study provides quantitative estimates and spatial distributions of insurance risk indicators under alternative climate scenarios,and explores an insurance pricing approach based on simulated loss data in the absence of claims records,thereby enhancing transparency and reproducibility of risk assessment.The findings offer scientific support for insurers in implementing refined pricing,optimizing product design,and improving capital management,and provide a quantitative basis for governments to develop regionally differentiated flood-control policies and advance climate adaptation strategies.

Key words：climate change;flood risk;multi-model ensemble;insurance risk quantification