Tropical cyclones (TCs) are devastating extreme weather events, impacting millions annually and causing significant economic losses. While the immediate economic damage is substantial, mounting evidence suggests that the long-term effects on economic growth can persist for more than a decade, leading to overlapping repercussions from successive events in vulnerable regions. This prolonged economic downturn can severely hamper development prospects, particularly in low- and middle-income countries (LMICs). The projected increase in frequency and intensity of TCs under global warming further exacerbates this issue, potentially amplifying long-term growth losses without additional adaptation measures. This study addresses a critical gap in current climate policy assessments by explicitly incorporating the long-term economic impacts of TCs into estimates of the social cost of carbon (SCC). The SCC, a key metric used to inform climate policy, typically focuses on the direct, immediate costs of greenhouse gas emissions. However, it generally fails to account for the cascading and persistent consequences of extreme weather events like TCs, leading to a potential underestimation of the true societal costs of climate change and inadequate planning for adaptation needs. Therefore, this research develops a robust framework to quantify the long-term economic consequences of TCs and integrate them into SCC estimates, thereby enhancing the accuracy of climate policy evaluations and adaptation strategies.

Existing research on TC damage functions varies widely in complexity, from simple estimations linking damages to global mean temperature (GMT) and socioeconomic factors to more sophisticated event-based approaches that consider individual storm characteristics. Many studies focus solely on the USA or extrapolate US-based damage functions to other regions, neglecting the unique climatological and socioeconomic vulnerabilities of other TC-prone countries. While some studies have explored region-specific damage functions for countries like the Philippines and China, comprehensive global sets of region-specific functions accounting for damage persistence are limited. Previous projections often rely on statistical links between TC predictors and reported direct damages, neglecting the persistence of these damages within the economic system, leading to potential underestimation of future damages. This study builds upon previous research by explicitly incorporating the persistence of TC impacts on economic growth and generating regionally specific damage functions to address this limitation, leading to more accurate and regionally informed policy recommendations.

This study employs a three-component modeling framework to quantify future TC damages. First, it estimates the historical growth responses of 41 TC-affected countries to TC strikes, accounting for the persistent nature of the economic damage using a three-way fixed effects panel model that includes temperature and TC exposure effects as exogenous predictors. This model avoids endogeneity issues of earlier studies and employs a maximum entropy bootstrapping method to account for uncertainty in historical growth estimates. The model is designed to identify the independent contribution of TC impacts on economic growth, independent of the effects of temperature. Second, it projects probabilistic, event-based TC damages using a TC emulator to generate synthetic time series of TCs with landfalls under various emission (Representative Concentration Pathways – RCPs) and socioeconomic development (Shared Socioeconomic Pathways – SSPs) scenarios. This allows for a systematic assessment of the compounding uncertainties involved. Third, it derives event-based, country-specific, temperature-dependent damage functions that incorporate the persistence of damages in the economic system from the growth losses calculated across scenario combinations. The use of synthetic TCs enables the model to account for changes in TC climatology under global warming, while the inclusion of multiple lag years in the growth response analysis captures the persistent effects on the economy. The framework then projects two key policy-relevant metrics: i) discounted annual damage (DAD), reflecting additional economic burdens from climate change, and ii) the TC-induced contribution to the social cost of carbon (SCC). These metrics are projected under different RCP-SSP-discounting combinations to assess the sensitivity of the results to various assumptions and uncertainties. The DAD is calculated by summing the discounted differences in national GDP trajectories with and without additional warming across the future period of 2010–2100. The SCC calculation follows the approach by Ricke et al. (2018), adapted to incorporate the TC-induced damages using the temperature-dependent damage functions. It accounts for uncertainty in historical growth response, TC impact response to future emissions, and the response of global mean temperature to emission pulses.

The analysis reveals several crucial findings: First, historical data shows that TC strikes reduce economic growth in the short, medium, and long term. Cumulative growth losses increase monotonically with the number of lag years considered, plateauing after 6–13 lags, highlighting the importance of accounting for the accumulation of persistent effects. The degree of growth loss varies significantly across countries, with heavily exposed countries like Japan and the Philippines experiencing considerably larger losses than the USA. Second, future projections of TC damages, factoring in six compounding uncertainty dimensions (future greenhouse gas emissions, socioeconomic development, discounting, historical growth response to TCs, the response of TC impacts to emissions, and the temperature response to emission pulses), show a significant global economic impact. Under a middle-of-the-road scenario (RCP6.0-SSP2, Ricke’s discounting), the median global discounted annual damage (DAD) is estimated to reach 0.18% of global GDP in 2021. Higher emission scenarios and lower discount rates lead to substantially higher DAD estimates. In absolute terms, high-income countries show higher per-capita DAD, but this difference diminishes when DAD is considered relative to average household income. This relative measure reveals that some LMICs face comparable or even greater economic burdens than the USA. Third, country-level temperature-dependent damage functions demonstrate a robust relationship between increasing GMT and TC-induced growth losses for 37 out of 41 countries. This relationship holds across various income groups, indicating that economic growth in high-income countries is also negatively affected by TCs, highlighting the limitations of historically implemented adaptation measures. Fourth, the inclusion of TC impacts substantially increases the global SCC. The median global SCC rises by 22%, from US$173 to US$212 per tonne of CO₂ under the main specification. The largest increases in country-level TC-induced SCC are observed in the USA, Japan, Taiwan, China, and India—major greenhouse gas emitting countries heavily exposed to TCs. The impact of discounting is the largest source of uncertainty across all scenarios, with lower discount rates significantly increasing both DAD and SCC.

This study significantly advances our understanding of the long-term economic consequences of TCs and their contribution to the social cost of carbon. By explicitly integrating the persistent impacts of TCs into the SCC framework, the findings demonstrate that current SCC estimates may significantly underestimate the true costs of climate change. The considerable increase in SCC when accounting for TC impacts underscores the need for more comprehensive assessments that consider the cascading and long-lasting effects of extreme weather events. The disparity in relative economic losses between high-income and LMICs, when considering DAD relative to average household income, emphasizes the disproportionate impact on vulnerable populations. The strong relationship between GMT and TC-induced growth losses across income levels highlights the urgency of mitigation and adaptation efforts. The methodological framework presented in this study offers a valuable tool for future assessments, applicable not only to TCs but also to other extreme weather events. By providing robust, regionally specific damage functions, the research contributes significantly to informed climate policy decisions.

This study provides a comprehensive assessment of the long-term economic impacts of tropical cyclones, incorporating their effects into the social cost of carbon. The findings highlight the significant underestimation of climate change costs when neglecting the persistent effects of these events. The framework developed can be extended to other extreme weather phenomena, providing valuable insights for policy makers. Future research could refine the methodology by incorporating more detailed factors, such as storm surge and rainfall damages, and examining the effectiveness of various adaptation strategies.

The study assumes that future growth responses to TCs will mimic historical responses, neglecting potential future adaptation measures or maladaptation effects. The damage estimates are based solely on wind fields, excluding the impacts of rainfall and storm surge. The analysis focuses on TCs, potentially underestimating the overall contribution of other extreme weather events to the global SCC. The model does not endogenously model mitigation responses or their impact on global warming and socioeconomic development.