自然灾害的增长效应：1970-2023年全球新数据集的证据

The Growth Effects ofNatural Disasters Evidence From A Novel Global DatasetOver 1970-2023 Ha Nguyen, Mehdi Raissi, Bruno Versailles and Alice TianboZhang WP/26/106 IMF Working Papersdescribe research inprogress by the author(s) and are published toelicit comments and to encourage debate.The views expressed in IMF Working Papers arethose of the author(s) and do not necessarilyrepresent the views of the IMF, its Executive Board,or IMF management. 2026JUN IMF Working Paper Strategy, Policy and Review Department The Growth Effects of Natural Disasters: EvidenceFromA Novel Global Dataset Over 1970-2023Prepared byHa Nguyen, Mehdi Raissi, Bruno Versailles and Alice Tianbo Zhang Authorized for distribution by Eugenio CeruttiJune2026 IMF Working Papersdescribe research in progress by the author(s) and are published to elicitcomments and to encourage debate.The views expressed in IMF Working Papers are those of theauthor(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management. ABSTRACT:We construct standardized climate anomalies from daily observations and carefully calibratephysical thresholds to identify storms, floods, droughts, heatwaves, and cold snaps across 196 countries overthe period 1970–2023. Using a local projections framework, we estimate the contemporaneous and 2-yeareffects of each disaster type and collectively on real GDP growth. We find that storms, floods, droughts, andheatwaves significantly reduce growth on impact (by roughly 0.1–0.2 percentage points on average), with thelargest effects observed in emerging markets and developing economies. Cold spells have no statisticallysignificant impacts. Our estimations also indicate that the initial drop in GDP growth is often not fully offset by aquick rebound, leaving GDP below the pre-disaster trend in the subsequent two years. Severe disastersimpose far larger costs. Catastrophic floods can lower growth by up to 3 percentage points (with once-in-100-year storms or heatwaves reducing growth by ~0.5pp and extreme droughts by ~1pp). By combining theestimated global coefficients with each country’s own disaster intensitystatistics, we translate the aggregateresults into localized growth effects and cross-check them against estimates from a dynamic heterogenouspanel model. Finally, rolling-window estimates indicate that the contemporaneous growth impact of storms andheatwaves has attenuated in recent decades, whereas droughts have become increasingly damaging,reflecting divergent adaptation and vulnerability trends over time. RECOMMENDED CITATION:Nguyen, H., Raissi, M., Versailles, B., & Zhang, A. T. (2026). The GrowthEffects of Natural Disasters: Evidence From A Novel Global Dataset Over 1970-2023. IMF Working Paper No.WP/26/106. International Monetary Fund The Growth Effects of NaturalDisasters: EvidenceFroma Novel Global Dataset Over 1970-2023 Prepared byHa Nguyen, Mehdi Raissi, Bruno Versailles and Alice TianboZhang1 Contents 1.Introduction.....................................................................................................................................................32. Methodology....................................................................................................................................................53. Results...........................................................................................................................................................104. Extensions.....................................................................................................................................................215. Conclusions...................................................................................................................................................25References.........................................................................................................................................................26Appendix I. Data and Code Availability...........................................................................................................28Appendix II. Comparison with EM-DAT...........................................................................................................29Appendix III. Impact of Extreme Weather Events on GDP Growth...............................................................32 1.Introduction Natural disasters are increasingly recognized as critical macroeconomic shocks. A substantial body ofempirical research has examined their macroeconomic consequences(Mitra et al. 2025), with studiesoffering mixed conclusions. Skidmore and Toya (2002) suggested that frequent natural disasters mightspur long-run growth through capital renewal and institutional reform. Subsequent analyses, includingNoy (2009) and Raddatz (2007), documented significantnegativeshort-run growth effects, especially inlow-income countries and small island economies. More recent work has emphasized the impo