Bistuer, DavidChuliá Soler, HelenaUribe Gil, Jorge Mario2026-01-202026-01-202025https://hdl.handle.net/2445/225788Previous development studies have documented a positive relationship between economic complexity and better environmental outcomes, as well as highlighted policy avenues that could leverage economic complexity as a roadmap for decarbonization and green growth. We build on this perspective by empirically demonstrating—using recent advances in explainable and causal machine learning—that economic complexity is also meaningfully linked to climate change resilience. Specifically, we show that more complex economies tend to be less vulnerable to climate change due to their stronger adaptive and coping capacities. These capacities are evidenced by stronger institutions, better long-term health outcomes, and, notably, a higher proportion of people employed in R&D. Our findings also reveal a positive association between exposure to climate risk due to geography and complexity, but only in cases of extreme exposure. While exposure to climate change itself is beyond the reach of policy intervention, vulnerability is not. By using an economic complexity framework combined with investments in knowledge intensive intangibles and large-scale long-term health interventions, policymakers can align the seemingly divergent goals of climate resilience and decarbonization, which is crucial, especially for developing nations.46 p.application/pdfengcc-by-nc-nd, (c) Bistuer et al., 2025http://creativecommons.org/licenses/by-nc-nd/4.0/Avaluació del risc ambientalIntel·ligència artificialAprenentatge automàticEnvironmental risk assessmentArtificial intelligenceMachine learninginfo:eu-repo/semantics/workingPaperinfo:eu-repo/semantics/openAccess