Beat the Heat, the Role of Heat Waves and Droughts in Regional EU Economies

Europe is increasingly exposed to heat waves and droughts, yet their short-term sector-specific economic impacts remain difficult to quantify in real time. We propose climate-augmented predictive models for regional real growth in gross per capita value added in 1,117 EU regions over 2002–2022, combining standard economic indicators with high-frequency climate variables capturing heat waves and different types of drought at different temporal horizons. With mixed-frequency designs and machine learning (ML) methods, we show that climate predictors using different ML models, Random Forest and XGBoost, improve predictive accuracy relative to lin-ear benchmarks for the climate-sensitive sector of agriculture. The results indicate that, while alternative sectoral aggregations exhibit modest gains in predictive performance from the applic-ation of our ML models, climate-augmented specifications do not significantly outperform their economic benchmark counterparts. Heat wave indicators consistently contribute to predictive performance, while the importance of droughts varies by sector. XGBoost simulations of an ex-treme compound heat-and-drought scenario indicate that agricultural growth can occasionally decline by nearly ten percentage points; industry (encompassing mining, manufacturing, energy, and water supply) experiences smaller losses, and manufacturing alone is the most resilient sec-tor. In general, the results demonstrate that ML models better capture nonlinear, seasonal and spatial climate–economic interactions, highlighting the value of climate-augmented predictive modelling for early warning, regional fiscal planning, and targeted adaptation.

Keywords: climate extremes, heat waves, droughts, regional predictions, machine learning. JEL Classification: C53, E37, Q54, R15.