This article presents an assessment of territorial vulnerability to natural hazards in Europe at the regional level (NUTS3). The novelty of the study lies in assessing vulnerability to natural hazards through a composite indicator analysis over a large extension (1395 territories in 32 different countries), and in analysing the relation between vulnerability and economic impacts of past disasters. For responding to the first goal, a principal component analysis (PCA) was performed over 25 indicators, previously grouped into susceptibility and coping capacity, and subsequently combined to obtain the final vulnerability. The main result is the spatial distribution of vulnerability to natural hazards across Europe through a normalised and comparative approach, which indicates that 288 out of 1395 regions presented a high or a very high level of vulnerability. They are concentrated in Eastern and Southern Europe, and in the Baltic Region, and the sum of their population lives in territories with high or very high vulnerability level, representing 20% of the total sample, i.e. 116 out of 528 million inhabitants. Regarding the methodology for analysing the relation between vulnerability and economic impacts, a spatial regression model has been used to combine hazard, exposure and vulnerability. The outcomes indicate a high level of agreement between vulnerability and the distribution of past economic impacts, which confirm that the indicator-based approach is a good proxy for assessing vulnerability to natural hazards. Knowing the distribution of vulnerability is of high relevance for targeting disaster risk management and climate change adaptation actions to the highest priority regions.