In this thesis I use a novel approach to estimate the economic impact of the 2021 Limburg flood using high-frequency transaction data from ABN AMRO bank. High- frequency bank transaction data have previously been proven valuable in accessing the economic impact of the COVID-19 pandemic (Neuteboom et al., 2021). To the best of my knowledge, this type of data has not been used to estimate the economic impacts of a natural catastrophe. I focus on a synthetic difference-in-differences methodology to estimate the impact. I find that the economic impact is 18,045 EUR on average of extra spending per inundated individual in Valkenburg aan de Geul, the most heavily hit area. This is very similar to the damage estimate of the ENW (2021) of 18,713 EUR on average per person for the inundated in Valkenburg aan de Geul. Further- more, the duration of the economic impact for the inundated is roughly 35 weeks on average. Finally, I did not find a measurable economic impact to uninundated and evacuated individuals. In summary, high-frequency bank transaction data paired with a synthetic difference-in-differences model is a reliable gauge of the economic impact of a flood and should be used to estimate the economic impact of future natural disasters. Additionally, it can be be the empirical foundation for calibrating existing damage models.