This paper develops a framework to assess physical climate risk in asset management by extending the Vasicek firm value model with downward jumps that represent extreme climate events. The model is calibrated using ND-GAIN country vulnerability scores and firm-level asset intensity (property, plant, and equipment over revenues), allowing firms to be grouped into clusters with similar risk profiles. The authors apply the model to a global dataset of over 10,000 stocks and compute expected and unexpected portfolio losses under climate-stressed scenarios. Results show that expected losses increase over time, with significant variation across sectors and regions; high-vulnerability and high-asset-intensity clusters suffer the most. The framework enables the calculation of climate-related Value-at-Risk add-ons, supporting the integration of physical risks into risk management and suggesting the need for additional capital buffers for exposed portfolios. The approach is practical, relying on publicly available data, and can be adapted to more granular loss estimates if available.