AbstractFollowing a disaster, residents of a community may be displaced from their damaged homes, leading to expensive and lengthy disruption, with many choosing to move away permanently. Population losses may hinder recovery and exacerbate inequalities across neighborhoods. This study considered household place attachment and identified groups with low place attachment along with expensive and slow postdisaster recovery. We developed a framework to integrate place attachment considerations into housing recovery simulations. We used data from the American Housing Survey to develop housing and neighborhood satisfaction models and identify the neighborhoods with the least-attached residents. A computational simulation framework was used to simulate postearthquake housing recovery for a community and assess expected costs and time frames. We used the triad of low place attachment, high cost, and slow recovery to identify households prone to permanently moving away from their communities. A case study of housing recovery after a hypothetical earthquake near San Francisco demonstrated the application of the methodology. We found that about 10% of the population in some neighborhoods are prone to moving away after a large earthquake. Households with low income, renters, and those in older buildings are most likely to have low place attachment and experience costly and slow recovery. Whereas existing approaches rely on heuristics, the approach and results in this paper provide quantitative means to assess potential population losses and inform efforts to reduce them. The framework to integrate place attachment into housing recovery simulations is versatile and employs publicly available information making it transferable to other communities.