How Long Do Buildings Live: A Systematic Review and Meta-Analysis of Global Building Lifetimes

Building lifetime is a foundational parameter for modeling stock dynamics, informing environmental impact assessment and urban development, yet current lifetime assumptions frequently lack empirical grounding. Although well-established in medicine and demography, survival analysis concepts have not penetrated building stock research to the same extent. This study bridges this gap through a systematic review and meta-analysis of 258 median building lifetime estimates across 46 global studies. Central to our analysis is right-censoring, when a building's demolition remains unobserved at the end of the observation period, which, if unaccounted for in estimation, leads to underestimated lifetimes. We systematize five estimation approaches based on data type and the capacity to account for right-censoring, describe the currently available empirical evidence, and apply a meta-regression to identify significant predictors of median lifetimes: world region, use class, structural material, and right-censoring bias, the latter underestimating building lifetimes by 50% (95% CI: 49-51%). Of the five estimation approaches identified, three inherently consider right-censoring, one inherently cannot correct for it, and one requires an active correction step that is frequently omitted. Our work identifies critical data gaps and offers targeted recommendations for users and producers of lifetime estimates to improve transparency, reproducibility, and methodological rigor.