Energy efficiency plays an essential role in energy conservation and emissions mitigation efforts in the building sector. This is especially important considering that the global building stock is expected to rapidly expand in the years to come. In this study, a global-scale modeling framework is developed to analyze the evolution of building energy intensity per floor area during 1971–2014, its relationship with economic development, and its future role in energy savings across 21 world regions by 2060. Results show that, for residential buildings, while most high-income and upper-middle-income regions see decreasing energy intensities and strong decoupling from economic development, the potential for further efficiency improvement is limited in the absence of significant socioeconomic and technological shifts. Lower-middle-income regions, often overlooked in analyses, will see large potential future residential energy savings from energy intensity reductions. Harnessing this potential will include, among other policies, stricter building efficiency standards in new construction. For the commercial sector, during 1971–2014, the energy intensity was reduced by 50% in high-income regions but increased by 193% and 44% in upper-middle and lower-middle-income regions, respectively. Given the large energy intensity reduction potential and rapid floor area growth, commercial buildings are increasingly important for energy saving in the future.

Carbon footprints and other environmentally extended input–output indicators are obtained as aggregations of emissions embodied in supply chains (EESCs), which express the emissions occurring in a specific production activity to satisfy a given volume of final demand. Here we derive theoretical approximations of the expectations of and covariances between EESCs, as a function of the expectations of and covariances between source data (technical coefficients, emission coefficients and final demand volumes) through a Taylor expansion. We report an empirical test of those approximations, using a sample of 5 global multi-regional input–output models in the year 2007, of which we extract 22 single-region input–output systems with 17 sectors. We find that approximations of multipliers perform better than those of EESC, and approximations of expectations perform better than those of covariances.