Urban scaling laws measure how urban quantities scale with population size. Residential electricity consumption, being fundamentally tied to basic household needs, is often assumed to scale linearly with population. However, empirical studies have reported divergent scaling behaviors for electricity use across urban systems in different countries. Here, we analyze scaling relationships of residential electricity consumption across Chinese cities from 2002 to 2023. Results show that residential electricity consumption exhibits persistent super-linear scaling with population (β = 1.17, 95% CI: 1.14-1.20 in 2023), meaning large cities consume disproportionately more residential electricity per capita. This super-linear pattern remains robust over time (β: 1.11-1.18). Comparatively, total and industrial electrical consumption generally also show a super-linear pattern but with a declining trend towards linear pattern. Scale-adjusted metropolitan indicators reveal pronounced spatial heterogeneity: over-consumption in residential electricity concentrates in southeastern coastal and southwestern inland cities, while under-consumption prevails across northern regions. Regression analyses identify economic growth, population density, and climate condition as key determinants of these spatial deviations. Our findings demonstrate that urban energy scaling is not universal, but varies across sectors and urban systems. • Residential electricity use rises disproportionately in larger cities. • Super-linear scaling persists across Chinese cities from 2002 to 2023. • Industrial and total electricity drift toward linear scaling. • Scale-adjusted metrics expose strong regional disparities. • Economy, density, and climate explain spatial deviations.
Shen et al. (Sun,) studied this question.