The electricity industry in most developed countries has been restructured over recent decades with the aim of improving both service quality and firm performance. Regulated segments (e.g. transmission) still provide the infrastructure for the competitive segments and represent a significant share of the total price paid by final customers. However there is a lack of empirical studies that analyse firms' performance in the electricity transmission sector. In this paper an empirical analysis of US electricity transmission companies is conducted for the period 2001–2009. We use alternative stochastic frontier models that allow us to identify the determinants of firms' inefficiency. These models also permit us to control for weather conditions, potentially one of the most decisive uncontrollable factors in electricity transmission. Our results suggest that weather conditions clearly have an influence on transmission costs and that there is room for improvement in the management of US electricity transmission systems. Regulators should also be aware that more adverse conditions generate higher levels of inefficiency, and that achieving long-term efficiency improvements tends to worsen firms' short-term relative performance.