We build a frontier function model with technical and cost efficiency measures to assess the impact of energy costs on competitiveness in the aluminum industry, a heavy energy consumer, by identifying what may be attributed to price and quantity effects. First, we estimate a data envelopment analysis model and measure the efficiency scores using the Farrell and Tone approaches. Then, we explain efficiency scores with a set of environmental factors using a truncated regression model, applying the double-bootstrap method proposed by Simar and Wilson in 2007. By considering the Tone efficiency scores, we highlight the fact that efficiencies are differentiated depending on geographic region and production technologies. This could explain the loss of competitiveness observed in certain industrial units in recent years. We also highlight other factors that impact cost efficiency, such as global market share, multinational status and the exchange rate with respect to the dollar. These factors have a significant, positive effect on relative efficiency in terms of aluminum smelter costs. Our results provide a better understanding of the reasons for the changes in this industry, which has been marked by major technological and economic shifts, and allow us to assess the role of energy in these changes.