Summary: Energy is consumed both by passenger and freight traffic, and it is generally not possible to apportion energy consumption exclusively to either of the traffic without making restrictive assumptions on the energy consumption norms of vehicles. The use of data envelopment analysis (DEA) has been suggested earlier in the literature for comparing energy efficiencies without requiring any such restrictive assumptions. In this paper, the applicability of DEA is further extended to estimate energy consumption of rail and/or road transport that would result in a pre-specified DEA efficiency. Using this new approach, effects of capturing some modal split in favour of rail transport on future energy consumption and emissions of carbon dioxide are presented in the paper via a scenario analysis. It is shown that if rail could capture 50% modal share in the years 2005–2006 and 2020–2021, it would save nearly 35% energy consumption and emissions compared to the situation when existing patterns of modal split are allowed to continue without policy intervention.