The question “What has been achieved by a reform of the power sector” remains important for governments since the beginning of liberalisation 20 years ago. Usually it is difficult to say if liberalisation brings efficiency advantages due to the absence of a perfect counterfactual. This study uses a unique setting – two regions in Russia that have similar load, generation and transmission characteristics – to identify a causal relation between liberalisation and subsequent performance. Given that the only difference between these two regions is that one was liberalised and the other one remained a natural monopoly, we can attribute any differences to changes in policy only. Moreover, our focus on the impacts in infrastructureidentifies changes that have not previously been explored. We find that primary energy efficiency increased in the monopoly region and decreased in the market region. There are three possible explanations for this effect. The first explanation is change in unit commitment in both regions resulting in different overall efficiency. The second one is deterioration of heat sector caused by decoupling of power and heat in the liberalised region. The third one is investment in more expensive and more efficient equipment in the monopoly region.