The application of stochastic frontier panel models in economic regulation: experience from the European rail sector

This paper shows how stochastic frontier panel techniques can be used by economic regulators to benchmark regulated firms against international best practice. We utilise a unique, panel dataset of European rail infrastructure managers (1996–2006). A time-varying inefficiency model, with firm-specific time paths for inefficiency, is adopted. The results were used in the 2008 regulatory review of the British infrastructure manager, Network Rail, and showed that the company faced an efficiency gap of around 40% against European best practice – in line with engineering-based evidence. More widely, the paper highlights the advantages of the inefficiency specification adopted for use in economic regulation.