Abstract

In this paper, we conduct a pan-European efficiency analysis to investigate the performance of European railways with a particular focus oil economics of vertical integration. We test the hypothesis that integrated railways realise economies of scope and, thus, produce railway services with a higher level of efficiency. To determine whether joint or separate production is more efficient, we apply a data envelopment analysis super-efficiency bootstrapping model which relates the efficiency for integrated production to a reference set consisting of separated firms which Use a different production technology. We find that for a majority of European railways, economics of scope exist.

Abstract

In their continuous battle against congestion and pollution, governments nowadays promote rail as an environmentally friendly Single Occupant Vehicle (SOV) alternative. For comparison, the energy consumption per person kilometre for rail is 0.14 kWh/pkm while that for a car is 0.48 kWh/pkm (Boussauw and Witlox 2009), and electric trains have no direct emissions of air pollutants like PM 10 and NO x. Modal shift policies often target the daily commute since home to work travel is concentrated in the congested peak hours and commuters' travel behaviour is more regular. This regular character seems to fit well with the scheduled nature of rail, and the central location of railway stations fits with the relative concentration of jobs in congested cities (Riguelle, Thomas, and Verhetsel 2007). Furthermore, both cities and public transport are predominant in the sustainable mobility paradigm (Banister 2008). Therefore, the location of jobs in city-regions and the link with rail use, are at the core of our analysis. Besides environmental objectives, employment concerns motivate governments to invest in railways. Indeed, rail can keep jobs in city centres accessible. This indicates that rail can contribute to all three dimensions of sustainability and sustainable transport, the environmental, the social and the economic dimension (Boschmann and Kwan 2008).