ECONOMIC EFFICIENCY MEASUREMENT BY DATA ENVELOPMENT ANALYSIS : The Case of European Railway



ECONOMIC EFFICIENCY MEASUREMENT BY DATA ENVELOPMENT ANALYSIS : The Case of European Railway

Authors

EMMANUEL BOUGNA, Laboratory of Transport Economic, YVES CROZET, Laboratory of Transport Economic

Description

This study use Data Envelopment analysis and a logistique regression in order to measure economic efficiency in rail sector and to assess their determinants by the use of time series data of 40 European railways company between 2000 and 2010.

Abstract

By the end of the twentieth century, railroads were in dire straits, most national railway companies were heavily subsidized but the shares of railroads in total (inter-modal) transportation were, at best, stable.
In most European countries, rail passenger traffic largely decreased but the amount of subsidy is increasing one year to another. The question which arises is therefore whether how to revive rail transport in Europe. This question entails two main questions: from a passenger, point of views what determines the demand for rail transport? From a supply perspective what are the determinants of the efficiency of a railway companies. ? It is the last question that held our attention. In the econometric literature, several model and technical estimation were developed to measure efficiency of decision-making units (DMUs). These models include linear regression models and stochastic frontier analysis. Data Envelopment Analysis (DEA) is one of the latest additions to the bracket of these techniques.
DEA is essentially an optimization algorithm, which develops efficiency scores for all DMUs on a scale of zero to 100%, with units receiving 100% efficiency score being called efficient. Further a simple modification in the DEA model also accounts for scaling efficiencies especially for large sized DMUs. In this study Economic efficiency measurement of European Railway Company was done using the data on a sample of 40 European railways company between 2000 and 2010. A logit model subsequently allowed us to identify the main determinants of the economic efficiency. The finding shows that there is a positive correlation between economic efficiency and factors such as the degree of filling of passenger train (number of passengers per train), index of the use of infrastructure by freight services (the number of train km freight per km of the line) and the level of technology (the percentage of electrified lines). On the other hand there is a negative correlation between economic efficiency index and the degree of filling of train freight. We have shown that an undertaking, which has a very high density of train freight, does not necessarily realize a good performance in passenger transport and vice versa. A horizontal separation between freight services and passenger services can therefore lead to a reduction in the cost of train miles.

Publisher

Association for European Transport