Eliciting the regulation of an economic system: the case of the French rail industry

Based on the modern theory of regulation, the analysis aims to characterize the effective economic regulation of the French railway industry. The methodology consists in econometrically testing various scenarios of regulation and determining which of these best fits the data. Using aggregate data on the overall passenger traffic for the incumbent French rail operator (RO), SNCF, the two behavioral hypotheses of reference which we consider –absence of regulation of the rail operator which acts as a pure monopoly, and price regulation of services supplied by the RO– are both statistically significant and do not subtract from each other. This result is certainly related to the fact that passenger services include both high speed train services, for which the RO has some entrepreneurial freedom, and regional transport services, which are regulated by local authorities. In any case however, as the presence of unobservable efforts exerted by the RO to improve its productivity is statistically relevant, one concludes that the RO is not fully and properly regulated. This emphasizes that the design of policy reforms must account for the incentives they create on the RO. The analysis also shows that the most statistically significant scenarios are the ones in which the access tariff imposed by the infrastructure manager is such that the revenue generated by the access tariff is equal to the infrastructure spending. The pricing of the access to the infrastructure network therefore does not seem to be governed by economic principles, but more by budget considerations. While data limitations does neither allow to understand all the facets of a complex reality, nor to claim a high level of precision in the measure of all the parameters of interest, we believe however that we provide an objective methodology to characterize the optimal economic policies for the railway sector, in particular because it yields realistic estimates of the main structural parameters. Indeed the empirical results suggest that the railway industry as a whole exhibits increasing returns to scale, which incidentally is not compatible with the presence of multiple firms. In addition, the elasticity of demand for railway transport is relatively high, an indication of the competitive constraints this mode of transport faces from other transport modes or induced traffic