One cornerstone of EU's railway policy is that track user charges should be based on marginal costs for infrastructure use. This paper updates knowledge about the marginal cost of maintaining Sweden's railway network. Using an extended panel dataset, now comprising 16 years, we corroborate previous results using a static model framework. However, the results from the dynamic model show that an increase in maintenance cost during one year increases costs in the next year, which is contrasting previous estimates on a shorter panel dataset. We conclude that more data made a difference in a dynamic setting, but the estimated cost elasticities are rather robust in a European context.

This paper proposes a new, two-stage methodology to estimate the relative marginal cost of different types of vehicles running on the rail infrastructure. This information is important particularly where the infrastructure managers wish to differentiate the track access charges by vehicle type for the purpose of incentivizing the development and use of more track-friendly vehicles. EU legislation requires that the European infrastructure managers set the access charges based on the incremental (marginal) cost of the running trains on their networks. The novelty of the approach derives from the combination of: (1) engineering simulation methods that estimate the track damage caused by the rail vehicles; and (2) econometric methods that estimate the relationship between the actual maintenance costs and the different damage mechanisms. This two-stage approach fills an important gap in the literature, given the limitations of the existing ''singlestage'' engineering or econometric approaches in obtaining the relative marginal costs for different types of damage. The authors demonstrate the feasibility of the method using 45 track sections from Sweden, for which the data on maintenance costs are available together with relevant track and vehicle data for 2012 (supplied by the Swedish Transport Administration). The authors demonstrate the feasibility of producing summary, section-level damage measures for the three damage mechanisms (wear, rolling contact fatigue, and track settlement), which can be taken forward to the second stage. The econometric results of the second stage indicate that it is possible to obtain sensible relationships between cost and the different damage types, and thus produce relative marginal costs by the damage mechanism and in turn the vehicle type. Based on this feasibility study, tracksettlement has been found to be the most expensive (in terms of maintenance cost) of the three mechanisms, followed by the rolling contact fatigue and then the wear. Future applications should focus on larger datasets in order to produce the required degree of precision on the estimation of the marginal cost.

In this paper, we analyse how railway maintenance costs are affected by different levels of railway line capacity utilisation. Previous studies have focused on the wear and tear of the infrastructure, while this paper shows that it is important to also acknowledge the heterogeneity of the maintenance production environment. Specifically, we estimate marginal costs for traffic using econometric methods on a panel dataset from Sweden and show that these costs increase with line capacity utilisation. The results are significant considering that current EU regulation (2015/909) states that track access charges can be based on marginal costs, with the aim of creating an effective use of available infrastructure capacity.

No abstract

No abstract

In this paper, we combine engineering and economic methods to estimate the relative maintenance cost of different types of damage on the Swedish rail infrastructure. The engineering method is good at predicting damage from traffic, while the economic method is suitable for establishing a relationship between damage and cost. We exploit the best features of both methods in a two-stage approach and demonstrate its applicability for rail infrastructure charging, based on a sample of 143 track sections comprising about 11,000 km of track. The paper implements for the first time the method previously proposed in Smith et al., whilst also enhancing the method in several respects. We demonstrate how the estimated relative maintenance costs related to different damage mechanisms can be used to calculate the marginal cost of different vehicle types. The results are relevant for infrastructure managers in Europe who wish to differentiate their track access charges such that each vehicle pays its short run-marginal damage cost, which can support more efficient use of the rail infrastructure and influence vehicle design to minimize system costs.

In this paper we study the effect of contract design on the performance of railway maintenance in Sweden, using a panel data set over the period 2003-2013. The marginal effect of incentive intensity is estimated, showing that the power of incentive schemes improve performance as measured by the number of infrastructure failures. In addition, the performance incentive schemes result in a reallocation of effort from failures not causing train delays to failures causing train delays.

No abstract