This paper updates knowledge about the marginal cost of railway maintenance. Using a panel dataset comprising 16 years, we test whether more data makes a difference to conclusions. In contrast to previous estimates using a shorter panel, maintenance costs are now demonstrated to exhibit a positive dynamic effect; an increase in maintenance cost during one year indicates the need for more maintenance also the next year. Moreover, the marginal cost from the dynamic model is larger than its static counterpart. We conclude that the use of dynamic models on longer time series may have charging implications in several EU member states, considering that their track access charges are based on econometric studies that use static models and short panel datasets.

本文更新了有关铁路维护边际成本的知识。我们使用包含16年的面板数据集，测试是否有更多数据对结论有所影响。与以前使用较短面板进行的估算相比，现在已证明维护成本具有积极的动态效果。一年中维护成本的增加表明第二年也需要更多维护。此外，动态模型的边际成本大于静态模型的边际成本。我们得出结论，考虑到动态跟踪模型的使用费是基于使用静态模型和短面板数据集的计量经济学研究得出的，因此在较长的时间序列上使用动态模型可能会对收费产生影响。