This paper tests the elasticities to time, frequency and interchange implied by an approach which combines these three variables into a single term and compares this approach with models which estimate separate elasticities. The forecasts obtained from different model forms can be appreciably different.

The productive efficiency of the railway systems in 19 OECD countries is analysed. The empirical results show that: (i) railway systems with high dependence on public subsidies are significantly less efficient than similar railways with less dependence on subsidies; (ii) railways with a high degree of managerial autonomy from regulatory authorities tend to achieve higher efficiency.

It is concluded that aggregate approaches to forecasting demand may be appropriate for cheap investments, such as new stations, or an initial assessment of a wide range of options. For detailed consideration of expensive investments, such as new rail services, disaggregate methods based on RP and or SP data should be considered.