Cost sharing in passenger transport models: specification, implementation, and impacts

A common practice in many transport models is to allocate all costs to the car driver meaning that car passengers are essentially free riders. This specification provides an easy way of integrating drivers and passengers and fits well with the general understanding that car drivers rarely trade passenger seats for money. However, in this paper, we question this premise and argue that cost sharing can, and most likely will, take place in a variety of forms different from simple out-offpocket money exchange. We also stress that from a modelling perspective, this issue is far from irrelevant as it influences not only demand elasticities but also the modelling of car occupancy rates which is one of the key variables when considering congestion, climate impacts, and network utilisation. In the paper we discuss the mechanisms by which costs are shared, consider how this can be operationalised in transport models, and finally, consider the consequences of cost sharing from a transport policy perspective. Based on empirical evidence from Denmark, we find that for trip purposes with high occupancy rates, cost elasticities and occupancy rates tend to be biased if cost sharing is not properly accounted for.

Automated summary

Many transport models allocate all costs to the car driver without considering the cost sharing among passengers. This paper questions this premise and argues that cost sharing can occur in various forms, which should be properly accounted for in transport models. The empirical evidence from Denmark suggests that not accounting for cost sharing may result in biased cost elasticities and occupancy rates.