The Impact of Rail Fares Complexity on Demand

The Impact of Rail Fares Complexity on Demand


Paul Metcalfe, PJM Economics, Chris Heywood, Accent, Rob Sheldon, Accent


This study uses an innovative stated preference choice experiment to explore, and quantify, the effects on the demand for rail travel due to simplifying, or making more complex, the range of ticket types offered to passengers.


The Impact of Fares Complexity on Demand
Paul Metcalfe, PJM Economics
Chris Heywood and Rob Sheldon, Accent
Tony Magee, Association of Train Operating Companies

Before the 1980s, fares were simple and easily understood. Yet from the 1980s onwards, especially since the time of privatisation, the variety of fares available has increased dramatically, leading to concerns regarding complexity and lack of clarity. In response to these concerns, the fares regime was simplified somewhat in 2008. Since then, however, Advance fares have occupied a growing proportion of all fares bought, and there have been continued concerns voiced regarding the overall complexity of the fares on offer.

The impact of fare complexity on demand has been explored in two recent studies: Accent (2013) “Demand Impact of Fares Structures and Simplification”, commissioned by the UK Passenger Demand Forecasting Council (PDFC), and SPAFT (2011) “Understanding and Testing Passenger Perceptions of Complexity in Relation to Fares and Ticketing”, commissioned by the Office of Rail Regulation (ORR). Accent (2013) deliberately excluded Advance fares, however, and so it is still unclear what effect simplifying the fares regime with respect to Advance fares would have on demand, or what type of fares regime would be optimal. SPAFT (2011) did incorporate an exploration of Advance fares; however, it did not go so far as to quantify the elasticities of demand or revenue with respect to simplification, however that might be defined, and nor did it appraise the desirability of alternative fares regimes.

The present study was commissioned by PDFC to explore, and quantify, the effects of a range of potential simplifications to the fares regime on demand for rail travel. The principal objective was to quantify the trade-offs between fare structure complexity, passenger demand and revenue with a particular focus on the effect of including Advance fares in rail fare structures. This included estimating the deterrence effect on passenger demand of fare structures which contain Advance fares over simpler fare structures, based solely on walk-up single and return tickets.

Additional objectives included the following:

• To assess a number of intermediate fares structures (between simple and complex, based on the range of tickets, number of restrictions, conditions and validities applicable to tickets) to establish the structure which offers the optimal balance between measurable deterrence and revenue return

• To develop coherent guidance on the demand impacts (quantified if feasible) of fares structures including Advance fares that can be used alongside existing guidelines to calculate demand elasticities

• To develop a tool to enable train operating companies to identify the optimum degree of price differentiation for a given market.
Our approach to delivering the objectives of the study was based around an innovative type of stated preference (SP) choice experiment, which involved varying the complexity of the ticket type choice. Both rail users and non-users were interviewed and both segments were shown the same SP exercise.

It was understood at the outset that in order to explore variations in complexity, the current ticket-buying choice situation needed to be replicated as closely as possible, with all its existing complexity. Rather than attempt to create a simplified choice situation, as is generally considered good practice in SP design, our approach therefore involved generating choice situations that mirrored extremely closely the online booking engines actually used in practice by the Train Operating Companies (TOCs). In addition, all ticket type variants currently used by the TOCs, and on offer through the booking engine, were made available in the choice experiment, including those involving restrictions over which TOC’s services could be used on the ticket chosen.

The experimental variation in the choice experiment focussed on the types of ticket types available, their fare levels, and the duration of the services available to choose from under different ticket types. Importantly for the focus of the study, respondents were given the option not to travel by train. Given this experimental variation, a model could be estimated that explored the impact of complexity on whether or not to travel by train, and on ticket type choice, while controlling for fare levels and service durations.

The research for this study is currently underway, and as of the time of submitting this abstract, there are not yet any results. The intention is that the study will be complete by May 2016.


Association for European Transport