Research Group: Economics and Discrete Choice
The Economics and Discrete Choice research group has two main areas of activity; transport economics, and discrete choice.
Transport Economics
For more than 40 years, the economics of the transport sector has represented a constant element of the ITS research portfolio. Whilst the conceptual and practical issues facing transport economists have changed with the times, economics as a discipline remains as relevant as ever within policy, practice and research.
Throughout four decades of sustained output, ITS has built a reputation as an international centre of excellence in transport economics, noted for its esteemed alumni, such as Professors Quarmby, Williams and Ortúzar; its prestigious established chair, occupied by only two long-serving Professors, namely Gwilliam and Nash; and its Masters programme in transport economics, unique in the UK.
ITS has always maintained a diverse range of interests across the field of transport economics, on both demand and supply sides, and including both theoretical and applied contributions. With regards to theory, some notable highlights include influential contributions on the value of travel time, marginal cost pricing, and transport and the economy. As regards applications, the Institute's economists are in frequent demand by UK and European policymakers, such as the Department for Transport, National Audit Office, Competition Commission, World Bank, Scottish Government, Welsh Assembly, Office of the Rail Regulator, to name just a few of our clients and sponsors. The following four strands have been identified as important foci which build upon existing strengths whilst responding to current and future research challenges and opportunities:
Demand
The Institute has a long-standing tradition of developing models of transport demand, at the levels of both the individual traveller/user and the market, and applying these models to policy analysis across a range of transport modes and contexts. Recent and on-going studies have included 'meta-analysis' of values of time and elasticities, bus soft factors and concessionary fares, all sponsored by the Department for Transport. ITS also has a long-term role in the development of the 'Passenger Demand Forecasting Handbook' (PDFH), and research on bus elasticities and values of time informs TRL Report 593 'The Demand for Public Transport'. Looking to the future, theoretical developments are needed to better represent traveller/user responses to particular features of transport systems, notably quality, capacity and reliability. More generally, modellers will be challenged to develop models that are founded on more flexible assumptions than simply an individual traveller making independent and static decisions under certainty. It is envisaged that there will be a greater call for models accounting for inter-relationships between decision-makers (e.g. in the context of household decision-making, or the competitive behaviour of freight shippers), where such decisions could be dynamic (e.g. patterns of vehicle ownership/scrappage over time) and have long-lasting implications (e.g. in the context of the carbon management agenda). On the latter, it is anticipated that there will be specific opportunities for demand forecasting research in relation to alternative-fuel vehicles, and forecasting of freight volumes and their economic and societal impacts based on governmental sustainability initiatives.
Costs and efficiency
ITS has considerable expertise in the area of efficiency analysis, drawing on a wide range of methodologies including the use of index number approaches to productivity comparisons, productivity and efficiency analysis using econometric stochastic frontier approaches as well as linear programming techniques, and cost accounting modelling. ITS carried out an econometric international benchmarking study which formed part of the evidence used by the Office of Rail Regulation (ORR) in reaching its efficiency determination in respect of Britain's rail infrastructure company, Network Rail. ITS has also carried out important econometric work in respect of British passenger train operating companies since privatisation, as part of a wider assessment of the impact of franchising in Britain.
Valuation
Drawing upon expertise in behavioural modelling from the 'modelling' theme, ITS has a proven track record in providing advice to government on the valuation of a diverse range of attributes and products including, for example, the value of travel time and travel time reliability, value of information, value of simplification, value of vehicle quality, and the value of environmental pollutants such as noise and vehicle emissions. This area is as vibrant as ever, evidenced by the likes of ongoing work for the Department for Transport scoping out the requirements for an update to the UK value of travel time.
A driving force behind this area of work has been ITS' expertise in so-called 'stated preference' experimentation, which involves the use of artificial experiments to elicit travellers' (and consumers' more generally) preferences among alternative travel options. The ITS profile in this area arose from early contributions to developments in stated preference methods during the 1970s and 1980s, as well as pioneering applications to the value of travel time. More recently, methodological and practical challenges have emerged which, to varying degrees, have caused Stated Preference practitioners to review and enhance their methods. This is seen as a continuing theme, creating need and opportunity for ITS to explore new avenues for Stated Preference methods, for example in relation to 'efficient' experimental design, the designing of experiments under risk/uncertainty (e.g. building on recent research on option values for the Department for Transport), and the correspondence between Stated Preference methods and the emerging field of experimental economics. On a more applied front, history shows that the market for valuation studies is ever diversifying; ITS continues to explore new market opportunities for the application of Stated Preference methods.
Regulation, pricing and incentives
ITS has substantial experience in evaluating cross modal competition, competition for the market through franchising and tendering, and competition between companies and products. Staff at ITS have advised the Department for Transport, Strategic Rail Authority, Office for Rail Regulation, Competition Commission and Office of Fair Trading on competition issues. Whilst this is a continuing market for transport economic research at ITS, there are new opportunities for a broadening of scope. More specifically, research has applied game-theory to the analysis of optimal regulation, e.g. competition between pricing authorities, resource allocation games in local transport, as well as competition between operators.
Discrete choice
Discrete choice models are used to understand, represent, and forecast human behaviour in a variety of contexts, with numerous transport applications, including for example mode choice and route choice behaviour. ITS is at the forefront of the development and application of such models, and the outputs from this work are routinely used to guide policy and infrastructure decisions. A number of crucial future directions of research in this area are identified below.
Foundations of behavioural models
There is inherent tension between the behavioural and theoretical foundations in travel behaviour research. On the one hand, there is appetite for grounding models more firmly in theories of behaviour per se, while on the other hand, an important application of behavioural models is economic analysis, which brings a need to ensure the validity of models in economic terms. An important avenue for work in this context is the development of models that are consistent with our observations of real world behaviour while avoiding violations of economic theory.
Reconciling modelled behaviour with real world behaviour
There is a growing realisation that while the widely used discrete choice models can be useful in representing individual choices, taking these choices out of their context and treating them in isolation and fixed at a specific point in time leads to a significant drop in realism. Similarly, it is important to recognise the potential role played by latent attitudes, strategies, and societal influences in models. ITS, with its extensive skills and experience in behavioural modelling and the psychological study of human behaviour, is particularly well suited for work looking at reconciling modelled behaviour with real world behaviour.
Experimental design for stated preference (SP) surveys
ITS has historically been one of the leading groups in the area of SP survey design. Given the growing reliance on SP data in behavioural modelling, and the realisation that standard design techniques are far from optimal, much research remains to be done in this field, and once again, ITS should be expected to be at the forefront of such developments. Alongside the development of surveys that allow more robust results to be obtained from smaller samples, a crucial area of work consists of analysing and reducing the effects of the miss-match between real world choice scenarios and those faced in SP questionnaires. Furthermore, the advantages of a new class of models that allow for an increasingly realistic representation of real world behaviour are potentially cancelled out if the data on which they are estimated are not representative of real world behaviour. A final important direction for ITS work in this area is the continued development of adaptive stated preference survey techniques, where ITS is one of only a few active groups globally.
Implementation and good practice
A disproportionally small share of the wider research effort is dedicated to the topic of model operationalisation, which includes for example aggregation and forecasting. The work required in this context includes bringing advanced models into applied work, where the difficulties faced in operationalisation increase rapidly with model complexity. A further weakness when compared with econometrics more generally is the lack of work on inference and diagnostics. ITS, with its long track record of advising other academics, practitioners, and policy makers, has a crucial role to play in developing and disseminating guidance for good practice in the field.
For further information about the Economics Research Group and its activities, please contact Dr Stephane Hess
