Economic Evaluation and Transport Modelling: Theory and Practice

A central role for Transport Modelling is to allow alternative strategies to be assessed. Most people involved in modelling are aware of the general principles of economic evaluation, but it remains a specialised topic, and tends to take second place in the literature to the more glamorous discussion of models. However, if the assessment tools are not integrated with the models themselves, much of the model sophistication may be wasted. It was therefore considered appropriate to commission a resource paper for the 10 IATBR Conference. The paper has a number of aims. It begins by discussing the theoretical foundations of evaluation in economics, and, without being excessively academic, attempts to give an understanding of how the theory has developed. It then contrasts this with current practice, examining the validity of the simplifications which are typically made, as well as the developments in this area. Continuing progress in the theory of discrete choice models, as well as the development of their use in transport modelling, has led to some key theoretical conclusions, which serve usefully to integrate modelling and evaluation. However, there are still some conflicts between the ideal theoretical requirements and what is considered practical. The paper is primarily aimed at mathematical modellers rather than at economists, most of whom will be familiar with much of the material. It includes the essential theory of the consumer that is available in economic textbooks, but without requiring the reader to become too involved in general questions of economic theory. Particular emphasis is placed on practical issues. Finally, the paper draws attention to a number of controversial areas, where further work is required, both theoretical and practical. An admission is due at the outset. Part of my intention for the paper was to effect a synthesis of the theory of modelling and evaluation in a way which I did not believe was available in an accessible form. I had not seen the excellent paper by Jara-Diaz & Farah (1988) which does indeed achieve such a synthesis, although I was aware of other work by Jara-Diaz et al. I believe that we are in substantial agreement, though there are some nuances of presentational difference. There remain some unresolved issues. It is hoped that, both at the conference itself and thereafter, this paper may assist in stimulating their resolution. It is also my intention that the final version of the paper should benefit substantially from advice and criticism received. In particular, appropriate contributions from others involved in this area will be gratefully received, incorporated, and acknowledged! Since the practical issues discussed in Section 3 are rather heavily based on the experience in the UK, examples demonstrating different approaches from other countries would also be most welcome.