The response of travellers to differing levels of service is of concern to policy-makers in two different ways. In the first place, there is the need to predict likely changes in demand, and in the second place, it is often important to be able to evaluate investment options which differ in the level of service offered. With evaluation, we are not merely interested in the change in demand, but also in the benefit (or disbenefit) perceived by existing users of a transport system. In the U.K. and certain other countries, these two areas have been largely integrated by means of the concept of generalised cost. In most work carried out in the late sixties and early seventies, generalised cost was usually a simple linear function of travel time and travel cost, in which the travel time was converted to cost units by multiplying by a 'value of time'. Although there are many aspects of time valuation which are controversial both in theory and in practice, the basic concept is sufficiently well known that the discussion in this paper will largely be conducted in these terms. With the subsequent development of discrete choice theory and random utility models, it has been possible to extend the notion of generalised cost considerably, and nowadays it would be more conventional to interpret generalised cost as 'utility'. Earlier problems about the units in which these concepts should be measured have been resolved, though care is still required, particularly for forecasting. However, the main concern of this paper will be to examine the implicit trade-off between components of utility in as far as they affect travel choices, with special attention being paid to values of time.(a) for the covering abstract of the conference see IRRD 290118.
Abstract