A Translog Cost Function for Urban Bus Transit

Abstract

SEVERAL authors-among them Fisher and Viton [8], Keeler et al. [i2], Pozdena [I 7], and Small [22, 23] have pointed out the apparent superiority of bus transit over other forms of mass transit. The conclusions of these analyses rest on the following framework: in a given transportation market structure, consider the full costs of providing a trip of certain characteristics. 'Full costs' are defined to include out-of-pocket costs of the providers of the service, as well as the time costs incurred by the users of the transit system and the external costs that the users impose on each other and the rest of the economy. That is, 'full cost' is the full resource cost of providing trips. If we compare, for different traffic volumes, the minimum full cost of providing trips on different modes, then bus transit is less costly than modern rapid-rail transit; and where travel volumes exceed about I OOO persons per hour, it is less costly than auto travel as well.1 In addition, bus transit has also been the source of one of the more interesting financial developments in the provision of urban transportation services. In New York City there are now some 50 bus routes operated under franchise from the New York City Board of Estimate which are able to set the quality of service offered (including the fare) in such a way as actually to make a profit (see Pignataro and Falcocchio [I6] for a description of the routes and Viton [30] for an analysis of the requisites for profitable service). Although other mass transit modes are sometimes able to cover variable operating costs, as is the case with the Lindenwold Line in the Philadelphia metropolitan area or the commuter rail service offered by the Illinois Central GulfW2 bus transit is the only mode which has shown itself able to cover both variable and capital costs. These developments impart considerable interest to the study of the economic characteristics of bus transit operations. We want to know the marginal cost of bus services, since this is a guide to optimal pricing. There are also several other economic effects of interest: for example, it is observed that the profitable bus operations alluded to are small in scale when compared to those provided by the major transit authorities. One hypothesis