Cost Functions for Australia’s Railways

Abstract

This paper utilizes a unique data set comprising almost hundred years of Australian railway history and applies the translog and symmetric generalized McFadden functional forms to the data in order to estimate cost functions for Australia’s seven state-owned railway systems. It discusses the advantages and disadvantages of each functional form in light of the data, finding the translog to be superior, and the shortcomings of the data. It also explores economies of scale and density and suggests a simple method to calculate long-run marginal cost, which may be of use in applications such as economic regulation.