Building new roads really does create extra traffic: a response to Prakash et al.

TLDR

Prakash et al. (Applied Economics, 2001) argued that induced travel effects do not occur. This paper challenges that claim by critiquing the methodology and assumptions used. The authors analyze UK road expenditure data and show that expenditure is not a reliable proxy for actual road capacity built.

Abstract

A recent article by Prakash et al. (Applied Economics, 33, 1579–85, 2001) asserted that induced travel effects do not occur. This paper is criticized on several grounds. It disregards much of the recent work in this area that has empirically estimated induced travel relationships. The models specified are inappropriate for properly addressing this question, both in their use of road expenditure data (based on a misunderstanding of how this may relate to traffic growth) and specification of a model that does not account for other variables that generally have a large effect on traffic growth (notably population and income growth). The evidence in the literature is summarized and an analysis of UK road expenditure data shows that expenditure is not a good measure of actual road capacity that is built.