Conventional endogenous growth models generate growth by using some kind of input in the production function that is for some reason not freely traded in a perfect market and, therefore, not provided at its true marginal costs (e.g. see Aghion and Howitt 1998; Barro and Sala-i-Martin 1995). Examples of such inputs include public inputs, which can be used in one way or another in a non-rival manner, or inputs that are protected by a patent. In this context, Bretschger (1996, pp. 89-90) argues that even if perfect markets would prevail on the firm level, because of spillovers on the aggregate level due to the (non-market) diffusion of knowledge, decreasing average costs result in the long run, which in turn represent the essential source for growth.
Moreover, empirical evidence indicates that the provision of infrastructure is an important condition, if not driving force, for economic growth. Aschauer (1989a,b,c), for example, explained the slow down of growth in the United States during the 1970s by the lack of public infrastructure provision. Although Aschauer’s findings have been challenged and questioned in various other studies since 1989, Poot (2000) finds that the vast majority of studies confirm a positive relation between growth and public investment in education and physical infrastructure. More recently, Colombier (2007, 2004) supports these findings with empirical evidence for Switzerland regarding public investments in education and transport infrastructure in a broad sense, this is, roads, electricity lines, gas and water pipelines.
Yet, over the last two decades the traditional way of providing such transport infrastructure by the government, and without any direct user charges, has been seriously challenged in various countries. Deregulation initiatives in the utility markets and a more widespread use of road pricing in several countries have led to a situation where transport infrastructure is increasingly provided subject to a user fee (e.g. see Niederprüm and Pickhardt 2002a,b; Bazart and Pickhardt 2004; Pickhardt 2005, 2007) . In particular, if the deregulation process is successful in the long run, competition in these sectors may eventually lead to marginal cost pricing or at least to efficient average cost pricing. But despite the positive effect of substantial gains in efficiency in the short run, for reasons noted above, a market like provision of transport infrastructures may not generate endogenous growth in the long run.
To this extent, the purpose of the present paper is to model transport infrastructure as an input in an endogenous growth model, by using alternative input specifications that allow for both a market and non-market orientated provision. In particular, we are interested in analyzing whether or not short run efficiency in providing transport infrastructure under market conditions is still compatible with generating endogenous growth in the long run.
The paper proceeds as follows. In the second section we introduce the endogenous growth model, discuss alternative input specifications and derive some results from incorporating these alternative input specifications into the endogenous growth model. Section three provides a discussion of the results and offers some policy recommendations. The final section concludes.