Three essays on infrastructure investment: the Australian experience

Abstract

This thesis studies a number of crucial topics in relation to infrastructure investments and their effects on economic performance. By focusing on the Australian experience, the thesis addresses several critical issues which have not received adequate attention in the earlier literature. Chapter 2 examines the relationship between public infrastructure and productivity by estimating a production function. The findings of this chapter suggest that while aggregate analysis produces an implausibly large infrastructure effect, results from an approach which controls for the specific characteristics of the states are more plausible. In another piece of evidence, the estimation of an error-correction model reveals that a long-run identification and modelling of the relationship reflects the important positive role of infrastructure on productivity. Short-run dynamics, however, provide no support for a positive effect. In addition, applying a causality test suggests a long-run unidirectional causality running from public infrastructure to productivity. Chapter 3 examines the impact of the provision of public transport infrastructure on the cost structure of Australian industries within a context that recognises interindustry spillovers. The study extends the symmetric normalised quadratic cost function by incorporating public transportation capital as an external input and adapting the spatial econometric techniques into an industrial context to allow for industry spillovers in the cost analysis. The study finds that while public transport has a productive effect in reducing the cost of production, neglecting interindustry spillovers noticeably reduces this effect. Chapter 4 examines the optimality of the provision of infrastructure using a system of Euler equations to represent intertemporal efficiency conditions. The estimation results suggest that dealing with individual types of infrastructure investments at the state level is helpful for reaching realistic conclusions about infrastructure provision. In particular, the study finds that while the efficiency conditions are satisfied at the aggregate level, a disaggregate analysis which examines individual components of economic infrastructure reveals inefficiency in the provision of some types of infrastructure across the states. In addition, contrary to other methods, the efficiency approach produces a relatively plausible estimate of the infrastructure effect.