The productivity of transport infrastructure investment: A meta-analysis of empirical evidence

Abstract

Investments in transport infrastructure have been widely used by decision makers to encourage economic growth, particularly during periods of economic downturn. There has been extensive research on the linkage between transport infrastructure and economic performance since the late 1980s, characterised by widely varying evidence. We conduct a meta-analysis of the empirical evidence on the output elasticity of transport infrastructure, based on a sample of 563 estimates obtained from 33 studies. Previous meta-analyses have focused on total public capital and hence cannot appropriately explain the wide variation in the productivity effect of transport infrastructure nor provide guidance to policymakers on the returns to investment in different types of transport infrastructure. Our results indicate that the existing estimates of the productivity effect of transport infrastructure can vary across main industry groups, tend to be higher for the US economy than for European countries, and are higher for roads compared to other modes of transport. The variation in the estimates of the output elasticity of transport is also explained by differences in the methods and data used in previous studies. Failing to control for unobserved heterogeneity and spurious associations tends to result in higher values, while failing to control for urbanisation and congestion levels leads to omitted variable bias. These findings can be used to inform future research on the choice of model specification and estimation and transport-related policy making.