In this paper, the authors put to trial the alleged role of investment in transport and communications infrastructure in economic growth in The Netherlands in the second half of the nineteenth century. To do so, they combine a new data set with data-oriented econometric techniques. Testing of the main hypothesis by applying Granger's concept of causation in a vector auto-regression model reveals that the development of infrastructure `caused' Dutch economic growth. In order to investigate the underlying mechanisms, the authors enhance the basic model with impulse-response analysis. This leads to a clear response pattern of GDP to a change in transport infrastructure investment. The response of GDP is explained as being induced by three underlying mechanisms: (1) positive forward linkage effects (mainly through reductions in the costs of transport and communications; (2) positive backward linkage effects (mainly through expenditure and income effects); and (3) negative transitional dynamics (mainly through changes in the spatial setting of the economy). (A)
Abstract