Spatial Autocorrelation of Alternative Infrastructural Indicators

Regional growth analysis often uses infrastructure endowment to explain regional development. In the literature, there are different methods to quantify public capital based on past investment or on data about infrastructure endowment. The aim of the paper is to use spatial autocorrelation across infrastructure indexes to assess the performance and quality of each infrastructure index. We expect that a good infrastructure index captures the spatial network characterizing infrastructure endowment. In this case the level of spatial autocorrelation across infrastructure endowments in Italian territorial areas should be significant and remarkable. The results of the spatial autocorrelation analysis show significant spatial patterns in the distribution of infrastructure endowment only for indicators expressed in physical terms. The unbiasedness of physical indicators with respect to the different efficiency of the public capital expenditure on the national territory can explain the results.