We propose two indexes to measure the agglomeration forces acting within and between different regions. Unlike the existing measures of agglomeration, our model-based indexes allow for simultaneous treatment of both aspects. Local plant diffusion in a given industry is modelled as a spatial error components process (SEC). Maximum likelihood inference on model parameters is dealt with, including the problem of data censoring. The statistical properties of standard agglomeration indexes in the data environment provided by our SEC model are then treated. Finally, our methodology is applied to Italian census data for both manufacturing and service industries.
Published in 2011 in: Regional Science and Urban Economics, v. 41, 3, pp. 266-280
