Spatial dynamics of public pharmaceutical expenditure

Abstract

A regression model for per capita public pharmaceutical expenditure is analyzed. The necessity of simultaneously controlling for dynamic patterns and spatial spillover in such analyses is demonstrated. In contrast to previous studies of impact of small-area variation, the present study exploits important aspects related to spatial dynamics as the effects of spatial spillover are analyzed and interpreted within a framework of spatial dynamics and spatial error-correction. It is shown that such dynamics bear important implications related to spatial convergence of a pharmaceutical market. The paper is accessible to an audience experienced with linear regression; basic exposure to spatial statistics is helpful but not strictly necessary.