Capturing Inter and Intra Sectoral Productivity Spillovers in Industrial Filière: A System of Dynamic Spatial Panel Data Models

Abstract

Given the relevance of spatial and sectoral interactions in shaping the productivity performance of economic systems and filières, in this study, we suggest addressing both aspects by estimating a system of simultaneous dynamic spatial panel data models. This approach allows to consider how the production level of a specific sector is influenced by the performance of other interconnected sectors in the same area (cross-sectoral spillovers) as well as the role of productivity spillovers originating from the same or other sectors in neighbouring territories (inter and intra sectoral spatial effects). The validity of the proposed approach is confirmed through the estimation of a simultaneous system for the production process of the Italian agribusiness filière. The results indicate the existence of meaningful inter-dependences within and between sectors and territories that materialises in spillovers of different natures (input factors, internal and external features), offering valuable policy implications for the overall economic growth of the agriculture filière.