The appraisal of plans and projects for the development of transport infrastructures and services is a key topic both for researchers and practitioners, particularly at a national and international scale. This is criticalwith reference to freight transport, whose dynamics are rapidly and strongly influenced through the governance of transport system, leading to substantial effects on market accessibilities and on the competitiveness of national economies. Modelling approaches for representing the interaction oftransport system and the economy can be substantially classified in input-output models and economic equilibrium models. The latter face severe computational difficulties even at a national level, and their application towider contexts requires a commodity classification and a geographical aggregation not in line with the requirements for effective transport applications. The former, in spite of the acceptable compromise between computational effort and transport modelling requirements, have not been applied extensively yet at a European scale for transport oriented analyses. This study proposes the specification of an European multi-regional input-output (MRIO) model with elastic trade coefficients. Firstly, the database required for model implementation is presented and discussed. In more detail, the integration of EUROSTAT, WTO and World Bank data sources lead to the availability of input-output tables for 20 European States with a 59 commodity sectors (CPA nomenclature) disaggregation. This allowed the calculation of all inputs needed for MRIO implementation except for trade coefficient,determined through the trade flow database COMEXT provided by EUROSTAT ona yearly basis following the NC8 nomenclature. Therefore, a table of correspondence between CPA and NC8 was established, and consistently all interzonal trade coefficients were firstly calculated and then normalized with the internal trade coefficient (i.e. the percentage of acquisition from the zone itself) calculable from the i-o (input-output) table. A comparison of productions provided through the implemented MRIO model with the true data led to satisfactory results (WMAPE=22.3%). However, a GLS correction procedure of trade coefficients was performed in MatLab in order to enhancemodel goodness of fit. Then, using the preceding databases, a statisticalanalysis of trade and technical coefficients was carried out. With reference to trade coefficients, their temporal variability from 1992 to 2006 was analyzed in order to provide information for a proper trade coefficient model specification. With reference to technical coefficients, both their temporal variability (analyzed thanks to the availability of input-output tables for different years) and the deviations of national values from thestudy area averages, allowed for a classification of European Countries accordingly with their economic efficiency. Moreover, effective informationfor technical coefficient updating in MRIO applications to future scenarios were determined. Then, an elastic trade coefficient model was specified. In more detail, the systematic utility of acquiring inputs from a country of the study area is modeled as function of acquisition prices, transport costs provided by means of an Euro-Mediterranean transport supply model,and attractiveness measures provided by EUROSTAT. In more detail, the estimation was performed through an aggregate GLS procedure. The implemented MRIO model with elastic trade coefficient was then applied to significant project and plans chosen among those recently funded by the European Unionand Italian Government, aiming at quantifying their impacts on freight flows and on GDP of the Countries within the study area. For the covering abstract see ITRD E145999
Samenvatting