Assessing classical input output structures with trade networks: A graph theory approach

We present data structures from multiregional multisectoral trade activities from the perspective of networks. To illustrate our approach we make use of trade patterns taken from three classical input-output models. Unlike other conventional approaches by which networks statistics are evaluated, here, emphasis is given on recovering the structure architecture of interrelations in the input-output model. By self-explanatory visual outputs we display the interaction of the trading partners, the number of trade links and the density of interrelations. Connectivity and density are quantified by evaluating the node degrees. Our network approach traces the feedback loops among regions and activities. Some global structural properties are also examined. Programming in Mathematica allows for the creation of iterative schemes explaining aspects of the nature of trade and the evolution of spatial trading/production cycles in growing trading systems. Mathematica’s environment enables interactive visual schemes and infinite number of experiments.