Stochastic trade networks

Abstract

This paper develops a simple network model to describe the dynamics of international trade flows. The result is achieved by means of the combination of two mechanisms of proportional growth: the first determines the formation of trade links (intensive margin of trade) and the second governs trade intensity (extensive margin of trade). With respect to recent network models of trade, our approach better fits the numbers of zeros and the degree distribution at different levels of aggregation from firms to national economies. Furthermore, we match a number of additional empirical regularities such as the high concentration of trade with respect to both products and destinations, and the relationship between node degree and node strength. These findings suggest that stylized facts are strongly interconnected across different levels of aggregation, so that a unifying explanation is called for. Our network model can be employed as a useful stochastic benchmark, similar to a generalized gravity model for both trade margins.