Statistical properties of the international seed trade networks for rice and maize

Abstract

International seed trade networks (iSTNs) are complex networks formed by the trade of seeds between economies. We investigate the statistical properties of the international trade networks for maize seeds and rice seeds from the viewpoint of temporal directed and weighted networks. We find that most properties, including node and edge numbers, average in- and out-degrees, total and average trade values, network density and clustering coefficient, showed an increasing trend, which indicates that the increasing globalization of seed trade has led to more and more complete and dense trade networks. We also find that links with larger trade values have greater stability. In addition, we observe positive correlations between in-degree and out-degree, link reciprocity and in- and out-degrees, link reciprocity and in- and out-strengths and clustering coefficient and in- and out-degrees. In contrast, there are negative correlations between clustering coefficient and in- and out-strengths. Furthermore, we find that both networks are assortatively mixed in most years, which indicates that large exporters and importers are more likely to trade with other large exporters and importers. The structural properties of the two iSTNs of maize and rice share many qualitative similarities, but exhibit quantitative differences.