Spatial distribution of the international food prices: unexpected randomness and heterogeneity

Global food prices are typically analysed in a times series framework to assess the causes of volatility and to highlight spikes, that are interpreted as a signal of food crises. We address the spatial dimension of the issue at hand by focusing on the spatial food price dispersion, at the country-scale, in the international food trade network (IFTN) for ten relevant commodities. We base our study on bilateral trade by focusing on both the "internal" variance, which indicates that an exporter sets di erent prices to different importers for the same commodity, and the "external" variance, that is a measure of market price competitiveness. We nd that spatial price dispersion is remarkable and persistent over time and that there exists a strict correlation between price spikes (in level) and peaks in spatial price variability. This entails that during price crises the market is more fragmented and a higher spatial price dispersion is found. Moreover, we implement a randomness test on the country-scale price distributions to test whether they can be replicated through a random process of extraction. It results that the actual distribution of prices of several commodities is well described by a random distribution. It follows that the process of data aggregation is not neutral because, in several cases, the information at the micro-level scale (firms' decision) is lost at the macro-scale due to the complexity of the international food trade network (IFTN). We suggest some possible economic explanations of this occurrence and we discuss the main methodological consequences.