Testing for the law of one price and identifying price-leading markets: An application to corn markets in Benin

Abstract

A data-determined method is proposed to test for the law of one price as a long-run equilibrium condition and to identify which markets in a network of spatially dispersed commodity markets quote the reference price. The method consists of supplementing Johansen's FIML cointegration procedure with a permanent-transitory decomposition and rules of inference in linear time-series models with unit roots. As an example, we apply our method to prices of six corn markets in Benin. We find that the law of one price holds in the long-run. It appears that two rural markets quote the reference price: their prices adjust fastest towards the permanent change induced by the common stochastic trend.