Price dynamics, LOP and quantile regressions

The Law of One Price is a dated but still a puzzling economic concept. Studies that found violations of the Law are frequent and numerous, although scholars have pointed that failures of the Law are likely to be due to lack of informative datasets. In addition, for storable commodities, the possible interactions of spatial and temporal arbitrage may hide the implications of the Law, invalidating the conclusions of the studies. Based on a simplified two-market model of spatio-temporal arbitrage, I review the implications of the Law of One Price and test for them with a rich dataset of weekly prices of storable commodities, and information on transaction costs, trade and storage. I conclude that most of the statements implied by the Law of One Price are indeed not empirically violated.