Nonlinear effects of financial factors on fluctuations in nonferrous metals prices: A Markov-switching VAR analysis

Abstract

Because of the irreversibility of financialization in nonferrous metals markets, the security of nonferrous metals resources has shifted from a supply security mode based on “production-supply” towards an economic security mode based on “trade-finance”. Thus, the pricing mechanism of nonferrous metals has become more complex and presents features such as nonlinearity, dynamic characteristics and structure dissimilation. Based on this background, we construct a theoretical framework for analysing the effects of financial factors on fluctuations in nonferrous metals prices and employ the Markov-switching vector autoregression (MS-VAR) model to conduct a nonlinear empirical analysis based on monthly data of international copper futures prices from August 2004 to October 2016. Further, we include supply and demand factors for a comparative analysis. The results show that copper futures price fluctuations present a regime switching dynamic that can be characterized as “steep drop”, “small drop” and “steady rise” periods. During each period, financial factors can fully explain the price volatility while in different mechanism, and the “Chinese factor” has apparently been exaggerated. In addition, oil price is key to monitoring the short-term volatility risks of copper futures prices in the downward cycle. The conclusions along with the established nonlinear econometric models provide a new intellectual framework and analysis tool for explaining the financialization of commodity markets.