Predictability in Commodity Markets: Evidence from More Than a Century

Abstract

Using more than 140 years of data, we comprehensively analyze the predictive power of a broad set of business cycle variables for risk and return in commodity spot markets. We find that industrial production growth and inflation are the strongest predictors for future commodity returns. Several further variables help predict future commodity volatilities. The introduction of derivatives generally reduces the predictability in the most active commodity markets but increases the predictability in others. Thus, derivatives likely make markets more efficient, but also attract most of the price discovery activity. Commodity spot volatilities generally rise after futures introduction.

Keywords: Commodities, Return Predictability, Derivatives Introduction, Business Cycle, Volatility Predictability

JEL Classification: G10, G11, G17