Dynamic Interaction Networks and Frequency Domain Features of Speculation and Volatility in US Energy Futures Markets

Abstract

This paper investigates the interplay between speculative and price volatility in the energy futures markets over various cycles, utilizing wavelet coherence and a double-layer network approach. Contrary to conventional wisdom, we find that long-term price volatility in individual futures markets, driven by extreme events, persistently leads to increased speculative trading, partly associated with increased hedging and risk management activity. The connectedness of the two-layer network system is dominated by speculation and volatility spillovers in the short and long term, respectively. The cross-layer spillover effects between price volatility and speculation are more pronounced in the long term. The direct and network effects of speculation reinterpret the interaction patterns among various futures markets. Specifically, the crude oil market, as a net receiver of spillover effects, exhibits an impact of speculation on price volatility driven primarily by network effects. However, the natural gas market is dominated by the direct effects of speculation.