Taming Volatility, Feeding Crashes: Evidence from Algorithmic Trading in China's Agricultural Futures Markets

Algorithmic trading is commonly used in agricultural futures markets. Using high-frequency order-book data for Dalian Commodity Exchange of corn and soybean meal futures, this paper studies how algorithmic trading affects price volatility and tail risk in China's agricultural futures markets. The evidence points to a dual effect: Algorithmic trading significantly lowers conventional realized volatility; At the same time, it increases downside tail co-movement and Asymmetry between each contract and agricultural futures market. This conclusion is supported by robustness checks that exclude the DCE 7.0 transition period, reconstruct tail-risk measures using 5-minute data, and use the combined day-and-night trading session. In addition, while the Symmetrized Joe-Clayton copula baseline check confirms the robustness of the empirical tail-risk results to an alternative tail-dependence measure, we also use instrumental variables to test how algorithmic trading affects volatility. Further analysis investigates the impact of quantitative trading on volatility under different volatility regimes and during the rollover periods of dominant contracts. This study provides a theoretical basis for the “dual-effect” mechanism in agricultural financial derivatives markets and proposes regulatory suggestions for algorithmic trading in agricultural futures markets.