This study investigates the asymmetric and nonlinear impacts of interacting uncertainty factors − geopolitical risk (GPR), financial stress and Twitter-based Economic Policy Uncertainty (TEU) − alongside green bond market dynamics on the price behavior of ten major agricultural and dairy commodities.
The empirical framework integrates a Nonlinear Autoregressive Distributed Lag (NARDL) model to capture long-run structural asymmetries with a Cross-Quantilogram (CQ) approach to identify state-dependent directional dependence. Using monthly data from January 2013 to January 2023, the analysis evaluates transmission mechanisms across deciles (0.1–0.9), capturing dynamics from bearish (0.1–0.3) to bullish (0.7–0.9) market regimes. A pre- and post-COVID-19 subsample analysis (2013–2019 vs 2020–2023) is also conducted to assess structural shifts.
The NARDL estimations reveal significant long-run asymmetries, indicating that financial stress and green bond dynamics exert persistent structural effects, contributing to an environmental risk premium in agricultural prices. The subsample analysis suggests that transmission intensity and adjustment dynamics become more pronounced during crisis periods. Furthermore, the CQ analysis uncovers strong regime-dependent dynamics: under bullish conditions (upper deciles, τ ∈ [0.7, 0.9]), negative directional dependence dominates, while under bearish conditions (lower deciles, τ ∈ [0.1, 0.3]), positive dependence prevails, consistent with the supply disruption channel.
This study contributes a unified nonlinear framework that conceptualizes agricultural price formation through four transmission channels. By capturing structural asymmetries, crisis-induced shifts, and decile-dependent predictability, it provides novel insights for quantile-contingent portfolio strategies and for policymakers aiming to enhance food security and market stability.
