Skip to main content
Article

Energy market volatility and food price stability: The role of oil and dollar shocks in coffee and wheat markets

Chellai FatihUniversité Ferhat Abbas, AlgeriaSalim Bourchid AbdelkaderDepartment of Business Administration, College of Business, King Khalid University, Saudi ArabiaPardaev JamshidDepartment of Finance and Tourism, Termez University of Economics and Service, Termez, UzbekistanKamel Si MohammedUniversité de Lorraine, CEREFIGE, Metz, F-57000, France
2026en
ABI

Abstract

This study investigates the complex interdependencies between energy markets, currency effects (USD Index), and food commodities (coffee and wheat) over a 25-year daily horizon (1999–2024). Employing an integrated multi-method framework combining structural break detection (PELT algorithm), artificial neural network, and Prophet forecasting models, continuous wavelet coherence, and cross-quantilogram analysis, we find that oil price shocks transmit to agricultural commodities through multiple channels: direct transportation costs and currency-mediated mechanisms. Critically, we find that the USD Index dominates short-term co-movements (1–30 days) across all commodities, explaining over 50% of daily correlations, while fundamental linkages dominate at longer horizons (60+ days). Wavelet coherence analysis reveals frequency-dependent transmission patterns: oil-wheat dependencies exhibit stronger coherence at 1–3 month scales, while oil-coffee transmission shows delayed responses at 3–6 month scales, confirming crop-specific supply dynamics. Cross-quantilogram results show asymmetric tail dependence: left-tail (crash) co-movements are 3–4 times stronger than right-tail dependencies, indicating that crises spread contagiously while booms remain commodity-specific. Our findings have significant implications for agricultural producers facing energy cost uncertainty, commodity traders designing cross-market strategies, policymakers managing food security risks, and portfolio managers seeking regime-dependent hedging solutions.

Identifiers

Citations and references

Cited by 050 references