The role of global trade agreements in global food supply chain and food security

Global trade agreements (GTAs) have transformed global food supply chains (GFSCs) by facilitating the international flow of food products and reducing barriers such as tariffs and quotas. However, uncertainties in GTAs lead to several challenges, including panic buying and fluctuating demand, transportation delays, expiration of perishable food products, and increased operational costs, all of which threaten the stability and resilience of food supply chains (FSCs). For instance, the termination of the North American Free Trade Agreement (NAFTA) can result in tariff increases of up to 150% in some products, leading to significant cost surges in cross-border food trade and disrupting supply continuity (Swanson & Granville, 2017). This study investigates the impact of such disruptions and the subsequent effects of panic buying on GFSCs and food security. To address the uncertainty in demand, this research develops a two-stage stochastic programming model to optimize a global multi-product FSC during disruptions, aiming to enhance profitability and resilience. Various products with different shelf lives and transportation modes are incorporated into the model to meet logistical requirements. The model integrates supply and distribution decision-making and considers multiple disruption scenarios. The Monte Carlo simulation is employed to simulate the demand parameter by generating a large number of scenarios based on historical data. The K-means clustering method is then applied to reduce the number of generated scenarios. Moreover, the sample average approximation method is employed to mitigate the large-scale nature of the problem and reduce the computational time. A real-world case study demonstrates the model’s applicability, particularly highlighting the impact of a potential NAFTA termination on the FSC between Canada and the United States. The results demonstrate that terminating NAFTA can reduce profit by up to 41.67% due to unmet demand, leading to food shortages first in Canada and then in the United States, with some products being less resilient than others. Finally, the numerical experiments confirm the model’s effectiveness across various problem sizes.