Supply chains and fake news: a novel input–output neural network approach for the US food sector

In this work, we focus on the following research question: “Could fake news extracted on Google be helpful in explaining the production and supply process in the food sector of the US economy?” In order to tackle this research question, we trace the supply chain of the US food sector based on Input–Output (IO) mapping. In fact, IO analysis is an essential tool for engineers, managers and decision makers across the globe, due to its direct link with the supply chain framework. The whole supply chain is perfectly captured based on the IO model. In this context, the paper studies the possibility of the production and supply processes being influenced by fakes news, captured by key phrases of Google searches, such as “collapse + US economy”. In this work, we incorporate fake news on the production process and estimate, using a suitably adjusted version of the traditional ARDL model augmented with Neural Network terms, their impact on the production process of the US food sector. We conclude that searches for fake news referring to the collapse of the US economy, could lead to significant improvement in the explanatory capability of the production process in the US food sector.