The causal links between adverse shocks (economic, climatic, conflict-related, health-related, etc.) and food insecurity is a core issue in development economics. Recently, the nexus between these dynamics (not least the recent pandemic shock) and positioning of smallholder households along the market chains has become an important research question, with relevant implications for policymaking. In particular, research is still scarce and ambiguous on whether closer connections to major markets can lower these households¿ vulnerability and avoid the long-lasting effects of shocks on their food security. This research project aims to shed light on these critical issues by using new data and new methods. Specifically, i) we combine new cross-country household data - 13 countries from three continents, for a total of almost 20,000 original household observations - from the micro-level impact assessment database of the International Fund for Agricultural Development (IFAD) with data on travel distances to major markets and population centers from a recently developed Global Map of Accessibility; ii) we apply a mix of causal inference (instrumental variable regressions) and machine learning methodologies (the Least Absolute Shrinkage and Selection Operator LASSO) to explore the causal links highlighted above. The research has relevant implications on policymaking since it provides broader empirical evidence on the role of market chains on households' vulnerability.
