This article examines the determinants of food security in the WAEMU countries. First, Principal Component Analysis (PCA) is used to compute a composite food security index in order to take into account the four dimensions of food security, namely availability, accessibility, stability, and utilization. The Pooled Mean Group (PMG) method is then applied to estimate an ARDL model. The results indicate the existence of a long-run relationship between food security and the explanatory variables. A disequilibrium of 38.7% observed in food security in the previous year is corrected in the following year. Moreover, the short-run determinants of food security include cereal yields, GDP per capita, and population growth. In the long run, the same variables, along with arable land area, significantly influence food security. This study makes an original contribution to the understanding of food security in WAEMU countries through a dual methodological approach. It proposes the construction of a composite food security index based on several key dimensions, in line with FAO recommendations. The research employs the ARDL model to examine the short- and long-run determinants of food security. The combination of these two approaches gives the study significant scientific added value. Methodologically, it enriches the literature on composite measurement of food security in the African context. Empirically, it provides WAEMU policymakers with a more reliable tool for assessment and monitoring, as well as a dynamic analysis of economic factors likely to sustainably improve food security.

Keywords: food security, WAEMU countries, principal component analysis, ARDL model, economic determinants.