The state of food insecurity measurement: A mix of methods, and a mix of messages

Headey, Derek D. 2024. The state of food insecurity measurement: A mix of methods, and a mix of messages. IFPRI Discussion Paper 2323. Washington, DC: International Food Policy Research Institute. https://hdl.handle.net/10568/169686

Robust food insecurity indicators are needed for monitoring development targets, humanitarian advocacy efforts, and rationally allocating foreign aid. Longstanding dissatisfaction with the FAO’s undernourishment indicator prompted the development of new metrics in recent decades, including the FAO’s Food Insecurity Experience Scale (FIES) and the unaffordability of healthy diets. However, no previous research has assessed whether food insecurity and poverty indicators are in broad agreement on which countries are insecure/poor, and whether global food insecurity is rising or falling. Unfortunately, this new mix of methods produces mixed messages. At the country level, FIES severe food insecurity is often higher in Latin America and the Caribbean than in Niger and other extremely poor African countries. On global trends, the FAO reports increasing undernourishment and FIES food insecurity over 2014-2022, whereas the World Bank reports monetary poverty declining and healthy diets becoming more affordable. Moreover, trends in FAO food security indicators are not statistically explained by hypothesized factors cited in FAO reports, such as conflict or climate change, and increases in the FAO’s calorie consumption inequality metric are inconsistent with declining income inequality reported by the World Bank. We provide four concrete suggestions to improve food security measurement and monitoring: (1) the FAO should cease modelling undernourishment; (2) new independent studies should re-evaluate the FIES and test new metrics; (3) international agencies should implement coordinated, high-frequency, multi-purpose, open-access surveys; and (4) researchers should further improve the “nowcasting” of poverty and food insecurity for data-scarce crisis contexts.