URBANA, Ill. (Chambana Today) — International humanitarian aid organizations rely heavily on the Integrated Food Security Phase Classification (IPC) system to monitor global food insecurity. This system is a crucial tool for directing assistance to areas most in need, but a new study from the University of Illinois Urbana-Champaign (UIUC) reveals that the IPC and similar analyses may be consistently underestimating the scale of hunger.
Published in Nature Food, the study—led by UIUC Professor Hope Michelson—finds that food insecurity data may miss up to one in five people who are at urgent risk of hunger. The findings suggest that global estimates significantly underestimate the severity of food crises, which could delay or prevent timely aid interventions.
The IPC, a collaboration of 21 organizations established in 2004, has long been considered a reliable system for assessing food security. It uses subnational data from countries facing extreme hunger to classify regions from “minimal” (Phase 1) to “famine” (Phase 5). These analyses help allocate billions in aid each year. However, the new research highlights a systemic flaw in the system.
“We were surprised to find that the IPC tends to underreport food insecurity, especially when faced with conflicting data on the ground,” said Michelson, a professor in the Department of Agricultural and Consumer Economics at UIUC. “The result is that many people in need may not be counted at all.”
In 2023, approximately 765 million people worldwide lacked sufficient food to meet their basic needs, and nearly a third of them faced acute hunger. Accurate identification of hunger crises is essential for directing international aid. Yet, according to the new study, many areas facing urgent need are classified just below the threshold for crisis (Phase 3), resulting in undercounting.
Michelson and her team analyzed nearly 10,000 food security assessments from 33 countries between 2017 and 2023. They compared these findings with IPC’s classifications and found that their estimates for people in crisis (Phase 3 or higher) were much higher—293 million, compared to IPC’s estimate of 227 million. This discrepancy suggests that over 66 million people—about one in five facing acute hunger—are not being counted.
The team also discovered that when data from various sources conflicted, IPC analysts tended to err on the side of caution, placing affected areas just below the critical threshold. This approach, while prudent, may lead to large groups of people going unnoticed by humanitarian responses.
“There’s a tendency to downgrade the severity of crises when data is inconsistent, even though the true situation on the ground might be worse,” said co-author Chungmann Kim, a doctoral student in Agricultural and Consumer Economics at UIUC.
While the study suggests room for improvement, it also affirms the importance of the IPC as a critical tool for global food security analysis. The researchers propose incorporating more advanced data collection techniques, such as machine learning, to enhance the accuracy of future analyses.
“There’s already a significant gap between the needs identified and the resources available to meet them,” Michelson added. “By underestimating the problem, we’re inadvertently depriving people of the help they need. This highlights the urgency of increasing both the scale of humanitarian aid and the resources dedicated to addressing food insecurity worldwide.”
The study also underscores the importance of continuous improvements in global food security monitoring. The authors recommend that automated systems and advanced algorithms be used to complement—but not replace—the existing consensus-based approach used by the IPC.