Blind Metrics: How Flawed Data Masks the True Scale of Myanmar’s Crisis

Antonio Graceffo, PhD As of June 2026, an estimated 3.77 million people in Myanmar are internally displaced, up from roughly 3.5 million in early 2025. Yet, the UN’s 2026 Humanitarian Needs and Response Plan estimates that 16.2 million people require life-saving humanitarian assistance, down from 19.9 million in 2025. OCHA attributes the reduction entirely to a narrower scope of analysis, stating it “by no means indicates any improvementof the humanitarian situation.” This counterintuitive finding is indicative of how flawed data collection or changes in methodology can present a false picture of the level of crisis facing people on the ground. Accuracy and consistency are crucial in assessing the humanitarian impact of the war because this data informs international policy interventions as well as potential donors and can result in funding cuts. At the same time, the junta can capitalize on the mirage of improving numbers to legitimize its rule. Roughly half of school-age children are missing basic education. Yet, according to the United Nations Development Programme, Myanmar’s Human Development Index (HDI) score held steady through the coup years, recovering to its 2019 level by 2023 despite a worsening situation on the ground. The HDI is a geometric mean of three components: life expectancy at birth, education (mean years of schooling for adults 25 and older, and expected years of schooling for children), and gross national income per capita in purchasing power parity terms. Myanmar’s score, published by UNDP, was 0.609 in 2019, the year before the coup. In 2021, the year of the coup, it was 0.605. By 2023, it had returned to 0.609, the same value as 2019. However, during the same period that the HDI was reported as unchanged and then recovering, displacement increased, as did the number of people fleeing the country. According to UNDP’s own projection, the percentage of the population estimated to be living in poverty nearly doubled, from 24.8 percent in 2017 to 46.3 percent. The number of children killed or maimed in the conflict also increased sharply, while meaningful indicators such as agricultural production, industrial activity, and the percentage of children attending school declined. One of the indicators that improved, creating the illusion of better conditions, was life expectancy at birth. Myanmar’s life expectancy at birth, as reported in World Bank data, fell from 66.612 years in 2020 to 65.553 years in 2021, the year of the coup. It then rose in each subsequent year: 66.506 in 2022, 66.889 in 2023, and 67.095 in 2024. The irony is that between 2020 and 2024, access to hospitals declined dramatically, while the likelihood of being killed in air or drone strikes increased. This figure is produced using the UN Population Division’s cohort-component demographic method, which projects forward from prior mortality trends and is updated when new vital registration or survey data become available. Myanmar does not have a national vital registration system that reliably captures deaths in contested or conflict-affected townships. Death registration completeness was already below 60 percent nationally before the coup, according to peer-reviewed assessments, with the largest gaps in rural areas with poor transportation and limited access to health facilities, the same areas most affected by the current conflict. Mean years of schooling, one of the two education inputs to the HDI, measures the completed education of the adult population aged 25 and older and has remained unchanged for years: 6.38 from 2019 through 2023. At the same time, the basic enrollment figures that actually changed show the opposite trend. Basic education enrollment in Myanmar fell from 9.7 million students in the 2019-20 academic year to 6.1 million in2025-26. The number of students sitting the matriculation exam fell from more than 900,000 in 2020 to about 200,000 in 2025. With fewer students attending school or sitting the matriculation exam, it logically follows that the mean level of education will eventually decline. However, because many of the children who have missed up to five years of schooling have not yet reached age 25, the figure has not yet been affected. Similar to other human development indicators, Myanmar’s education quality data are likely illusory. The World Bank’s Human Capital Project assigns Myanmar a Harmonized Learning Outcomes score of 424.6, on a scale where 300 represents minimal attainment and 625 advanced attainment. That same score appears for 2017, 2018, and 2020, indicating a single pre-coup data point carried forward across update cycles rather than repeated measurement, with no coup-era update since. The Academic Freedom Index, produced by the V-Dem Institute and Friedrich-Alexander-Universität Erlangen-Nürnberg, scored Myanmar at approximately 0.4 in 2020. After the February 2021 coup, the score fell to 0.02-0.03, among the three lowest scores measured globally, alongside North Korea and Eritrea, and rated “completely restricted.” Scholars at Risk reported Myanmar’s university student population shrank by 90 percent from pre-coup levels by 2024. Unlike the indicators above, this index is built from annual expert assessments rather than modeled demographic or national accounts data, and it clearly reflects the impact of the coup. The gap between illusion and reality reflects how the data is gathered. UNDP’s own November 2021 assessment of Myanmar admits, “given the currently limited data, we are not able to provide an accurate picture of the impact of COVID-19 and the takeover” on human welfare. Its Myanmar Development Observatory describes the country’s data environment as marked by “protracted crisis, conflict, and data scarcity.” A peer-reviewed study of eastern Myanmar found that “national figures do not include data from substantial portions of the country, including areas that are conflict-affected or controlled by non-state actors along Myanmar’s borders with Thailand and China.” Similarly, a separate review of Myanmar’s health sector found that national survey representation “was constrained by a 31-year gap between population censuses and by persistent active conflict in multiple states and regions.” Some data collection efforts use methodologies designed to reach areas outside government administrative control. The World Bank’s Myanmar Subnational Phone Surveys use a sample of more than 8,500 households across all 15 states and regions, covering an estimated 95 percent of the population through phone interviews. However, the methodology raises questions about the reliability of the results because much of the country has little or no cell phone service. The junta has imposed prolonged communications blackouts, cutting both mobile and internet access across nearly all resistance-controlled and contested areas. As of late 2025, roughly 131 townships remained cut off, with some in Sagaing, Chin, and Karenni states disconnected co