The Displacement Equation: What Refugee Flow Data Reveals About Conflict, Climate, and the New Geography of Flight

• In 2025, for the first time on record, conflict overtook climate as the world’s leading driver of forced displacement — but the two forces are increasingly indistinguishable in the world’s most fragile states, including across South Asia.
• At the end of 2025, 117.8 million people were forcibly displaced worldwide — refugees, asylum-seekers, and the internally displaced combined.
• The cleanest way to think about the conflict/climate relationship is not as two competing causes of displacement but as a feedback loop.
• Combined with the 15 percent of countries where displacement data itself is degrading, the practical result is that policymakers may soon be flying blind precisely as the conflict-climate feedback loop intensifies.

In 2025, for the first time on record, conflict overtook climate as the world’s leading driver of forced displacement — but the two forces are increasingly indistinguishable in the world’s most fragile states, including across South Asia.

Every year, the United Nations High Commissioner for Refugees and the Internal Displacement Monitoring Centre publish what amounts to a global stress test — a count of everyone forced from home by war, persecution, or the weather. Read side by side, the two datasets do something no single number can: they show when conflict and climate act as separate shocks, and when they compound into a single, self-reinforcing crisis. 2025 was the year the compounding became impossible to ignore.

At the end of 2025, 117.8 million people were forcibly displaced worldwide — refugees, asylum-seekers, and the internally displaced combined. That is roughly the population of Bangladesh, uprooted. Of these, 41.6 million had crossed an international border as refugees, while 68.7 million remained internally displaced by conflict and violence inside their own countries. A further 9 million were awaiting asylum decisions. The figure represents a rare year-on-year dip after thirteen consecutive years of increase — driven not by fewer people fleeing, but by a wave of returns to the Democratic Republic of Congo, Syria, and Ukraine, many under conditions humanitarian agencies describe as fragile rather than durable.

Two Drivers, One Crossover Year

For most of the past decade, disasters — cyclones, floods, droughts — displaced more people within their own countries each year than conflict did. That pattern broke in 2025. IDMC’s most recent Global Report on Internal Displacement recorded 32.3 million new conflict-related internal displacements, a 60 percent jump over 2024 and, for the first time since the data series began tracking this reversal in 2019, more than the 29.9 million people newly displaced by disasters.

The reversal is not a sign that climate risk receded — disaster displacement in 2025 was still 13 percent above the past decade’s average. It reflects, instead, an intensification of war. The Democratic Republic of Congo alone recorded close to 10 million new conflict displacements, a national record, as fighting in the east escalated. The Middle East and North Africa accounted for 40 percent of all conflict-driven movement globally, with Sudan’s civil war continuing to generate the largest internal displacement crisis on the planet — 9.1 million people, more than the population of Switzerland, still unable to return home.

Where the world’s displaced are concentrated

Displacement is not evenly distributed — it clusters. Just five countries — Sudan, Colombia, Syria, Yemen, and Afghanistan — account for 46 percent of the world’s internally displaced people. And 68 percent of all refugees are hosted not by wealthy states but by low- and middle-income countries, usually immediate neighbors: 65 percent of refugees remain in a country bordering their own. The burden of displacement, in other words, falls hardest on states that are themselves fragile, poor, or climate-exposed — a dynamic that compounds rather than resolves the original crisis.

Where Climate and Conflict Stop Being Separate Variables

The cleanest way to think about the conflict/climate relationship is not as two competing causes of displacement but as a feedback loop. A flood or drought degrades livelihoods and food security; that pressure feeds into existing political or ethnic fault lines; the resulting instability then makes it harder to invest in the flood defenses, irrigation, or early-warning systems that would have reduced the original climate shock. Sudan, Somalia, Yemen, and Afghanistan — four of the states with the largest displaced populations — all sit at this intersection: active conflict layered onto acute climate vulnerability, with humanitarian access and data collection degraded by the fighting itself. IDMC’s 2025 report noted that data availability fell in 15 percent of monitored countries — three times the share the year before — precisely because conflict makes climate risk harder to even measure.

The South Asia Case

South Asia illustrates the convergence acutely. Disaster-driven internal displacement in the region nearly tripled between 2023 and 2024, reaching 9.2 million — the second-highest regional figure recorded in over a decade, after a temporary El Niño-linked dip the year before. Pakistan, ranked the world’s most climate-vulnerable country by the 2025 Germanwatch Climate Risk Index, remains the starkest example: the 2022 monsoon floods affected 33 million people, displaced roughly 8 million, and caused an estimated $30 billion in damage. Pakistan was hit again in 2025, when monsoon flooding between late June and mid-September killed 946 people, including 255 children, according to UNICEF.

Bangladesh’s disaster-displacement figures have climbed for four consecutive years, reaching 2.4 million in 2024 — up roughly 600,000 from 2023 — placing it fifth globally for disaster-related internal displacement. India, meanwhile, experiences nearly every variant of the crisis within its own borders: Cyclone Fengal’s flooding in Tamil Nadu and Puducherry in late 2024, chronic drought-linked distress across the Deccan interior, and long-term coastal erosion pressure in the Sundarbans, alongside its role as the region’s principal destination for cross-border movement.

Why this matters beyond humanitarian accounting

• Border and internal security implications. Analysts at the Wilson Center have flagged that climate-driven rural-to-urban migration in Bangladesh, India, and Pakistan strains already overcrowded cities in ways that can heighten recruitment risk for extremist groups and sharpen communal tensions where displaced populations cross religious or ethnic lines.

• Fragmented regional response. With SAARC largely non-functional, climate displacement in South Asia is being managed almost entirely through unilateral national frameworks — Bangladesh has a National Strategy on Disaster and Climate-Induced Displacement, but no shared regional mechanism exists for anticipatory planning, data-sharing, or burden distribution.

• No legal category. “Climate refugee” carries no standing in international law, leaving an estimated 1.2 billion people projected to face climate-related displacement by 2050 without a defined protection framework — a gap that domestic policy will have to fill in the absence of international consensus.

• India’s dual exposure. India sits simultaneously as a major generator of internal climate displacement and the default receiving state for cross-border movement from Bangladesh, Nepal, and potentially a submerging Maldives — a combination that makes internal migration data and border management policy two sides of the same planning problem.

The Data Gap Ahead

The most consequential trend in this year’s figures may not be a displacement number at all, but a funding one: UNHCR operated in 2025 with 24 percent less funding than in 2024, assisting 16 percent fewer people even as needs stayed roughly flat. Only 37 percent of total humanitarian needs were funded, down from 48 percent the year before — meaning the agency is now responding to a crisis more than twice the size it managed a decade ago, with resources last seen when the crisis was half that size. Combined with the 15 percent of countries where displacement data itself is degrading, the practical result is that policymakers may soon be flying blind precisely as the conflict-climate feedback loop intensifies.

_References:

• _{UNHCR, Global Trends: Forced Displacement in 2025 — topline figures on refugees, IDPs, asylum-seekers, and returns.}
• _{UNHCR, Figures at a Glance — breakdown of displacement categories and hosting patterns.}
• _{UNHCR, Global Report 2025 — funding and operational response data.}
• _{Norwegian Refugee Council / IDMC, Conflict and Violence Become the Leading Driver of Internal Displacements (GRID 2026) — the 2025 conflict/disaster crossover.}
• _{IDMC, Global Report on Internal Displacement 2025 (GRID 2025, covering 2024) — regional breakdowns including South Asia.}
• _{PRISM Deep Dives, Internal Displacement in 2025 — mapped country-level 2025 flow data.}
• _{The Climate Watch, Disasters Displace 600,000 People in Bangladesh in Just One Year.}
• _{South Asia Monitor, Climate Refugees Are Rising: Is South Asia Prepared?}
• _{Wilson Center, Climate-Induced Displacement: South Asia’s Clear and Present Danger — security-dimension analysis.}
• _{Australian Institute of International Affairs, Why South Asia’s Biggest Threat Is Climate Change — Pakistan risk-index and rainfall data.}

Divyanka Tandon

Divyanka Tandon holds an M.Tech in Data Analytics from BITS Pilani. With a strong foundation in technology and data interpretation, her work focuses on geopolitical risk analysis and writing articles that make sense of global and national data, trends, and their underlying causes. Views expressed are the author’s own.