The international refugee system has collapsed

Displacement is no longer an interruption between two stable lives. It has become a condition of its own and the institutions built to address it were never designed for that reality.

When the architects of the postwar international order drafted the 1951 Refugee Convention, they had a specific crisis in mind. Europe was full of displaced people. Camps were filled with those uprooted by a war that had, mercifully, ended. The assumption embedded in every article of that convention was that displacement was an interruption, a temporary rupture that good institutions could bridge. Build the camps. Mobilize the aid. Wait for conditions to stabilize. Send the people home.

That model has not merely aged poorly. It has collapsed.

The numbers tell a story the international community has been reluctant to directly confront. According to the UN Refugee Agency, the number of forcibly displaced people worldwide passed 117 million in 2023, more than double the figure recorded a decade earlier. That doubling did not happen because the world suddenly became twice as violent. It happened because old crises are not resolving, while new ones accumulate on top of them. The pipeline flows in. Almost nothing flows out.

The clearest measure of this failure is duration. The average length of a major refugee situation now exceeds 20 years

Dr. Azeem Ibrahim

The clearest measure of this failure is duration. The average length of a major refugee situation now exceeds 20 years. Two decades. A child born in a camp at the moment of displacement grows to adulthood, forms a family and raises children of their own, all within a system designed to be temporary. The Afghan displacement crisis has, in various forms, persisted for more than 40 years. The Somali crisis for 30. The Palestinians, the original test case for whether the international community could manage protracted displacement, have been displaced for more than 75 years and counting. The emergency, in each case, became the permanent condition. The institutions responded by pretending otherwise.

This is not a resource problem, though resources are chronically inadequate. It is a structural one. The UN Refugee Agency’s mandate is organized around three durable solutions: voluntary repatriation, local integration in the country of first asylum and third-country resettlement. In theory, one of these three pathways resolves every refugee situation. In practice, all three are failing simultaneously.

Repatriation, historically the preferred and most common outcome, requires a safe, voluntary and dignified return. In the current landscape of protracted civil wars and authoritarian consolidation, those conditions rarely materialize.

Local integration is politically toxic in most host countries, which are overwhelmingly low and middle-income states already absorbing populations that dwarf anything the wealthy world accepts. Turkiye hosts more than 3.2 million registered Syrian refugees. Pakistan and Iran together host at least 4 million Afghans. Bangladesh carries nearly a million Rohingya in what is arguably the world’s most densely populated refugee settlement. These are not short-term burdens being temporarily shouldered. They are permanent demographic realities that host governments were never asked to formally accept and receive no proportionate support to manage.

Third-country resettlement, meanwhile, operates at a scale that borders on the symbolic. In 2022, the US, historically the world’s largest resettlement country, admitted fewer than 25,000 refugees from a global population requiring resettlement that the UN estimated at more than 1.5 million. The entire global resettlement system that year processed about 114,000 people. The math does not work and everyone involved knows it does not work.

Host states carrying disproportionate burdens without proportionate support eventually reach their political limits

Dr. Azeem Ibrahim

The Rohingya crisis illustrates what this failure looks like in human terms. Nearly a million people have now spent eight years in the camps of Cox’s Bazar, with repatriation blocked by conditions in Myanmar that have deteriorated rather than improved since the 2017 expulsions. Local integration is politically unavailable in Bangladesh. Resettlement proceeds at a pace that could absorb the population over several centuries.

A generation of children has grown up with access to informal education at best, no legal right to work and no documentation that would allow participation in any formal economic life. The system’s three solutions are all inoperative. What remains is indefinite management dressed up as temporary accommodation.

The geopolitical consequences of this management approach are becoming impossible to ignore. Host states carrying disproportionate burdens without proportionate support eventually reach their political limits. When they do, the results are not orderly. They are dangerous. Secondary displacement, sea crossings, stateless populations that become recruitment pools for criminal networks and armed groups — these are not hypothetical risks, they are the documented outcomes of a system that offers people no legal horizon.

What is required is not more humanitarian pledges. The pledge-making machinery runs without interruption and without discernible effect. What is required is for the meaning of refugee protection to be fundamentally reconceived in cases where there is no foreseeable prospect of return.

Displaced people need the right to work, to move and to build economic lives in the countries that host them. Host countries need binding, enforceable burden-sharing arrangements rather than voluntary commitments that wealthy states consistently fail to honor. Development finance, not just emergency relief, needs to flow to the communities — both refugee and host — that are absorbing the cost of crises they did not cause.

The 1951 convention was a serious document written by serious people for a world in which displacement was an exception. They could not have imagined, or perhaps chose not to imagine, a world in which it was a permanent feature of the international landscape. That world is here. The gap between the institutions we have and the problem we face grows wider every year. Filling it is not a humanitarian aspiration. It is a geopolitical necessity.

The emergency was always going to become permanent. The only question was whether the system would adapt before the cost became unbearable. The answer, so far, is no.

• Dr. Azeem Ibrahim is the director of special initiatives at the New Lines Institute for Strategy and Policy in Washington. X: @AzeemIbrahim