Climate Displacement Without a Passport: East Africa's Growing Internal Refugee Crisis

A woman hangs a school uniform on the fence of a church compound in Garissa County. The uniform is too clean for the setting, washed carefully in a plastic basin, pressed with the flat of a hand against a warm wall. The school it belongs to is underwater. The home where the child who wears it slept is underwater. But the uniform is ready, because Monday might come, and when Monday comes, a child should be dressed for school. This is what displacement looks like before the cameras arrive and long after they leave: a small, fierce insistence on normalcy in a place where normalcy has been revoked.

In March 2026, heavy rains across Kenya triggered flooding that killed dozens of people and displaced more than 34,000 others, with Tana River among the worst-affected counties. Within a week, the number appeared in headlines from Nairobi to Geneva. Within a month, it will have disappeared from most of them. The floods will recede. The water will go somewhere. So will the people.

Where?

The Number That Disappears

The Internal Displacement Monitoring Centre, based in Geneva, keeps count. It is one of the few institutions in the world that does. In 2023, Kenya recorded approximately 624,000 new internal displacements from flood events alone, according to IDMC data. In 2024, the figure was even higher, driven by the devastating El Nino-amplified rains that submerged entire districts in the Rift Valley and along the Tana River. Globally, disasters caused 26.4 million new displacements in 2023, the vast majority weather-related, and sub-Saharan Africa accounted for a disproportionate share.

The word "new" matters here. IDMC counts new displacement events, not only the total number of people living in displacement at any given time. It tracks both metrics: movements within a year, and the stock of people still displaced at year's end. But a person displaced in April 2023 who has not returned home by March 2026 does not generate a new count in 2024 or 2025. They have not been displaced again. They have simply not gone home.

This distinction is not semantic. It is structural. It means that the cumulative population of people living in a state of flood-related displacement in Kenya is far larger than any single year's figure suggests. Each rainy season adds. Nothing subtracts. The IDMC estimated that at the end of 2023, roughly 684,000 Kenyans were living in internal displacement from all causes, with weather events as a growing share. But even these stock figures are approximations, because many displaced people simply stop being visible to the systems that might count them.

The 34,000 from March 2026 will join this population. Some will return. Many will not, or will return to the same place and be displaced again next year. And the number will stop appearing in headlines, which is not the same thing as the situation being resolved.

What the Word 'Refugee' Cannot Hold

There is a word for someone who flees across a national border because of persecution: refugee. The 1951 Convention Relating to the Status of Refugees, drafted in the aftermath of the Second World War, established this definition and the rights that come with it. Its 1967 Protocol removed geographical and temporal limitations, making the framework universal. A refugee has a legal identity, a claim to protection, and an institutional architecture designed to serve them. UNHCR exists for this purpose. Dadaab exists for this purpose. Kakuma exists for this purpose.

There is no equivalent word, no equivalent convention, no equivalent institution for a Kenyan farmer whose land disappears under floodwater every other year. This person has not been persecuted. They have not crossed a border. They have simply been rendered homeless by rain, by a river that has broken its banks, by soil that can no longer absorb water because the drought before the flood had baked it into something closer to stone.

The UN Guiding Principles on Internal Displacement, presented by Francis Deng in 1998, attempted to fill this gap. They defined internally displaced persons as people forced to flee their homes "in particular as a result of or in order to avoid the effects of armed conflict, situations of generalized violence, violations of human rights or natural or human-made disasters." The definition includes flood victims. But the Guiding Principles are exactly what their name suggests: principles, not law. They create no binding obligations. They establish no rights that a displaced Kenyan can invoke in a Kenyan court.

And so the language fails. A person displaced by climate within their own country is not a refugee, not a migrant, and the phrase "internally displaced person" carries none of the institutional weight that the word "refugee" does. The absence of a word is not accidental. Language is a political act, and this particular silence reflects a political choice: the international system was designed for displacement that crosses borders, not displacement that stays within them.

How many people does this silence render invisible? More than most realize. At the end of 2023, an estimated 75.9 million people worldwide were internally displaced, compared to roughly 43.4 million refugees. The internally displaced outnumber refugees by nearly two to one. They receive a fraction of the funding.

The Promise of Kampala - and Kenya's Own Path

Africa saw the gap and tried to close it. The African Union's Convention for the Protection and Assistance of Internally Displaced Persons in Africa, adopted in Kampala, Uganda in October 2009, was the first legally binding continental instrument to address internal displacement. It entered into force in December 2012. As of 2025, over 30 of the AU's 55 member states had ratified it.

Kenya is not among them. Despite hosting one of the continent's largest displaced populations, Kenya has neither signed nor ratified the Kampala Convention. What Kenya did instead was pass its own domestic legislation: the Prevention, Protection and Assistance to Internally Displaced Persons and Affected Communities Act (No. 56 of 2012), which received presidential assent on December 31, 2012. The Act borrows heavily from the Kampala Convention's principles, including the obligation to protect persons displaced by natural disasters.

The law exists. But legislation and implementation are different verbs, and the distance between them is where displaced Kenyans live. The IDP Act established rights to protection, assistance, and durable solutions. Yet implementation has been widely criticized as inadequate. Regulations that would give the Act operational teeth have been slow to materialize. Institutional capacity at the county level, where displacement actually happens, remains thin.

What does this mean in practice? It means that when tens of thousands of people are displaced from flood zones in Tana River and other counties, there is a statutory framework that defines their rights to shelter, to resettlement, to assistance, but the machinery to deliver those rights barely functions. There are ad hoc responses: the Kenya Red Cross mobilizes, OCHA issues flash appeals, county governments open schools as shelters. But ad hoc is not the same as systematic, and response is not the same as the rights the Act was meant to guarantee.

The Kampala Convention imagined a continent where internal displacement would be governed by law rather than by charity. Kenya wrote its own version of that law. Thirteen years later, Kenya's flood-displaced still live in the gap between what the law promises and what it delivers.

The Trajectory of Displacement

Where do tens of thousands of displaced people go?

The answer, in Kenya as in most of East Africa, is rarely a camp. The image of orderly rows of white UNHCR tents belongs to the cross-border refugee infrastructure. Flood-displaced Kenyans mostly go somewhere else entirely: to relatives, to churches, to mosques, to school buildings that then cannot function as schools. IOM's Displacement Tracking Matrix has documented this pattern repeatedly. In the 2024 floods, the majority of displaced people in Kenya's Rift Valley and Tana River regions were absorbed by host communities, relatives, and religious institutions rather than by formal emergency shelters.

This sounds like resilience, and in one sense it is. But it is also a transfer of cost. The relatives who take in a family of six do so at the expense of their own food stocks, their own sleeping space, their own children's access to a room with enough floor. The church that opens its doors as a shelter cannot hold services. The school that becomes a dormitory cannot hold classes. UNICEF estimated that the 2024 floods disrupted education for approximately two million learners in Kenya, not because schools were all destroyed but because many were occupied by displaced families or rendered inaccessible by floodwaters.

And then what? Return, mostly. Return because there is nowhere else. Kenya faces a housing deficit of over two million units, concentrated in the low-income segment where flood displacement hits hardest. There is no surplus of safe, affordable land waiting for resettlement. The government has periodically announced relocation programs for people living in flood-prone riparian zones, but execution has been limited by funding, land availability, and political resistance.

So people go back. They rebuild with what they have, which is usually mud, timber, and corrugated iron. They know the floods will come again. The Kenya Meteorological Department tells them so, and they know it anyway because they have been through this before. IDMC has documented that a significant portion of displacement in East Africa is repeated, the same communities displaced by the same hazard in consecutive years.

This is not a cycle. A cycle implies regularity. This is something more disorienting: a pattern that everyone can see but nobody can break.

The Funding Gap Between Borders

Follow the money and you will find the architecture of what matters.

UNHCR's Kenya operation is substantial. Dadaab camp complex, near the Somali border, hosts approximately 430,000 registered refugees and asylum seekers. Kakuma camp in Turkana County, together with the adjacent Kalobeyei settlement, holds roughly 310,000 more. These populations have dedicated UNHCR offices, funding streams, school systems, health clinics, and legal representation. They are visible to the international system because they crossed a border, and crossing a border activated a legal framework, and a legal framework activated funding.

Internally displaced Kenyans have no equivalent infrastructure. When OCHA issues a flash appeal for Kenya flood response, the requested amounts are typically in the tens of millions of dollars and are routinely underfunded. The 2024 flood response plan requested approximately $23.1 million and received a fraction of what was asked. Compare this to the hundreds of millions allocated annually to the cross-border refugee operation in the same country.

This is not because donors do not care. It is because the international humanitarian system was built for one kind of displacement and has not been rebuilt for another. The UN's Cluster system activates for emergencies but does not maintain a permanent presence for internal climate displacement. There is no "camp" for a Kenyan displaced by Kenyan rain on Kenyan soil. The sovereignty principle - the idea that internal displacement is a domestic matter - creates both a legal and a practical barrier to international engagement.

The result is a funding disparity that compounds vulnerability. A Somali refugee in Dadaab has access, however imperfect, to international protection and resources. A Kenyan farmer displaced by floods 200 kilometers south of Dadaab has access to the generosity of relatives and the intermittent attention of an overstretched county government.

Both are displaced. One is seen.

The Invisible Population Grows

Here is the pattern that no headline captures: climate displacement in East Africa is cumulative. Every flood season adds to the population of people who have lost homes, farmland, livestock, or livelihoods. Every drought does the same. The Horn of Africa's devastating drought from 2020 to 2023, the longest and most severe in recorded history, displaced millions across Somalia, Ethiopia, and Kenya. Then the floods came, displacing many of the same people from the same regions that had just endured three years without adequate rain.

IDMC reported that in 2023, sub-Saharan Africa recorded approximately 6.3 million new disaster-related internal displacements, up from 1.1 million in 2009. East Africa accounted for a substantial share, with Somalia, Ethiopia, and Kenya among the most affected countries on the continent. These numbers will rise. Climate models project that East Africa will face shifting rainfall patterns and longer dry spells as global temperatures increase, a trajectory that will amplify displacement, though the precise regional dynamics remain subject to scientific debate.

But the projections exist in the language of modeling. The displacement exists in the language of a family sleeping in a school. And between those two languages, there is a gap wider than any convention has bridged.

There is also a political dimension that makes the problem harder to name. Climate-displaced people within their own country are also voters, also citizens, also potential constituents. But they are a diffuse constituency, scattered across dozens of counties, belonging to no single bloc, organized around no single demand. They do not march. They do not have a lobby. They rebuild their houses and wait. This is why governments can acknowledge the problem in principle and neglect it in practice without meaningful political consequence.

Six Months After the Water

Ask this question about the tens of thousands displaced in March 2026: where will they be in September?

If the pattern of previous years holds, most will have returned to the same areas. Some will have rebuilt on the same plots, in the same floodplains, using the same materials. A smaller number will have moved to Nairobi or Mombasa or Kisumu, joining the urban population without joining any formal registry. A few will have received assistance from the Kenya Red Cross, from church organizations, from relatives who can afford to help. Almost none will have received formal government resettlement to safe ground.

Some of their children will have lost a school term. Some will not return to school at all. The disruption, like the displacement, accumulates.

The Kenya government's National Disaster Management Authority coordinates response. The county governments of Tana River and other affected areas deploy what resources they have. But disaster response and displacement governance are not the same thing. Responding to a flood is a matter of boats and blankets and boreholes. Governing displacement is a matter of law, land, housing, and long-term investment. Kenya has the former. It has the law for the latter, on paper. What it does not yet have is the implementation that would make that law real.

And so the question does not resolve. Where will the displaced be in September? They will be wherever they can manage to be. Which is not the same as where they should be, or where they have a right to be, or where someone has planned for them to be.

They will not have crossed a border. They will not have become refugees. They will simply have become people for whom the word "home" has become conditional, a place that exists only until the next rain.

A school uniform dries on a fence. Monday might come.