Kakuma, Kenya — "We are close enough to communities to be useful and far enough from power to be safe."
Every few months at Solidarity Initiative for Refugees (SIR), the organisation I founded and lead in Kakuma refugee camp, we go through a ritual that has nothing to do with our mission. We say goodbye to someone talented.
A programme officer who spent two years building community trust and digital literacy curricula. A finance coordinator who learned to navigate compliance requirements no one designed with us in mind. A communications lead who could write a funding report and facilitate a community dialogue on the same day.
They don't leave because they stop caring, they leave because an international NGO or UN agency offered them three times the salary, formal contracts, and something refugee-led organisations almost never can: job security beyond the next project cycle.
Keep up with the latest headlines on WhatsApp | LinkedIn
This is the story the humanitarian sector rarely tells about refugee-led organisations (RLOs). Not the story of what we lack - funding - but the story of what we are systematically prevented from building: the institutional depth to grow, lead, and last.
Closest to the community. Farthest from the money
Everyone in the humanitarian sector says the same thing: RLOs are closest to the community. They say it in conference panels, they write it in strategy documents, and they repeat it in donor reports. They are not wrong, but then they keep us farthest from quality funding.
This is not a contradiction they have failed to notice, it's a contradiction they have chosen to maintain. Because our proximity is the product they sell to donors, to governments, to the public, while our poverty is the condition that keeps them indispensable.
As long as we are celebrated but starved, international intermediaries remain the necessary bridge between resources and communities. Remove our poverty, and you remove their justification.
We are close enough to communities to be useful and far enough from power to be safe.
The Grand Bargain, signed in 2016, promised that 25% of humanitarian funding would reach local organisations by 2020. It is now a decade later. In 2022, just 1.2% of humanitarian funding reached local actors.
RLOs, the most local of all, received $26.4 million globally. Nearly half of that came from a single private foundation. The gap between the rhetoric and the reality is not a failure of implementation but a declaration of intent.
What's Unsaid | Let refugees lead
A refugee advocate says it's time for the refugee-led organisation, or RLO, 'revolution'.
The trap no one names
The conversation about RLOs almost always centres on one thing: We need more money. This is true but incomplete. Because even when money does arrive, it arrives in a form designed for projects, not institutions.
It arrives for six months or twelve. It covers programme activities but not the people who sustain them. It requires reporting frameworks built for organisations with full-time finance teams and provides no funding to hire one.
The result is a structural trap. To access funding, RLOs must demonstrate organisational capacity, but the funding available prevents them from building that capacity. They must be professional enough to receive a grant, but the grant is never large enough or long enough to become professional in any durable way.
In Egypt, researchers have documented how RLOs cannot formally register and cannot open bank accounts, making it nearly impossible for donors to fund them directly, even when donors want to.
In Kenya, some refugee IDs are not recognised by the Kenya Revenue Authority or key government registration systems, meaning that founding and formalising an organisation requires workarounds that international founders never face.
These are not administrative inconveniences. They are walls built into law, into donor policy, and into the humanitarian system's basic architecture that keep RLOs permanently at the margins of the sector that claims to serve their communities.
The staff who leave
Behind every statistic about underfunded RLOs is a person who chose to stay until the day they didn't. At SIR, we have watched skilled, committed staff members leave - not because they stopped believing in the work - but because they could not afford to keep doing it.
The salaries we can offer are set by what project budgets allow, typically a fraction of what an international NGO, or even a national NGO, can provide.
The staff who leave take something irreplaceable with them: institutional memory, community trust, and the relational capital that takes years to build. The next hire starts at zero. The community notices. Donors do not.
This is not unique to Kakuma. Across the sector, RLOs are informal training grounds for the international humanitarian system. We identify talent in displacement contexts, invest in developing it, and then watch as larger, better-resourced organisations absorb it because they can offer what we cannot.
The humanitarian system benefits from this pipeline. RLOs pay the cost.
Research on localisation dynamics confirms the same pattern: local organisations take on the operational risk of crisis response while coordination structures and decision-making authority remain firmly in the hands of international actors.
We run the programmes. They run the meetings that decide the programmes.
Built to stay small
The subcontracting model where RLOs implement activities designed, funded, and evaluated by international intermediaries is now so normalised that many in the sector don't recognise it as a structural choice. But it is. And it is one that reproduces itself.
Under this model, RLOs bear the operational risk of project implementation with none of the institutional stability that comes from multi-year core funding.
Overheads and indirect costs - the budget lines that fund salaries, systems, and governance - are frequently denied or capped at the first intermediary level, never reaching the organisations doing the work on the ground.
The result, documented across contexts from Africa to the Middle East, is organisations serving as what one pan-African body described plainly as "hired hands supplying cheap labour for project-based aid". Capacity building, when it happens, rarely aims higher than producing a more reliable subcontractor.
Meanwhile, the same international organisations that restrict RLOs overhead costs maintain headquarters in big and expensive cities, with executive teams, communications departments, and governance structures all funded as core costs.
No one questions whether they have "capacity". The question is only ever asked of us.
The localisation agenda, which promised to shift power to local actors has in practice often deepened the imbalance.
International actors still determine whether local actors are worthy humanitarians, whether they are capable of receiving funding, how much core support they are allowed, and who may participate in coordination meetings.
Localisation, as currently practised, is less a transfer of power than a rebranding of control. The system has produced what it intended to produce: a tier of local organisations large enough to be useful, but never resourced enough to be threatening.
What genuine investment looks like
Some funders are beginning to do this differently, and their experience is instructive.
Longer-term, multi-year grants - not the one-year project cycles that dominate the sector - allow RLOs to retain staff, invest in systems, and plan beyond the next reporting deadline.
Funders who have moved to this model report that RLOs demonstrate impact just as effectively as international organisations, once they are given the time and stability to do so.
The Resourcing Refugee Leadership Initiative has shown that removing difficult registration requirements, using fiscal sponsorship and creative intermediary arrangements, can unlock direct funding to RLOs that would otherwise be unreachable.
These are not complicated innovations. They require only the political will to stop using risk-management frameworks designed for a different context.
At the policy level, Kenya's Shirika Plan - an initiative to transform refugee camps into integrated settlements, promoting development opportunities and self-reliance for refugees and host communities - represents a moment of genuine opportunity.
If the plan is to be more than a reworking of the same relationships, it must include core, multi-year operational funding for RLOs - not as project implementers, but as institutions with a legitimate, permanent role in the humanitarian and development ecosystem.
What needs to change
- Fund institutions, not just projects. Multi-year core funding must become the norm, not the exception. Staff salaries, systems, and governance are not overheads - they are the organisation. Without them, RLOs remain permanently fragile.
- Close the salary gap deliberately. Donors should require that competitive RLO salary benchmarks are included in budget negotiations and treated as programme costs, not indulgences. The brain drain from RLOs to INGOs is a predictable outcome of a system that funds one and starves the other.
- Remove the registration trap. Governments and donors must accept alternative formalisation pathways such as fiscal sponsorship and verified track records as legitimate proxies for institutional credibility. Demanding formal registration while preventing it is not due diligence: It is exclusion with paperwork.
- Redistribute decision-making, not just implementation. Programme design, needs assessments, and strategy development must include RLOs from the beginning with time and budget allocated for genuine participation. Consulting us after the PowerPoint is already prepared is not participation. It is performance.
Sign up for free AllAfrica Newsletters
Get the latest in African news delivered straight to your inbox
The cost of the status quo
When a talented staff member leaves SIR for an organisation that can pay them properly, I do not blame them. I congratulate them. They deserve security and recognition. But I also know that the community loses something when they go - and that the system responsible for that loss will never be the one to account for it.
The humanitarian sector likes to speak about resilience - the resilience of displaced communities, their ability to adapt and survive under extraordinary pressure. What it rarely acknowledges is that this resilience is not free. It is produced, sustained, and quietly subsidised by RLOs operating on scraps, losing their best people, and being asked to grow without the conditions for growth.
We are not failing to grow. We are being prevented from growing by legal barriers that deny us the right to exist formally, by funding models that cap our ambition at the next project deadline, by a subcontracting structure that positions us as implementers rather than leaders, and by a participation culture that invites our voices but rarely our decisions.
Everyone says we are closest to the community. It is time to ask who benefits most from keeping us farthest from the resources.
The system produced what it intended to produce. The question now is whether those with the power to redesign it have the honesty to say so and the will to build something different.
Bahana Mirindi Hydrogene, Founder and executive director of Solidarity Initiative for Refugees, a refugee-led organisation in Kakuma camp, Kenya
