Differentiated Assistance arrived in Kenya’s refugee camps in 2025 in the middle of the worst funding crisis in years, and it read to most observers as an austerity cut. The Government of Kenya, through the Department of Refugee Services, working with UNHCR and WFP, replaced thirty years of uniform, status-based support with a model that sorts every household into one of four categories: the most vulnerable (Category 1), those with limited means to meet basic needs (Category 2), the partially self-reliant (Category 3), and the fully self-reliant (Category 4). The timing made the story tell itself: budgets shrank, so rations had to be sliced and targeted.
That story holds up, but only for half of it: the funding crisis forced agencies to finally act on a recommendation that had sat unimplemented since 2014.
Blow the dust off the 2014 Joint Assessment Mission (JAM) report and a different chronology appears. The household vulnerability assessment WFP and UNHCR built in 2025 was first recommended in 2014, before the cuts, before the current discourse, before anyone called it Differentiated Assistance. Between 70 and 80 percent of refugees in Dadaab and Kakuma that year ran out of food before their next ration and turned to negative coping strategies.1 The 2014 team’s conclusion was blunt:
“The JAM recommends UNHCR and WFP to undertake Household vulnerability assessment in both Kakuma and Dadaab, to establish the feasibility and stimulus of differentiated assistance.”
That wasn’t a line buried in an annex. The report’s own summary of recommendations called for the same assessment to translate into an actual rollout within two to three years, putting implementation around 2016 or 2017.2 A separate chapter arrived at the same place from a different angle, noting that most households still depended entirely on food aid even though, in the report’s words, “opportunities for assistance differentiation exist[ed].”3 Two sections of one assessment, arriving independently at the same answer.
DA is eleven years overdue: the recommendation behind it dates to 2014, and this year’s funding crisis is why it finally moved forward.
Lost in Translation: How 2014's Call for Change Was Mired in Austerity Logic
The 2025 rollout was a trial by fire. It coincided with a sharp contraction in international aid, driven largely by cuts in US funding, that forced WFP to cut rations well below the levels DA was designed around. Category 1 and 2 households saw rations fall to 40 and 20 percent of the minimum food basket in the August/September cycle, against a design of 60 and 40 percent; Category 3 lost its 20 percent allocation entirely, per the October 2025 Differentiated Assistance Country Info Brief.4 To households already running on empty, it felt like the mechanism through which aid was being withdrawn, reform and retrenchment blurred into the same thing from where they stood. One community member put it plainly: sorting people into categories without trust invites conflict, because everyone in the camp already has nothing.5
The picture has since improved, on paper. Additional funding let WFP raise October/November rations to 55 and 35 percent for categories 1 and 2, and reinstate category 3’s 20 percent. The plan is to reach the original 60/40/20 design between December 2025 and March 2026.6 That doesn’t undo what the mid-2025 cuts did to trust. But it’s a useful tell: the rations are moving back toward the design the 2014 report pointed toward, not settling into permanent retrenchment dressed up as reform. The crisis was a detour on a road mapped out in 2014.
Editors Note: The food assistance levels have since been revised further up to Category 1: 80% Minimum Food Basket, Category 2: 60% Minimum Food Basket, Category 3: 20 % Minimum Food Basket based on this June 2026 brief.
Reclaiming the Vision the 2014 Report Already Drew
The same 2014 report that recommended differentiation also recommended how to build trust around it, and that second recommendation has been followed far less closely. The assessment singled out Food Advisory Committees, FACs, refugee-led structures that shared food-basket information and settled disputes at distribution points, as the vital link in making any assistance model legitimate. The numbers backed the claim: in one Dadaab feedback cycle, 76 percent of refugees said they were satisfied with the information they got about their food basket, and under-scooping complaints sat at 0.3 percent.7 The report’s instruction was blunt:
“The JAM recommends WFP to strengthen beneficiary complaints and feedback mechanisms in all GFP.”
DA’s current recourse mechanism, through which more than 14,600 households went to helpdesks between August and October 2025 and over 4,000 cases have already been reviewed and closed, answers part of that call.8 It gives one household at a time a way to contest its category, an appeal after the fact. The collective half of what FACs provided, a standing seat at the table before decisions get made, has no equivalent in the current design.
The gap runs all the way up. The Refugees Act’s own Refugee Advisory Committee, which advises the Cabinet Secretary on national refugee policy, seats Principal Secretaries, the Attorney-General, the Inspector-General, and a representative from the Council of Governors.9 Not one refugee. The Shirika Plan, four years later, repeats the structure: its National Technical Committee can co-opt refugees and refugee-led organisations for technical input, on a need basis, input, not a standing seat.10
What's Still on the Shelf
The 2014 report didn’t stop at Food Advisory Committees. It named a harder problem and gave it a deadline:
“One of the key limitations to refugee self-reliance and expansion of livelihood opportunities is restriction on movement and the limiting policy environment. UNHCR and the civil society should continue advocating with the national Government on freedom of movement for the refugees.”
The same year that recommendation was written, movement went the other way. In March 2014, after a string of attacks in Mombasa and Nairobi, the Interior Cabinet Secretary announced a plan to force more than 50,000 registered refugees out of Nairobi and other towns and into Dadaab and Kakuma.11 Recommendation 51 set its own clock: a full self-reliance strategy within two years, preceded by a household livelihoods survey.12 That clock ran out around 2016, the same year the assessment recommendation was also due. The survey itself arrived later still: DA’s current categorisation work is effectively that survey, a decade behind schedule.
The Refugees Act, 2021, came into force in 2022 and is described, in government planning documents, as a shift away from the old encampment policy. Section 35 has the Commissioner liaising with national and county governments so refugee matters get folded into development planning, almost the exact language Recommendation 51 asked for in 2014, just turned into law seven years late.13 But the same Act, in Section 31, still requires refugees to reside in a designated area, and Section 8 still has the Commissioner issuing movement passes for travel outside it.14 The break from encampment is real on the planning side and incomplete on the movement side, the same divide the 2014 report described.
The Shirika Plan, launched in March 2025 as the government’s own plan for refugee inclusion, doesn’t name freedom of movement anywhere in its text, a gap in government policy design, not in agency advocacy. The pillar closest to 2014’s self-reliance agenda, Sustainable Economic Development, gets $257 million, more than a quarter of the plan’s $943 million budget, second only to the Integrated Services pillar.15 Money for market access; nothing for the movement restriction the 2014 report said was blocking it.
Eleven years after Recommendation 50, one law and one plan later, refugees still need a pass to leave the place differentiation is supposed to help them leave.
What Finishing the 2014 Agenda Would Take
First, treat the DA Transition Framework, now being finalised, as what it actually is: a decade-old recommendation that a funding crisis forced into the open. Shirika already tells this story about itself: its own rationale blames coordination failure and duplicated services, not the funding crisis.16 DA can claim the same history instead of letting the crisis narrate it.
Second, bring back a modern Food Advisory Committee alongside the recourse mechanism already in place. One settles individual disputes; the other gives the community a hand in writing the rules before disputes start. DA needs both to hold onto people’s trust.
Third, close the self-reliance gap the 2014 report named and the funding crisis obscured: freedom of movement, market access, livelihood support that actually reaches categories 3 and 4. Differentiation only works as a pathway if the route it promises actually leaves the camp. “Graduating” from aid should be a choice earned through opportunity, never imposed as an exit.
Kenya, UNHCR, and WFP are managing a plan and a crisis at once: finishing a decade-old recommendation on assistance design while absorbing a funding shortfall nobody asked for. Movement is the harder case, a sovereign policy choice the agencies can advocate against but not overturn, and the record shows that advocacy hasn’t moved it yet. The 2014 report explains why the model looks the way it does. The 2025 crisis explains why it arrived now, and arrived rough. That doesn’t excuse how rocky the rollout was. The more urgent question is why it took this long, and what’s owed to the people who waited for it.
Notes
- WFP Food Security Outcome Monitoring, 2013, cited in 2014 JAM Report, p. 16.
- 2014 JAM Report, Summary of Key Recommendations (Coordination: refugee influx, voluntary repatriation and self-reliance, item 1).
- 2014 JAM Report, p. 56. Verbatim: “the majority of Dadaab and Kakuma refugees still depend entirely on food aid while opportunities for assistance differentiation exists.”
- Differentiated Assistance in Kenya: Country Info Brief, October 2025 (DRS/UNHCR/WFP), p. 2.
- Differentiated Assistance in Kenya: Country Info Brief, October 2025, “Voices from the community,” p. 4.
- Differentiated Assistance in Kenya: Country Info Brief, October 2025, p. 2.
- CARE beneficiary complaint feedback analysis, Dadaab, first cycle May 2014, cited in 2014 JAM Report, p. 17.
- Differentiated Assistance in Kenya: Country Info Brief, October 2025, p. 3.
- Refugees Act, 2021 (Cap. 173), s. 9(1).
- Shirika Plan, 2025, p. 44 (National Technical Committee membership).
- Human Rights Watch, “Kenya: Plan to Force 50,000 Refugees Into Camps,” 26 March 2014.
- 2014 JAM Report, Recommendation 51, p. 56.
- Refugees Act, 2021 (Cap. 173), s. 35.
- Refugees Act, 2021 (Cap. 173), ss. 8(1)(o), 31.
- Shirika Plan, 2025, p. 52 (Resource Requirements: total budget $943M; Sustainable Economic Development pillar $257M; Integrated Services pillar $382M, the largest single pillar).
- Shirika Plan, 2025, p. 19 (“1.3 Rationale of Shirika Plan”: cites parallel service provision and duplication, not funding levels, as the driver).
