n.V1-5.04| WHAT ARRIVES
01| Money enters this system at the top from a small group of donor governments and leaves it at the bottom in a settlement. Between those two points it passes through organisations, and each passage costs something. That is not scandal; it is how any delivery chain works. What matters is the shape of the chain and who it is arranged to protect.
02| The UN's deduction is standardised and public. The inter-agency working group endorsed "the use of 7% as a harmonized indirect PSC rate" for multi-donor trust funds and country-level joint programmes, and the same figure recurs across UNDP, UNICEF, UNFPA and UN Women. INGOs charge seven to ten percent on institutional grants, plus a further three to eight percent of programme management embedded in the budget itself. National sub-grantees typically receive contracts with lower overhead recovery, or none.
03| A donor dollar passes through two to four organisational layers before it reaches a settlement. Independent estimates of how much arrives as cash or in-kind value range from thirty-five to sixty percent, with wide variance by intervention type.
04| That range is doing a lot of work in a lot of arguments, and it should be treated with suspicion by anyone using it, including us. A spread of thirty-five to sixty is not a measurement. It is a placeholder for a measurement nobody has taken. For any specific settlement — Mathare included — no actor publishes what arrives at ground level against what is consumed in intermediation. The figure is not withheld. It does not exist in a form a resident, a journalist or a donor could query.
05| The more revealing pattern is not the percentage but the geometry. Four things change as a contract chain approaches the ground. The contracts get smaller. The durations get shorter. Overhead recovery falls toward zero. And reporting requirements get heavier relative to contract size.
06| Local organisations in a settlement like Mathare typically work on annual or sub-annual contracts of twenty to two hundred thousand dollars, with no margin for institutional growth and no funding for the multi-year planning the actors above them treat as routine.
07| Follow what that does over time. An organisation that cannot build reserves cannot survive a funding gap. It cannot hire staff it does not yet need. It cannot bid for anything larger than its last contract. It carries delivery risk without being permitted to accumulate the capital that would let it absorb risk.
08| Then the sector observes that local actors lack capacity, and cites the observation as the reason it cannot fund them directly.
09| This is the part worth being precise about, because it is not hypocrisy and does not require anyone to be lying. The contract geometry manufactures the capacity deficit. The deficit is then reported accurately. The accurate report justifies the geometry. Every step is defensible and the loop is closed.
10| The same loop shows up in the localisation numbers with unusual clarity. The 2016 Grand Bargain committed signatories to "an aggregated target of at least 25% of humanitarian funding to local and national actors as directly as possible." The reported position for 2024:
"Globally, just below 10% of international humanitarian funding reached LNAs, directly and indirectly, in 2024 – far below the Grand Bargain target of 25%."— ODI, The state of international humanitarian funding to local and national actors
11| Direct funding was 3.8 percent. Among the signatory donors themselves the direction reversed: their direct funding fell to 0.5 percent in 2024, from 0.6 percent in 2023.
12| Now look at who is meeting the target. Fifteen signatory aid organisations reported providing at least twenty-five percent of their funding directly to local actors — UNHCR, WFP, UNICEF, UNFPA among them. Four donors reported hitting it with one intermediary in the chain. So the agencies that pass money onward are compliant. The donors that originate it are not, and their share went down.
13| The intermediation the commitment was designed to reduce is where the compliance is happening. That is not a loophole someone exploited. It is what happens when a target is written about a percentage and attached to no consequence for missing it, over eight years, with annual reporting.
14| Underneath all of this sits the mandate cascade, which is supposed to give the whole structure direction: SDG 11 on sustainable cities, then the New Urban Agenda, then a national urban policy, then a county development plan, then settlement plans, then projects. It is loose at every joint, and the measurement system is why. SDG indicator 11.1.1 measures the proportion of urban population living in slums, informal settlements or inadequate housing, reported at national level, with city-level disaggregation a recent improvement. There is no settlement-level reporting.
15| So a project in Mathare can claim alignment with SDG 11 and there is no indicator at which its contribution could register. Alignment is claimable without being demonstrable — and, equally, without being refutable. The mandate chain is not a control system that directs behaviour. It is a paper trail that legitimises it.
