n.V1-4.02 | THREE THINGS CALLED OWNERSHIP

01| An inter-agency review of community-based child protection, led by Michael Wessells, worked through 160 documents across 60 countries to find what makes such mechanisms work. The answer was not money, technical design or expertise. Community ownership was the strongest determinant of effectiveness.

02| Its second finding is the one to sit with. Most NGO-established child welfare committees were top-down and achieved low ownership, and the evidence for that was not a metric. It was what people said. Residents described the committee in their own village as a Save the Children project, and, understanding it as somebody else's responsibility, exercised little of their own.

03| The committee existed. Residents sat on it. Meetings happened. It belonged to someone else.

04| That gap is the subject here, and the reason it survives is that one phrase covers at least three different arrangements. They come apart in every direction, and the differences decide what is standing after the funder leaves.

Decisions

05| The commonest meaning is that residents decide. They pick which street gets drainage, they set priorities, they sit on the allocating committee. Real, and not nothing.

06| Two things limit it. The menu is set elsewhere — a community choosing between drainage and street lighting has not chosen that the intervention should be infrastructure, because the programme design settled that before anyone was consulted. And decision rights of this kind are not transferred, they are lent. They exist because a project confers them and they end when it ends. A community that chose something owns, afterwards, nothing.

07| This is the type that fails Wessells' test while appearing to pass it. Residents participated in every decision and still called it a Save the Children project, because participating in someone's decisions is not the same as it being yours.

Assets

08| The second meaning is legal title. The clearest worked example is the community land trust, because its mechanics are written into an enforceable document. A mission-driven nonprofit holds the land permanently; residents own or lease the buildings on it. A long-term ground lease governs who may occupy, what is paid for the land, what mortgage may be taken, and what happens on resale — and the resale formula caps how much the price may rise, so the home stays affordable for the next qualifying buyer.

09| Something has genuinely changed hands: a mortgageable asset, held by a household, independent of any programme continuing.

10| Notice what the structure retains. The formula deliberately limits how much wealth the owner can accumulate — necessarily, since unlimited appreciation would defeat the purpose. Ownership here is real and conditional, and the conditions were drafted by the body holding the land, not by the people living on it. Worth adding, since the model attracts more enthusiasm than evidence: remarkably little rigorous research exists on how these trusts perform over long periods.

The entity

11| The third meaning is the one usually skipped. Who controls the organisation — its board, its decisions, and above all its money.

12| A community development corporation may own substantial property in a neighbourhood while its board is professionals and funders: community assets, non-community control. A residents' association may control itself completely and own nothing: it can decide and cannot act. Participatory budgeting hands over real decisions about public money and transfers neither asset nor entity.

13| The variant that combines most is the savings-based federation — Muungano wa Wanavijiji in Kenya, and the wider Slum Dwellers International network. The organising unit is collective savings: members' own money, pooled over years. That produces a form of control no grant replicates, for a structural reason rather than a moral one. A funder can stop funding. A funder cannot withdraw capital that was never theirs. The limit is scale — savings accumulated in informal settlements are small against the cost of the infrastructure those settlements need, so the leverage is political and informational rather than financial. It buys a seat at the table with the county, not a sewer.

Who can be removed

14| Entity control has a harder test than a board's composition, and it is the one question that settles the rest: can the people the organisation exists for remove the people who run it?

15| For most nonprofits the legal form answers before any vote is held. In a self-perpetuating board, "the board itself decides who it wishes to elect as members." Such an organisation "may allow people to become 'members,' but these 'members' do not have governing power or voting rights. They simply make a financial donation to support the organization." The governance literature states the consequence plainly:

"The board is not accountable to another body. Without external accountability, a board can become self-reinforcing."— Governance guidance on member-elected versus self-perpetuating boards (Governance Matters)

16| So a legal aid centre or a policy institute may spend forty years advocating for poor people while offering those people no route to replace its leadership. Not misconduct. That is what the form is.

17| One category in this field has an outside enforcer, and only one. Under the US Labor-Management Reporting and Disclosure Act, local unions must elect officers by secret ballot at least every three years, and a member in good standing "has the right to nominate candidates, to be a candidate subject to reasonable qualifications uniformly imposed, to hold office, and to support and vote for the candidates of the member's choice." Where a union's own rules provide no adequate removal procedure, the Secretary of Labor can find as much and members can vote an officer out anyway. Everywhere else in this sector, the rules that would remove a leader are written by the people the rules would remove, and can be amended by them.

18| Federations of member groups score highest on origin and lowest on documentation. Their claim is Batliwala's — the base groups created the federation rather than the reverse — and what is rarely written down anywhere is how a base group removes an officer, when elections fall due, or who adjudicates a disputed one. That is a serious gap in the category with the strongest moral claim. An organisation genuinely created by its constituents can still, twenty years on, be run by people no constituent can vote out.

19| A lock on power is structural, not personal, and needs nobody to want it. It has three parts. An undefined membership, so nobody can be shown to have standing to vote. An election interval the incumbents control, which means there is no election. And no external enforcer, so the only remedy for a broken rule is an appeal to the people who broke it. Every organisation that was ever captured had meetings, and most had bylaws.

What survives

20| The test that separates the three meanings is awkward and simple: what is left when the funder goes home. Decision rights disappear, because they were never transferred. Assets remain, on whatever terms the instrument specifies. Entity control remains only to the extent the entity has resources of its own — which is why savings-based structures behave differently from grant-dependent ones even when their constitutions read alike.

21| It would be tidy to conclude that legal ownership is the real thing and decision-making is theatre. The evidence does not allow it. Wessells' finding is about the first type, not the second. No land changed hands in those villages. What separated a committee that worked from one that did not was whether people had taken the problem on or been convened to help with somebody else's programme.

22| So there are two variables and they are not substitutes. One is legal: what was transferred, on what terms, and what survives contractually. The other is closer to authorship: whose problem this is understood to be, by the people living with it. A structure can deliver the first and fail the second — a housing trust residents experience as a landlord. It can deliver the second and fail the first — a village that owns its problem and owns nothing else, whose effort dissipates when the people carrying it move on.

23| One word collapses both, and the collapse is what lets an organisation claim ownership on the strength of a well-attended meeting.