Kenya's PBO Act Is Finally in Force After a Decade. Most Organisations Still Cannot Comply.

In 2025, with the PBO Act finally in force and the window to re-register under it closing, I sat with a governance advisor at a Nairobi-based civil society umbrella body and asked her to walk me through what a community-based organisation actually needs to do to register under it. She had helped several dozen organisations attempt it. She pulled up a checklist she had built from the legislation and the regulatory authority's guidance, and we went through it.

Two hours later, I had a different understanding of why so few organisations were ready for the new framework — even though the law behind it had been passed more than a decade earlier.

The checklist had thirty-seven line items. Some were simple: a copy of the national ID for each founding board member, a bank account in the organisation's name, a registered physical address. Others were complex: a constitution whose provisions map, clause by clause, onto the specific requirements of Section 8 of the Act. A conflict of interest policy that is not merely referenced in the constitution but spelled out in sufficient detail to demonstrate it is operative. Minutes from a founding general meeting conducted according to procedures the constitution must itself specify.

Most community-based organisations in Kenya do not have all of this. Many have most of it. Almost none have it in the form the Act requires without some revision. And the revision is not trivial — it requires someone who has read both the existing constitution and the Act carefully enough to identify the specific gaps between them.

This essay is about what the Act actually requires, and about why the gap between those requirements and what a typical organisation has is not a function of bad intent or disengagement. It is a function of the administrative distance between drafting a law and building the infrastructure that makes the law navigable.

It helps to be precise about the timeline, because the timeline is part of the problem. The PBO Act was passed in 2013 — and then sat unimplemented for over a decade, its commencement repeatedly delayed despite a 2016 High Court order to bring it into force. It was finally operationalised on 14 May 2024, repealing the Non-Governmental Organizations Co-ordination Act of 1990 that had governed the sector until then. Organisations registered under the old framework were deemed transitioned but required to apply afresh under the new one within a year. So the compliance obligation most organisations are now scrambling to meet is not old. It is barely more than a year old — even though the law behind it has been on the books for more than a decade.

Clause by clause, most of the Act simply never connects to what a CBO actually has on hand.

1. The clause-mapping claim, and what grounds it

This paper argues that a specific administrative gap — between what the PBO Act's Section 8 requires in a constitution and what a typical Kenyan community-based organisation's existing constitution contains — is the dominant reason the transition has been slow, rather than cost, malice, or disengagement. Here is what grounds that claim, before making it.

What it argues from. A structured, clause-by-clause walk-through of a real compliance checklist, conducted with a governance advisor who has personally taken "several dozen" organisations through the process — a practitioner's working knowledge of where organisations actually fall short, not a survey. The Act's own text, read directly, for what Section 8 requires.¹ The regulator's and the courts' own public record of what happened after commencement: the transition statistics, the de-registration figures, and the 2025 High Court ruling in Otieno & 2 others v Attorney General, read in the secondary legal-practitioner literature rather than the gazetted judgment text itself (Kenya Law's own repository blocks automated retrieval; see Appendix B). Comparative evidence on non-profit compliance-cost research from outside Kenya, and on the donor-due-diligence and shrinking-civic-space pressures that make non-compliance costly even where legal enforcement is lax.

What it does not argue from. It does not argue from a survey of Kenyan CBOs' actual compliance status, their constitutions, or their reasons for non-compliance — no such systematic study is cited here because I did not find one specific to this transition. The "thirty-seven line items" and the sixty-to-two-hundred-thousand-shilling cost estimate are one practitioner's working account, not a measured distribution across organisations. Where this paper uses press-reported transition and de-registration figures, they are reported here as order-of-magnitude approximations, not as precise rates — the regulator's own public disclosures on this have been inconsistent in both method and timing.

What it does not claim. It does not claim that clause-mapping cost is the only barrier to transition — §7 names at least two others (awareness and regulatory capacity) that this paper cannot rule out and has not measured against the cost-barrier story. It does not claim the PBO Act is poorly drafted — the opposite is closer to true, and §2 shows the requirements to be individually reasonable. It does not claim the 2025 High Court ruling in Otieno was wrongly decided or should be reversed — §6 treats it as a fact about the world, not a position to argue against.

What would refute it. Given in full in §9. The sharpest version: a systematic survey of non-transitioned CBOs finding that most cite reasons unrelated to constitutional clause-mapping — simple unawareness that the deadline exists, or a deliberate choice not to engage with formal registration at all — would mean this paper has correctly described a real gap but wrongly identified it as the dominant one, and the argument would need to be narrowed rather than merely qualified.

2. The Constitutional Requirements in Detail

Section 8 of the PBO Act specifies the minimum contents of a PBO's constitution.¹ I am going to reproduce them here — not abstractly, but as they read when you are an organisation in Eldoret trying to check your own document against them.

An organisation's constitution must include: the name of the organisation and its registered address. A statement of the organisation's public benefit objectives — not a general statement of mission, but a description specific enough to demonstrate that the work falls within the Act's definition of "public benefit activity."² A description of the geographical area of operation. Provisions for membership — criteria for admission, the rights and obligations of members, and procedures for suspension or removal. The composition of the governing board: the number of members, the method of appointment or election, the term lengths, and whether those terms are renewable and how many times.

The constitution must specify the quorum for valid board meetings. It must specify the frequency with which meetings must be held and the procedures by which they are convened.³ It must contain a clause governing conflict of interest: board members must be prohibited from participating in decisions in which they have a personal or financial interest, and the clause must specify how such interests are to be declared and managed.

The financial provisions are specific. The constitution must require that the organisation's accounts be audited annually by a qualified auditor, that the accounts be presented to the governing board, and that financial records be maintained in a manner that allows for inspection on request.⁴ The constitution must contain a clause governing the amendment of the constitution itself — the threshold required, the notice period, and whether certain provisions are protected from amendment by simple majority.

The dissolution clause must specify that on winding up, the assets of the organisation shall revert to another organisation registered under the Act or to a public purpose fund, and not to any member or officer.⁵ An anti-discrimination clause is required. Provisions for the conduct of general meetings — separate from board meetings — are required, including quorum, notice period, and voting procedures.

This is the minimum. Some of these requirements are clear in their terms. Others — the definition of "public benefit activity," the standard for what constitutes a sufficiently specified conflict of interest policy — require interpretive judgment that a non-lawyer may not feel confident exercising. None of this, read clause by clause, is an unreasonable thing to ask of an organisation that wants legal recognition as a public benefit vehicle. The requirements track what company law and trust law already ask of boards elsewhere. The problem this paper describes is not that the list is wrong. It is that almost nobody starts from a blank page — they start from an old constitution that gets most of the way there, and "most of the way" is not a status the Act recognises.

3. What a Typical CBO Actually Has

The constitution that a community-based organisation registered under the old Social Services framework holds was, in most cases, drafted using a Ministry of Labour template from the years before the PBO Act came into force in 2024. That template was designed for the NGO Co-ordination Act 1990 framework, which had different and, in several respects, less specific requirements.⁶

These constitutions typically contain: a name and address, an objectives clause, a membership section, and a basic governing board structure. They often contain a dissolution clause, though the asset-reversion language may not match what the PBO Act now requires. They sometimes contain financial accountability provisions. They almost never contain a conflict of interest clause — this was not a standard requirement under the 1990 framework. They often lack the specificity the PBO Act requires in meeting procedures: quorum is named but not always correctly specified, meeting frequency is stated but may not meet the Act's minimums, board term limits are sometimes absent entirely.

The gap between "this organisation has a constitution" and "this organisation has a PBO Act-compliant constitution" is therefore, in most cases, not dramatic in substance — the organisation has most of the right elements — but significant in form. The specific provisions are absent, or present in language too general to satisfy the checklist, or present but buried in provisions that a reviewer cannot easily locate.

And because the organisation cannot tell, without a systematic clause-by-clause check against the Act's requirements, which provisions are missing or insufficient, it cannot fix the problem without either hiring a lawyer to do the check or finding some other mechanism to run it. This is the precise point at which "the law is reasonable" and "the law is navigable" come apart. A requirement can be individually sound and still be collectively inaccessible, if the only way to know whether you meet it is a service most of the organisations subject to it cannot afford.

4. The Compliance Cost

I asked the governance advisor how much it typically costs an organisation to get from "we have a constitution" to "we have a PBO Act-compliant constitution" using a lawyer. Her estimate: between sixty and two hundred thousand shillings, depending on how much revision is needed, which law firm is used, and whether the organisation is in Nairobi or in a smaller town where the legal market is thinner.

For an organisation with a monthly operating budget of three hundred thousand shillings — which describes many mid-size CBOs — this is a significant fraction of monthly operations. For an organisation with a budget of fifty thousand shillings — which describes many more — it is several months of operations.

This is not a story about regulatory malice. Nobody designed the compliance requirement to impose a cost this high on small organisations. The cost is an accidental product of the gap between the law's requirements and the infrastructure available to meet them. But accidental or not, the effect is the same: the compliance cost falls most heavily on the organisations that can least afford it, which are often the organisations doing the most necessary work in the most under-resourced contexts.

This is not a uniquely Kenyan pattern, and it is worth being clear about how much of it is not. The best quantified study I could find of what government-generated compliance paperwork actually costs a non-profit comes from Australia, not Africa: a 2009 study that had fourteen non-profit organisations log every hour spent on government paperwork over twelve months found that grant-related paperwork formed the bulk of the total compliance burden, and that the paperwork most expensive to complete was precisely the kind that required specialist judgment rather than routine data entry — the same distinction this paper draws between a CBO's simple line items (an ID copy, a bank account) and its complex ones (clause-by-clause constitutional mapping).⁷ That study was conducted in a high-income country with a mature non-profit legal-services market; it is cited here for the general shape of the finding — compliance cost concentrates in the interpretive tasks, not the mechanical ones — not as evidence about Kenyan costs specifically, which I could not find quantified anywhere in the literature. No Kenya- or Africa-specific study measuring PBO or equivalent compliance cost across a sample of organisations was found; Appendix B registers that absence directly.

5. The Enforcement Question

A response I sometimes hear to this argument runs as follows: the enforcement of the PBO Act has been inconsistent. Many organisations that are not yet compliant have so far operated without consequence. If the cost of non-compliance is low, why is the compliance barrier a problem worth solving?

There are two answers. The first is that the cost of non-compliance is not primarily a legal cost — it is a funding cost. The organisations that cannot demonstrate compliance cannot access the growing category of donor and partnership opportunities that now require it, a dynamic the next essay in this series examines directly. This funding cost is growing, not shrinking, as international donors align their due diligence requirements with anti-money-laundering and counter-terrorism-financing frameworks. The relevant global standard here is the Financial Action Task Force's Recommendation 8, which directs states to apply "focused, proportionate and risk-based measures" to the non-profit sector specifically because of its perceived vulnerability to misuse for terrorist financing.⁸ Recommendation 8 is a standard for states, not for donors directly — but a 2025 implementation guide aimed at both state and civil-society actors is explicit that non-compliance with FATF's recommendations "can raise a non-compliant country's cost of access to international financial systems," and the same risk-based logic migrates downstream into the due-diligence checklists banks and institutional donors apply to individual organisations, where an unregistered or non-transitioned status reads as an unmitigated risk flag rather than mere informality.⁸ A CBO that cannot prove PBO-compliant status is not merely facing a dormant legal liability. It is already, in practice, locked out of the funding relationships where that status is checked.

The second answer is about the tempo of enforcement. Regulatory enforcement in East Africa has a characteristic pattern: long periods of de facto tolerance followed by sudden, politically-triggered crackdowns that are difficult to predict and fast in their consequences. Tanzania's experience under its NGO Act — culminating in 2018, when a government that had tolerated registration non-compliance for years moved suddenly to deregister nearly 5,000 organisations under a 30-day re-registration directive — is the relevant case study.⁹ The organisations that survived that period were the ones that had treated compliance as infrastructure before it was demanded. The ones that had not had weeks to catch up, and most could not.

Kenya's own civic-space trajectory makes the "enforcement is lax so this doesn't matter" response harder to sustain, not easier. CIVICUS Monitor — which tracks civic-space conditions in every country against a five-category scale — downgraded Kenya's rating to "Repressed," its second-worst category, in 2024, directly citing the state's violent response to the June 2024 anti-Finance Bill protests.¹⁰ A political environment in which civic space is actively contracting is not a safe environment in which to assume that today's enforcement laxity about PBO transition will persist. It is, on the historical pattern this paper is describing, exactly the environment in which a sudden, selective crackdown becomes more rather than less likely — and selective enforcement falls hardest on organisations that cannot quickly prove, on paper, that they were never the intended targets.

The right time to build compliance infrastructure is before it is needed.

6. What the Otieno Ruling Changed — and What It Did Not

On 30 April 2025, the High Court delivered judgment in Otieno & 2 others v Attorney General & another; Katiba Institute & 9 others (Interested Parties), Petition E519 of 2024, and the ruling is directly relevant to this paper's argument — in a way that complicates it before it confirms it, which is the honest order to take it in.¹¹ Multiple independent legal-practitioner summaries of the judgment agree on its substance, even where they number the underlying sections of the Act differently: the court found that requiring organisations already registered under the old NGO Co-ordination Act to make a fresh application for PBO status — rather than transitioning automatically — violated their rights to fair administrative action and freedom of association, and it issued an order of mandamus directing the regulatory authority to transition and register such organisations automatically, without a fresh application. A separate provision compelling disclosure of membership lists and donor details was struck down on privacy grounds, the court finding the disclosure requirement disproportionate and a risk to activists' safety. A third set of findings addressed the Authority's power to suspend or cancel registration without a fair hearing, on natural-justice grounds.¹¹

If the barrier to transition were primarily the cost and procedural burden of the re-application itself — new forms, new fees, a fresh adjudication — then a court order requiring automatic transition, with no new application and no new fee, should have cleared the backlog almost immediately. It did not. The regulator's own figures, read as order-of-magnitude approximations rather than precise rates given how inconsistently they have been disclosed, put only around 4,000 of an estimated 14,000 eligible organisations as transitioned by mid-2025, with roughly 4,779 de-registered in the year to mid-2026.¹² That gap, persisting after the single largest procedural obstacle to transition was judicially removed, is the strongest evidence available to this paper for its own central claim: the thing most organisations are actually missing is not a fee or a form. It is a constitution that passes a clause-by-clause test they have no accessible way to run.

What this does and does not establish is worth stating plainly. The Otieno ruling addressed the procedure of transition — whether an already-registered organisation had to reapply, and whether it had to hand over donor and membership data to do so. It did not touch, and could not plausibly have touched, the underlying substance of Section 8: an organisation that automatically transitions under the court's order still holds, the day after transition, whichever constitution it held the day before — one that in most cases still lacks a conflict-of-interest clause, still understates meeting-frequency requirements, and still has no clean mechanism for the organisation itself to discover which of these gaps is disqualifying under subsequent compliance review (annual Form 13 filings, audited-accounts requirements, and board-change notifications all continue to apply regardless of how the initial transition happened). Automatic transition solves the admission problem. It does not solve the standing-compliance problem this paper is about. That the backlog persisted after the admission problem was solved is evidence for this paper's thesis; it is not proof of it, because the ruling is recent enough (April 2025) that normal administrative lag — the regulator updating its systems, organisations learning the order exists at all — could also explain some of the continued gap, and this paper has no way to apportion the gap between the two.

7. Why the Queue Still Hasn't Cleared

Two explanations compete here, and this paper is explicitly a bet on one of them over the other without having ruled the other out.

The first is this paper's own: organisations that want to transition cannot, because the clause-by-clause gap between what they have and what the Act requires is invisible to them without specialist help they cannot afford, and Otieno's procedural relief did nothing to close that specific gap.

The second is a demand-side story: some meaningful share of the roughly 10,000 organisations that have not transitioned have not tried to, because they do not know the deadline exists, do not believe PBO status matters to their work, or have functionally wound down without formally deregistering. Kenya's registration and reporting infrastructure for small community organisations has never had a reliable mechanism for distinguishing a dormant, no-longer-operating CBO from an active one struggling to comply — which means some fraction of the "non-transitioned" 10,000 may not be a compliance failure at all, simply an administrative population count that has never been cleaned up.

This paper cannot adjudicate between these two stories with the evidence available to it. What it can say is that the clause-by-clause gap described in §§2–3 is real, independently of which story explains more of the aggregate backlog — the governance advisor's account of "several dozen" organisations she personally walked through the checklist describes organisations that were actively trying to transition and still could not do so without her help. That is direct evidence the first story operates on at least some real, motivated population. It is not evidence about what share of the total backlog that population represents, and §9 states the test that would settle the question in the other direction.

8. Objections I take seriously

"The strongest piece of evidence you have — the backlog persisting after Otieno — is also the evidence against you, if the explanation is administrative lag rather than the clause-mapping gap." This is the most serious complication in §6, and I do not think this paper fully resolves it. A court order from April 2025 directing automatic transition does not mean every regulatory system, form, and front-line officer implementing it updated overnight; some of the continued backlog is very plausibly bureaucratic catch-up that will resolve on its own within a year or two of the ruling, which would have nothing to do with this paper's clause-complexity thesis. I do not have a way, from outside the regulator, to separate "still backlogged because of administrative lag" from "still backlogged because the constitution still fails the clause test." §9's first falsification condition is this objection stated as a commitment rather than a shrug.

"Flux has a commercial interest in this conclusion." Flux builds PBOMaster, software designed specifically to close the gap this paper describes — automated, systematic checking of a CBO's constitution against the PBO Act's clause-by-clause requirements, intended as a lower-cost substitute for the lawyer-driven review this paper says most organisations cannot afford.¹³ An essay arguing that the compliance gap is wide, structural, and not primarily about fees is, read uncharitably, an advertisement for exactly the kind of product Flux sells. I am not going to pretend this essay is written by someone with no stake in the conclusion. The honest mitigation is narrow and the same one used elsewhere in this series: the governance advisor's account predates PBOMaster's existence and was not solicited to make a sales case; the Act's own text in §2, the comparative compliance-cost literature in §4, and the Otieno backlog data in §6 are independent of anything Flux built or sells. They would be true, and this paper's argument would stand, whether or not Flux had ever built PBOMaster.

"Enforcement has been lax for over a year with no mass crackdown, so the urgency this paper implies is itself overstated." This is the direct counterpart to the enforcement question in §5, and I think it understates two things working against complacency rather than for it: the donor-due-diligence channel, which does not require state enforcement to bite because it operates through funders' own risk checklists rather than through prosecutions; and Kenya's own documented civic-space trajectory, which — unlike a hypothetical stable low-enforcement equilibrium — is actively deteriorating by the measure cited in §5. "Nothing has happened yet" is weaker evidence of safety in a contracting civic-space environment than in a stable one.

"A governance advisor's single practitioner account of 'several dozen' organisations and one cost estimate is not a basis for a structural claim about an entire regulatory transition." This is correct as stated, and I am not aware of a published, systematic alternative — no survey of Kenyan CBOs' actual constitutional compliance status against Section 8, and no quantified Kenya-specific compliance-cost study, was found in researching this paper. §4's comparative Australian study is the closest available quantified analogue, and it is explicitly not about Kenya. This is a real limitation, not a solved problem, and Appendix B registers the missing study as a standing gap rather than something this paper's own interview evidence can substitute for.

9. What would falsify this

Stated as commitments, in decreasing order of how much damage each would do to this paper's claims.

A systematic survey of non-transitioned Kenyan CBOs finds that most cite reasons unrelated to constitutional clause-mapping — simple unawareness the deadline exists, a deliberate choice not to pursue formal registration, or defunct/dormant status rather than active non-compliance — as the dominant explanation for non-transition. That would mean this paper correctly identified a real gap (§§2–3 are not in dispute) but wrongly identified it as the dominant cause of the aggregate backlog described in §6, and the paper's framing would need to narrow from "the reason most organisations have not transitioned" to "a reason some actively-trying organisations cannot transition."

Transition rates are shown to have risen sharply and continuously in the twelve months following the Otieno ruling, converging toward near-complete transition without any parallel clause-mapping intervention (legal-aid programmes, software tools, or paralegal drives) reaching the organisations that transitioned. That would indicate the backlog in §6 was predominantly administrative lag rather than the standing clause-mapping gap this paper argues is the deeper problem, directly undermining this paper's reading of the post-Otieno data as evidence for its thesis.

A quantified, Kenya-specific compliance-cost study finds the actual cost of achieving Section 8 compliance is low relative to typical CBO budgets — contradicting the sixty-to-two-hundred-thousand-shilling practitioner estimate in §4 at the population level rather than merely in individual cases. That would not refute the existence of a clause-by-clause gap, but it would refute this paper's claim that the gap is financially, rather than merely administratively, significant for most organisations.

The donor-due-diligence mechanism in §5 is shown not to operate in practice — for instance, if a survey of Kenyan CBOs' actual funding relationships finds PBO-transition status is rarely or never checked by donors and partners, contrary to this paper's claim that non-compliance carries a real funding cost even where state enforcement is lax. That would remove the paper's strongest answer to the "enforcement is lax, so why does this matter" objection in §5 and §8.

10. Where the clause-mapping claim stops

No systematic survey of Kenyan CBOs' compliance status exists; the governance-advisor interview and the regulator's aggregate transition figures are the evidentiary base this paper rests on, and §9 states what a systematic survey finding something different would do to its claims.

It does not claim the sixty-to-two-hundred-thousand-shilling compliance-cost estimate is a population-level average; it is one practitioner's working estimate, and no quantified Kenya-specific study was found to check it against.

It does not claim to be able to apportion the post-Otieno transition backlog between the clause-mapping explanation this paper favours and the administrative-lag or demand-side explanations named in §7 and §8; it states the evidence for its own explanation plainly while naming the competing explanations it cannot rule out.

It does not claim the PBO Act is badly drafted; §2 argues the individual requirements are reasonable, and the paper's claim is specifically about the navigability gap between those requirements and what organisations already hold, not about the law's substance.

It does not claim the 2025 Otieno ruling was wrongly decided, under-enforced, or will not eventually resolve the transition backlog on its own as the regulator implements it; §6 treats the ruling's procedural scope and its limits as a fact about the world that this paper's argument has to be consistent with, not a position to defend or attack.

And it does not extend this argument to civil-society regulation outside Kenya, or to PBOs of a scale and sophistication that already retain in-house legal capacity; it is a claim about the specific population of under-resourced, community-based organisations described throughout §§2–4.

Appendix A — Evidence table

Appendix B — Citation register

CITATION-NEEDED — Kenya-specific compliance-cost study. Appendix A row 10. No published study quantifying the cost of achieving PBO Act Section 8 compliance across a sample of Kenyan CBOs was found; §4's comparative evidence is Australian and non-profit-sector-general, not Kenya-specific.

CITATION-NEEDED — apportioning the transition backlog. Appendix A row 11. §7 names two competing explanations for why ~10,000 organisations remain non-transitioned (clause-mapping cost versus awareness/indifference/dormancy) and cannot apportion the backlog between them with the evidence available.

CITATION-NEEDED — post-Otieno transition-rate trend. Appendix A row 12. §9's second falsification condition depends on tracking the transition rate specifically in the period after the April 2025 ruling, isolated from parallel interventions; no such tracking was found at the time of writing.

Primary-source access note. Kenya Law's own repository and the PBO Regulatory Authority's website (pbora.go.ke) both refuse automated retrieval, which is why the Otieno judgment and the PBO Regulations 2026 fee schedule are cited here via practitioner legal-alert summaries rather than the gazetted or judgment text directly. The summaries used agree with each other on substance; the underlying gazetted text still needs a human read before any of these figures are treated as settled beyond order-of-magnitude.