Devolution and CDF: Supreme Court Limits on Funds

The Supreme Court’s 2022 decision in Institute for Social Accountability v National Assembly is often invoked in arguments about the National Government Constituencies Development Fund. That is understandable, but it can also produce a serious legal mistake. The Supreme Court decided the constitutionality of the Constituencies Development Fund Act 2013. It did not, by its formal orders, strike down every later constituency-based fund simply because it had a similar name.

The distinction has become especially important in 2026. The separate National Government Constituencies Development Fund Act 2015 was declared unconstitutional by the High Court in September 2024, but the Court of Appeal set that result aside on 6 February 2026 and upheld the 2015 Act except for part of section 43(9), which it severed. A notice of appeal to the Supreme Court was filed after that judgment. Any current article must therefore explain both the binding 2022 constitutional principles and the different present appellate position of the 2015 Act.

What the Supreme Court actually decided in 2022

Institute for Social Accountability & another v National Assembly & 5 others, Petition 1 of 2018 [2022] KESC 39 (KLR), arose from the long-running challenge to the Constituencies Development Fund Act 2013. The Supreme Court unanimously restored the central constitutional conclusion reached by the High Court: the 2013 Act was unconstitutional.

The Court examined devolution, public finance, division of revenue, Senate participation and separation of powers. Its method was structural. Instead of asking only whether constituency projects were socially useful, the Court asked where the Constitution located the relevant functions, who should administer them, how public money should move and whether legislators had crossed from oversight into implementation.

That method remains the decision’s strongest continuing contribution.

A constituency is not automatically a level of government

The Constitution creates two levels of government: national and county. Constituencies are central to political representation, but they are not a constitutionally established third level of government.

In the 2013 CDF scheme, the Supreme Court considered constituency machinery capable of intruding into county functions and creating parallel local service-delivery structures. The constitutional problem was not that national government may never spend money in a county. National institutions necessarily operate throughout the country.

The problem arises where an ordinary statute uses constituency structures in a way that displaces county functions, duplicates planning or blurs responsibility for public services. A constitutional review must therefore look beyond the name of the fund and examine what its institutions actually do.

Why separation of powers mattered

Members of Parliament have strong reasons to advocate for local needs. They represent constituents, debate legislation, appropriate money and scrutinise executive action. The 2022 Court nevertheless emphasised the constitutional difference between those functions and executive administration.

If legislators control project selection or implementation and later oversee the same expenditure, accountability can become circular. The person expected to scrutinise the use of public money becomes connected with the decision being scrutinised.

This does not mean the Constitution requires branches of government to operate in sealed compartments. Some interaction is expressly contemplated. The practical question is whether one institution has entered the core operational domain of another in a way that weakens checks, balances and traceable responsibility.

The revenue-sharing rule is equally important

The Supreme Court also rejected the idea that Parliament could validate the 2013 fund merely by describing the money as part of the national government’s share. The Constitution requires nationally raised revenue to pass through the vertical division between national and county governments before the national government allocates money to its own agencies.

That sequencing matters because fiscal labels cannot alter constitutional structure. Public finance is not only about whether money is eventually spent on a legitimate public purpose. It is also about whether the appropriation followed the constitutional route and whether the responsible institution can be identified and audited.

The IMF Fiscal Transparency Code offers a useful, non-binding international comparison. It emphasises comprehensive fiscal reporting, credible budgeting and clarity about public-sector responsibilities. Kenya’s constitutional rules are more specific and legally binding, but both approaches recognise that fragmented or opaque fiscal structures can weaken accountability even when individual projects appear beneficial.

Why the 2015 NG-CDF Act cannot be treated as already decided

The Supreme Court itself was careful to identify the statute before it. The 2015 NG-CDF Act was later legislation. Although some provisions and constitutional questions overlapped, a court still had to examine the later statute’s actual text and operation.

That examination produced major litigation. In Gikonyo v National Assembly, the High Court in September 2024 declared the 2015 Act, as amended in 2022 and 2023, unconstitutional and ordered the Fund to cease operating on 30 June 2026.

The legal position changed before that date. On 6 February 2026, the Court of Appeal in National Assembly v Gikonyo & 9 others [2026] KECA 214 (KLR) set aside the High Court judgment. It held, among other things, that the 2015 Act did not generally violate devolution, separation of powers or public-finance principles and that Senate involvement was not required in the way the High Court had found. The Court severed the unconstitutional part of section 43(9) tying the Fund Account Manager’s tenure to the parliamentary term and left the rest of the Act standing.

What the 2026 development means for precedent

The 2026 Court of Appeal judgment does not erase the 2022 Supreme Court decision. A Court of Appeal cannot overrule the Supreme Court. The two decisions address different legislation and therefore have to be read with care.

The 2022 Supreme Court ruling remains binding on the constitutional defects it identified in the 2013 Act and on the principles of constitutional interpretation it established. The 2026 Court of Appeal concluded that the later 2015 Act, especially after amendments, was materially different in ways that avoided most of those defects.

That produces a useful litigation rule. A party challenging a modern fund should not simply cite the 2022 case and assume invalidity. The party must identify the provision of the current law, show how it reproduces a constitutional defect recognised by binding precedent and address any later appellate judgment that has already considered the revised structure.

Likewise, a public body cannot answer every challenge by saying the 2015 Act survived the 2026 appeal. A particular exercise of power may still be unlawful, procedurally unfair or inconsistent with public-participation and public-finance duties even where the statutory framework itself remains operative.

The current position as at 7 September 2026

As at the date of this article, the Court of Appeal’s 6 February 2026 judgment is the latest final appellate merits decision identified in the public sources reviewed for the 2015 NG-CDF Act. The respondents filed a notice of appeal indicating an intention to challenge that judgment in the Supreme Court.

A notice of appeal is important, but it is not itself a Supreme Court reversal. Until a later competent order changes the legal position, lawyers must distinguish an intended appeal from a decided appeal. The Court of Appeal judgment therefore remains essential to any current explanation of the 2015 Act.

This is precisely why legal publishing should state dates and procedural posture. A reader who sees only the 2022 Supreme Court decision could wrongly conclude that the current NG-CDF framework is already void. A reader who sees only the 2026 Court of Appeal decision could wrongly conclude that the older Supreme Court structural principles are irrelevant. Both conclusions are too simple.

A practical constitutional design test

The strongest continuing value of the 2022 case is a method for testing public funds. Identify the constitutional function the project serves. Identify whether that function belongs to national or county government. Trace the source and route of the money. Identify who selects, approves, implements and audits the project. Determine whether legislators retain oversight distance from executive administration. Consider whether the Bill in substance affects county institutions and what bicameral procedure the Constitution requires.

The same method can be used by parliamentary counsel before legislation is enacted, by auditors reviewing a fund and by litigants framing a constitutional petition. It replaces arguments based on political popularity with questions that a court can actually decide.

The method also improves public understanding. Citizens can support a development programme while still asking whether responsibility is clear enough to know who should be praised for success, blamed for failure or investigated for misuse.

Why the 2022 decision still matters

The Supreme Court’s 2022 CDF judgment remains a major case on devolution, public finance and institutional accountability. Its importance is not a permanent judicial ban on the word “constituency.” Its importance is the insistence that ordinary legislation must fit the governmental architecture created by the Constitution.

The 2026 Court of Appeal decision makes accurate citation more demanding. It requires lawyers and legal publishers to separate the invalid 2013 Act from the later 2015 NG-CDF framework and to explain the current procedural position honestly.

That distinction also helps avoid low-value legal writing. The useful question is not “Is CDF constitutional?” in the abstract. The useful question is which statutory design is being examined, what functions and financial routes it creates, what the latest binding or controlling court has said, and whether the challenged provision reproduces a defect that the Constitution does not permit.

Source note

This article is based on Institute for Social Accountability & another v National Assembly & 5 others [2022] KESC 39 (KLR), the Constitution of Kenya 2010, Gikonyo v National Assembly [2024] KEHC 10886 (KLR), National Assembly v Gikonyo & 9 others [2026] KECA 214 (KLR), the notice of appeal reported after the 6 February 2026 judgment, the current NG-CDF statutory framework, and the IMF Fiscal Transparency Code as a non-binding international comparison.

Suggested citation: 

Ronald Serwanga, “D Devolution and CDF: Supreme Court Limits on Funds” East Africa Legal Insight (7 September 2026).