The national government would not matter much if devolution delivered

Almost 15 years since the 2010 Constitution was promulgated, 12 years after the first post-constitution election that birthed our current system of devolved government and eight mega conferences later, what is the true state of devolution today?  Is it dead, or dying, as some suspect?  The short and easy answer here is our devolved system of government is still a work in progress.  

Lest we forget, devolution was touted as the crown jewel in our progressive and transformative constitution.  It was the antidote to our long and dark history of concentrated political and economic power; the response to years of marginalisation and exclusion for large swathes of the country; the promise of equity in access to resources, services and development; the final ascent to our desired mountaintop of inclusive and progressive political, social and economic order. 

As many still romanticise today, “if devolution worked, the presidency wouldn’t matter so much”.

This was another way of speaking to the expectation of limited national government in our lives. 

Which leads us to a more immediate question: do we still have too much national government? 

Let’s answer this through a couple of data stories, beginning with the 2014 and 2024 Division of Revenue Acts (DORA) for 2014/15 and current 2024/25 fiscal years. 

• .

  Keep Reading

• 
  Kiambu reaps as Ruto allies roll out empowerment project
• 
  Why Ruto should shun centralised system and fully support devolution
• 
  MPs earn the highest for resting the most
• 
  Treasury to revise Budget estimates

In 2014, respective national and county shareable revenue amounted to Sh800 billion and Sh226 billion respectively. 

In 2024, we have two sets of numbers, before and after Finance Bill 2024.  Before the bill, total shareable revenue in DORA of Sh2.948 trillion was shared as Sh400 billion for counties; Sh2.598 trillion for national government.  After the bill collapsed, the amended DORA shared Sh2.587 trillion as Sh387 billion for counties and Sh2.2 trillion for national government. 

Here’s the picture that emerges.  The county share in revenue fell from 22 per cent in 2014 to 13.6 per cent in the original 2024 DORA before squeaking back to 15 per cent in the amended version.

 Comparing 2014 and 2024, total shareable revenue was up by 152 per cent (171 per cent in the original DORA); national government’s share grew by 175 per cent (218 per cent in the original DORA) and the county share advanced by 71 per cent (77 per cent in original DORA).  Does this mean that national government is the faster growing entity in our system of devolution? 

For the record, the 2025 Division of Revenue Bill (DORB) allocates 14.2 per cent of projected 2025/26 shareable revenue to counties, and 85.3 per cent to national government (the balance is for the badly neglected and poorly resourced Equalisation Fund).  What’s wrong with this picture? 

Remember, officialdom will tell us, as is the case with the 2025 DORB, that the county share is 25.8 per cent of shareable revenue.  Not for 2025/26, but based on 2020/21 data, the last audited revenues approved by Parliament!  The 2024 DORA claims 24.7 per cent on the same lines! 

For the record, the current county share is based on 2020/21 audited revenues of Sh1.571 trillion, but we have audited accounts up to 2023/24 currently sitting in Parliament!  It would be interesting to see what sort of share counties would have received if our MPs were more diligent and less obsessed with CDF and other funds, but it is not unfair to speculate that the Sh3.3 trillion dished out since 2013 might have been at least 50 per cent more at around Sh5 trillion by now. 

Which leads us to a second data story.  The whole idea behind the use of a revenue rather than expenditure share for counties probably rested on the heroic assumption that the national government would be responsible in its borrowing. 

Yet, a quick calculation shows that total public spending since the advent of devolution, including this year, will probably hit Sh31 trillion.  At Sh3.3 trillion to date, are we saying that the effective county share to date is only 10.6 per cent? 

In effect, what’s happened is national government’s debt spree since 2013 — because debt is essentially future taxes — is a clever “pre-payment” of their equitable share, to the detriment of counties. 

Listen to our Treasury nabobs argue that our constitution (including devolution) is too expensive as a borrowing rationale.  Hence the radical suggestion by some that the shareable revenue calculation must include borrowing so counties may share these “revenues in advance”! 

Let’s close with our third story based on Treasury and Economic Survey date for three sectors to 2023/24. 

We find that we spent Sh851 billion on agriculture since 2013/14.  Of this total, national government spent Sh657 billion (77 per cent), while counties did Sh194 billion (23 per cent).  Agriculture is a devolved function.  On health, we spent Sh1.786 trillion, split between national at Sh862 billion (48 per cent) and counties at Sh924 billion (52 per cent).  Health is also a devolved function.  Then there’s transport, a concurrent function, that covers roads, rail, ports, air, marine and pipeline.  Sh3.1 trillion total; 2.744 trillion (88 per cent) national (including CDF) and Sh356 billion (12 per cent) counties.  Parastatals, you say, but doesn’t devolution come first? 

If this isn’t enough, don’t forget that the Intergovernmental Technical Relations Committee (IGTRC) recently costed transferable county functions still being run nationally at Sh272 billion across water, energy, tourism, and, yes, agriculture, health and transport. 

It should not surprise anyone that, as we walk into the 2025/26 financial year, the National Assembly has approved the county equitable share at Sh405 billion, the Commission on Revenue Allocation proposed Sh417 billion, the Senate thought it should be Sh465 billion and the Council of Governors wanted Sh536 billion.  

One stakeholder suggestion during DORB hearings by the National Assembly’s Budget and Appropriations Committee spoke to a three-year rolling average allocation at Sh441 billion. 

There are many other data stories we may consider, from cash flow predictability to revenue and financing potential, but it would seem that much of today’s picture reflects an unsettled division of labour between our national and county governments, and the hundred of agencies in between.

Yet this is not about next year’s allocations. 

It’s about a two-level devolved system of government where at the lower level (counties) the rubber meets the road and the higher level (national) is one of unified county interests as the national interest in a bottom-up, not top-down, fashion. 

Granted, counties could perform better in moving from project activities to actual service delivery outputs and household wellbeing and welfare outcomes.  Yes, local corruption is more easily evident than national corruption. 

But, we should have graduated from devolution’s nascent institutional and infrastructure building phases to investment and innovation.

Think economic blocs, fintech hubs, industry clusters, cultural hotspots, science parks and great sites of history. Counties are not payroll houses but service delivery points and economic development pivots. 

Sadly, the data question is clear — how does this happen with overbearing national government?