Article 223: How Treasury has turned emergency exit into a highway

Controller of Budget Margaret Nyakang’o. [Elvis Ogina, Standard]

When the Controller of  Budget (COB), Margaret Nyakang’o, appeared on national television last week, her remarks were treated by mainstream media as a series of shocking revelations.   First, she has no “visibility” over “funds and levies” that do not interact with the Consolidated Fund that parameterises her mandate.  In simple English, there are parts of the publicly funded budget that are beyond the control of the Controller of Budget responsible for public funds.

Second, increased weaponisation of Article 223 of the Constitution through National Treasury means that what was originally designed as a narrow escape valve for urgent and unforeseen crises, like a famine or a national security emergency, has been converted into a regular, off-budget financing pipeline.  And it isn’t spending that wasn’t foreseen; it is highly predictable stuff. By the time these items are brought to Parliament for retroactive “regularisation”, commitments are complete, funds have been spent, and the legislature is presented with a fait accompli.

Actually, if you want to predict the scale of this abuse in advance, simply refer to the Sector Reports presented earlier in the budget preparation cycle.  In these reports, MTEF sectors and individual MDAs present resource bids far in excess of the resource ceilings to which they are eventually restricted.  But, as said before, these MDAs do not cut the planned activities they have promised the public to fit with the lower ceilings,  thus creating space for “planned emergencies”. 

As a rule of thumb, consider maybe a third to a fifth of this bid-ceiling shortfall comes right back.   And then remember Article 223 abuse comes long before we see actual Supplementary Budgets.   Don’t forget the 10 per cent allowable excess spending rule now applies to the entire budget, not individual votes, making it easier to switch this excess from priority needs to political whims.

It was not without irony that Dr Nyakang’o noted that “we have ensured that the budget complies with the law, which does not mean the spending will comply with the law”.  Mic-drop right there!  Especially when this ongoing fiscal drift might accelerate exponentially as we head into 2027. After all, you need money to maintain a vast transactional network of patronage for political survival.

But it wasn’t just these obvious highlights that will raise eyebrows.  For example, the fact that the COB can only confirm the veracity of a spending approval (e.g. confirming that approval to buy a pen actually bought a pen) through post-facto expenditure returns rather than real-time tracking based on IFMIS access calls into question the utility of budgetary control if it’s simply a post-mortem. 

Essentially, while Article 228 of the Constitution mandates the COB to authorise (and report on implementation) every single withdrawal of public funds, the lack of live, transaction-level visibility reduces this office from an independent gatekeeper to a blindfolded bookkeeper receiving data supplied by the same institutions that the office is supposed to be policing.

Then there’s the revenue question.  We heard about the lack of visibility over funds and levies (the tip of the iceberg) but not the huge, parallel revenue engine that is the e-Citizen portal through which internally-generated non-tax revenues - Appropriations in Aid (A-in-A) - flow.  This centralisation under e-Citizen sought to “seal revenue leakages” as the Executive claimed.

Yet, as the Auditor-General (another independent post-mortem office) repeatedly flags, we have billions of shillings flowing through unreconciled commercial settlement accounts managed by private intermediaries outside the immediate oversight of the Consolidated Fund.  This represents the ultimate loophole: a dual exchequer that extracts from the public through digital means, yet remains analogue in insulating itself from independent transaction-level scrutiny. 

It is highly instructive that COB reports exclude A-in-A to the extent that it doesn’t flow into the Consolidated Fund.  As said, part of the budget is out of the control of the budget controller.  Hard pejoratives like “financial black hole” and “off-grid digital dark pool” easily come to mind.

There was much more that the TV interview covered; for example, the lack of official transparency in Treasury regarding the hidden fees, costs and charges in our “low-cost” external debt.  Yet one impression you get from our independent PFM institutions and offices like COB (and, say, the Commission on Revenue Allocation (CRA)) is that, rather than seeing themselves as counterweights, a fourth arm of government, they are happy to bifurcate between a “big brother lens” on the 47 county governments, and “velvet gloves” for national government.  Is this a colonial hangover where officials view counties as “the natives” and national as “the elites”?

In COB’s case, by example, why do counties suffer far greater budget scrutiny than the national government?  Why are the national government’s budget estimates “box ticked” while those for counties are “audited to the final zero”?  We could ask the same question of CRA, which seems more eager to pursue county revenue enhancement when its Article 216 mandate calls for them to “define and enhance the revenue sources of the national and county government”  (so, it's from them, and not National Treasury as the Ministry of Finance, that tax/revenue policy should emerge?).

That’s before we connect the vital “whole of government” fiscal responsibility mandates of COB (in practice through implementation reporting) and CRA (as specified in the Constitution).  But let’s conclude our story today by taking a helicopter view of our Public Finance Management (PFM) architecture, and offering three closing reflections to take forward, working top-down.

First, we haven’t quite established PFM as a continuous governance loop, what I call the PPBIR (or Policy to Results) cycle of policy/programming-planning-budgeting-implementation-results.

Very simply, policy and programming must inform planning; planning must dictate budgeting; budgeting must guide implementation; and implementation must yield results evaluated through monitoring, evaluation, reporting and learning (MERL) and auditing.  At COB level, this failure manifests as spending approvals based on budget “space” rather than workplan-driven data; for example, how do you stop an exchequer request for half the volume of an item at twice the price?

Second, the question of systems integration and functionality, including and beyond IFMIS.  There are two parts to this.  The first on integration must address the reality of parallel systems as described earlier.  The second, on functionality, comes from the fact that we simply lack the gold standard South Korean, or even Rwanda’s homegrown, “PPBIR” system that adds value to the work of institutions like COB.  It is an open secret that our text-based “non-financial reports” (including NIMES) have near-zero links to financials, hence our ad-hoc performance reporting.

Beyond these technical reflections is the PFM political one: the case for the independent National Treasury, as separate from a Ministry of Finance, as an independent, professional state organ serving the entire republic, including the 47 counties, with equal fidelity.  By collapsing the two into one in the 2012 PFM Act, did we create a conflict of interest where the apex political actor who spends money is the same actor who polices the purse?   More bluntly, did we create a PFM architecture of deception, and was this by design or default? These are not easy questions.

And they might even seem irrelevant until you remember where Article 223 shenanigans begin; why Treasury and COB reports don’t tally, or why COB and CRA as two independent PFM, or quasi-Treasury,  institutions, kowtow to the same Treasury that reports only to National Government, while acting “nyapara” with our counties.  One takeaway from this televised COB interview might be that it gives us a moment to reflect on the PFM architecture we truly need.  A big part of this reflection must seriously explore what an independent Treasury might look like.