The cost of donor reporting for Ugandan NGOs

On the last Friday of a quarter, a finance officer at a mid-sized organisation in Gulu has the same three months of expenditure open in four places at once. One funder wants the figures by activity, the second by budget heading with staff costs grouped differently, the third in a currency the organisation does not bank in, and the fourth has returned a spreadsheet with last quarter's tabs still in it. Nothing about the programme has changed between the four windows — what changes is the shape each funder wants the money poured into, and pouring it is the afternoon's work.

Multiply that Friday across the country.

A field officer in Mbale re-codes the same household visits because the second funder counts a beneficiary differently. An executive director in Fort Portal signs an annual return whose figures already sit in three other offices. A community organisation's chairperson in Kabale carries audited books to a sub-county committee and then the same books to a district one, and none of them has produced anything that did not exist before they sat down.

The reporting exists for a reason, and the reason should come first. Money raised from one public to be spent on another has to be accounted for, and an organisation that cannot show where its money went has no claim on anyone's confidence. Uganda's statutory framework for this is neither unusual nor, on its own, heavy; what has grown heavy is everything stacked on top of it.

THE LAW ALREADY ASKS FOR MOST OF IT

Section 39 of the Non-Governmental Organisations Act, 2016 sets the floor. An organisation keeps accounting records, draws up financial statements within six months of its year end, and within two months of that tells the National Bureau for Non-Governmental Organisations whether it has complied with the Act. It files annual returns with audited books certified by an auditor, and declares its income and expenditure estimates, budget, work plan and sources of funds to the district technical planning committee and the district and sub-county committees of every area in which it operates.

The cash cost is modest: under the NGO (Fees) Regulations, 2017, filing annual returns costs UGX 50,000 for an organisation other than a community-based one, though drifting past permit renewal attracts one hundred currency points — UGX 2,000,000 in the same instrument — for every month in default. The multiplier is geography, not money: the duty attaches to the area of operation, so a programme in six districts lodges the same budget and work plan with six district committees and the sub-county committees beneath them.

AGREEMENT ON THE GOAL, DISAGREEMENT ON THE INDICATOR

Then the funders arrive with questions of their own, and this is where the cost is made. Uganda's Development Partnership Review, published by the Ministry of Finance, Planning and Economic Development with the United Nations Development Programme in 2020, counted 180 donors and implementers reporting aid transactions for the country in 2018 in IATI data. The Global Partnership for Effective Development Co-operation's 2023-2026 monitoring round, drawing on twenty partners active in Uganda, found that they and government want much the same results and measure them differently.

The distance between the first figure and the other two is the problem in a line. Agreement on the objective costs nothing to maintain; disagreement on the indicator is paid for in fieldwork, in forms and in somebody's evening. The same round records that most partners use indicators generated inside their own organisations, and that only 31 per cent of funding to Uganda's public sector ran through government public financial management systems. A 2017 Global Public Policy Institute study put it from the other end: duplication is less costly than the multiplicity of donor standards, indicators and frameworks.

Uganda has paid this bill once already. Kiyemba and colleagues, in African Health Sciences in 2023, compared the two national tuberculosis reporting systems that ran side by side across all 112 districts until 2018: in 2016 the parallel system recorded 42,570 incident cases against 48,434 in the ministry's DHIS2, a discrepancy of 12.1 per cent narrowing to 7 per cent the year after, and running both, the authors write, further strained the meagre resources available. Two systems did not produce two useful views of tuberculosis. They produced two numbers and a bill.

NOBODY HAS EVER COUNTED THE HOURS

The uncomfortable fact is that the cost goes unmeasured nearly everywhere it lands. The United Nations Joint Inspection Unit's 2017 review of donor reporting requirements found that organisations struggled to estimate what their reporting cost, that most do not track it separately at all, and that the difficulty stems partly from the absence of methodologies for computing it. Reviewing the harmonised reporting pilot two years later, the Global Public Policy Institute found no benchmark for what counts as a normal amount of time to write a report.

The few measured figures in circulation are not Ugandan. The International Council of Voluntary Agencies surveyed agencies in six countries in 2016 and found that, once informal requests were counted, they were filing a report every twenty-four hours, 40 to 59 per cent of it perceived as duplicative; the Norwegian Refugee Council put its own saving at 90,000 staff hours a year if its nine largest funders harmonised formats, cost categories, budgeting and procurement, a figure the same 2017 study carries. What travels to Kampala is the shape rather than the quantity, along with the finding that reporting on a small grant is proportionately the dearest reporting there is.

RWANDA DECIDED TO GRADE ITS FUNDERS

Rwanda named this cost in policy twenty years ago and built machinery against it. The Rwanda Aid Policy endorsed by cabinet in July 2006 records that donors place significant demands on government in time, reporting needs and other resources, commits it to moving partners onto the national system for evaluation and reporting so as to reduce the burden on both sides, and notes a development assistance database put in place to act as an interface between donors, non-governmental organisations and government.

What gave that teeth was the scorecard. Rwanda's donor performance assessment framework has graded each partner in an annual dashboard against targets for the share of aid recorded in the national budget and the share disbursed through government budget execution, financial reporting and procurement systems. Uganda's own 2020 review reprinted that dashboard as the example to follow, urged Uganda to publish individual donor scores yearly, and argued that partner strategies should follow the national plan rather than each donor's incompatible results framework.

WHAT IS THE WAY FORWARD?

Several moves are available and none asks anyone to report less. The Bureau and the district committees should treat the statutory return as one filing the lower tiers draw from, not a document delivered to every office the Act names. Funders should adopt a common core: that 2017 analysis of nineteen donor countries' templates found the thirteen most frequent questions cover 77 per cent of what donors ask, leaving genuine differences to a short appended section. The Ministry of Finance should publish a partner-by-partner score, as its own review proposed, because a habit with no name costs its owner nothing. Organisations should record the activity once, in a structure a report can be cut from, which is what the current generation of locally built field monitoring platforms is for. And somebody should commission the Ugandan measurement.

The arithmetic holds until somebody changes it. The finance officer in Gulu will open the same quarter in four windows again in December, and every hour she spends moving figures between them is an hour her organisation was funded to spend somewhere else.