How URA plans to catch fake intra-group charges

A bank, a parent company, and a number with 12 figures in it, is the outline of Uganda’s biggest live transfer-pricing fight.

Stanbic Bank and its holding company are contesting a Shs117.8b assessment, filed by KAA in June 2025 in the Tax Appeals Tribunal.

The dispute is, at bottom, an argument about a franchise fee, specifically, whether parts of it were really payment for genuine support, or whether the parent was charging Uganda twice for things it should have absorbed itself, or things it had already billed elsewhere.

Stanbic says the arrangements reflect legitimate cost-sharing. URA disagrees. It would be easy to read that as a peculiarly Ugandan argument. It isn’t.

It is the local branch of a fight the Organisation for Economic Co-operation and Development (OECD) has been trying to settle since intragroup service charges first became a recognisable category of dispute.

What makes this particular moment unusual is timing: the global rulebook on exactly this question is being rewritten right now, in Paris, while Kampala’s tribunal is mid-argument on the very issues the rewrite addresses.

Old rules, new invoice

On June 1, 2026, the OECD opened a public consultation on Transfer Pricing Guidelines, the chapter governing intragroup services, with comments due by July 22, and a follow-up discussion at the OECD’s Paris conference centre in November.

OECD says the draft seeks to ‘update and modernise existing provisions,’ not to rewrite the underlying principles.

The plan is to do most of the reframing in four shifts.

First: it’s not enough to show the service happened. You now have to show that, before you spent the money, there was a good reason to think it would help.

Before, you just needed to show, but now you need to show why you paid for a service in the first place.

Second: the way you split a shared cost between different parts of a company has to actually match who benefits, not just be a rough guess.

Splitting it equally used to be fine. Now you need to split it based on who actually benefited more.

Third: when one activity does two different jobs, you can’t lump them all together anymore; you have to separate the parts.

Fourth: it’s not about looking good on paper anymore, it’s about proving you actually thought it through at the time.

Before, you mainly needed to just justify the pricing, but now, you need real proof for the pricing before spending.

The suggestions are a shift that has already generated debate.

In a Transfer Pricing Insight, Ann Barnshaw Kengaaju, a senior transfer pricing advisor at BDO East Africa, notes that the benefit test ‘is evolving into a business case assessment.’

In practice, she argues, this moves the burden of proof away from tax departments constructing justifications after the fact, and onto businesses themselves, who generate the relevant evidence months or years earlier, simply by requesting and approving services in the normal course of operations.

The wider implication, in her assessment, is that OECD’s revisions place far more weight on whether a service can be shown to have genuine commercial substance, rather than on whether the supporting paperwork looks technically correct.

‘Audits and disputes going forward, she suggests, ‘will turn less on how polished a transfer pricing file is and more on whether a multinational or local group can actually demonstrate why a service was needed, who it benefited, and what value it was expected to deliver.’

No opt-out clause

Most jurisdictions get to watch this process from a comfortable distance and decide later whether to adopt whatever the OECD finally agrees. Uganda doesn’t have that luxury, structurally.

The Income Tax (Transfer Pricing) Regulations reference OECD guidelines ‘as updated from time to time.’

Whatever eventually emerges from this consultation becomes law automatically, with no statutory instrument and no implementation lag.

The discussion-draft window closing on July 22 is, for taxpayers, the only point at which the rule can still be shaped before it simply applies.

URA, for its part, has been building the institutional muscle to use rules like these. Its use of cross-border information exchange, which enables it to interrogate intercompany arrangements, grew from two requests in 2012 to more than 170 between 2014 and 2022, recovering upward of Shs259.9b in that period.

‘Transfer pricing is a sensitive area globally,’ John Musinguzi, URA’s Commissioner General, has previously said.

The Tax Appeals Tribunal has, fortunately, already established the basic posture it needs to take.

In the Total E and P Uganda case in 2021, the ruling placed the burden squarely on the taxpayer to prove that a service actually delivered economic benefit, rather than on URA to disprove it.

OECD’s new benefit test simply formalises, globally, a standard URA that was already applying.

Different neighbourhoods, different speeds

Kenya is some distance ahead of Uganda on procedure, if not necessarily on outcome.

Kenya Revenue Authority (KRA) issued dedicated intragroup-services guidelines in 2021, moved to a full OECD-aligned three-tier documentation framework in 2023, and, critically, got advance pricing agreements into its Finance Act 2025.

Tanzania and Rwanda sit somewhere in the middle of the regional pack; South Sudan and DRC are still building the basics.

There is no East African Community-wide transfer-pricing framework tying any of this together, which means a bank or telecom running shared services out of a regional hub is, in effect, negotiating five separate versions of the same argument.

Across the wider continent, the constraint is the infrastructure.

The African Tax Administration Forum and OECD ran joint capacity-building workshops reaching over 10 tax officials in more than a dozen African countries between 2025 and 2026, with the recurring complaint being a shortage of reliable local comparables.

That scarcity cuts an odd, double-edged way under the new rules. A test built around documented business rationale rather than benchmarking arithmetic is, in principle, easier to apply where comparable-company data is thin.

It is also a test that rewards groups with disciplined internal paperwork and punishes those without it.

Show your working

None of this means management fees are wrong. Things like a dedicated team running IT systems, treasury management, and/or risk analysis across different countries are ideal.

But the new rules want to change how you prove. Before, you could basically say, ‘look, this is the normal market price, here’s a study showing other companies charge similar amounts,’ and that would get you through an audit.

However, under the new rules, you will need everyday proof that the service was planned, delivered, and made at the time. The rules will also require the head office sending the bill to have better record-keeping.

Leave a Reply

Your email address will not be published. Required fields are marked *