Transfer Pricing in the UAE: Related-Party Rules That Apply to More Than You Think

I would call it the single most costly belief in the UAE SME tax market, and it comes from a reasonable place. Every headline number in the framework describes a large enterprise. AED 200 million for a Local File, AED 3.15 billion for a Master File, AED 100 million of controlled transactions for an APA.

What the belief misses is the distinction between the documentation threshold and the arm’s length obligation. Those are two separate things. The thresholds determine what formal records the FTA can compel you to produce. Article 34 applies to every related-party transaction of every taxable person, regardless of size and regardless of whether any threshold is triggered.

A business with AED 15 million of revenue, a mainland LLC and a free zone subsidiary has a transfer pricing obligation. The management fee between them must be at arm’s length. The intercompany loan must carry a commercial rate. Article 34 does not say “unless your revenue is below AED 200 million.”

The Second Version: “It Only Applies Cross-Border”

This one circulates because transfer pricing is historically associated with multinational profit shifting. In the UAE, the most immediately relevant exposures are domestic. Between a mainland entity and a free zone subsidiary, between two mainland companies under common ownership, between a business and its owner-director.

Domestic transfer pricing matters here because of the rate differential. A mainland entity sits at 9%, a QFZP at 0% on qualifying income. That is a nine percentage point difference on the same dirham depending on which entity earns it. It is exactly why the framework flags domestic transactions between differently taxed entities, and why they are eligible for domestic APAs.

This is the finding I hit most often in a first review of any multi-entity group. Not as an isolated error but as a structural assumption. Money moving between two entities under common ownership does not need a commercial basis, and if no interest is charged there is nothing to document. Both halves of that are wrong.

The pattern in my files is always similar. In the early years, when the second entity needed working capital, the first lent it money. The transfer was recorded as a loan, with no agreement drafted and no rate set. The owner thought of it as money moving within his own structure, which commercially it was, and the absence of interest felt appropriate because there was no counterparty to negotiate with.

Under Article 34 that reasoning does not hold. An independent lender providing funds for an indefinite term would charge interest reflecting the borrower’s creditworthiness, the term, the currency and market conditions. The absence of interest between related parties is a departure from arm’s length terms whatever the commercial rationale behind it.

Where the Cost Comes From

1. The deemed benefit. On an AED 3 million interest-free loan carried for three years at a market rate of around 7%, roughly AED 630,000 of interest should have been recognised. At 9% CT that is about AED 56,700 of additional tax, before penalties.
2. The QFZP dimension. If the free zone entity is the lender, the imputed interest on a loan to a mainland borrower is non-qualifying income. Recognising it could push non-qualifying income above the de minimis limit, and that costs the 0% rate for five periods.
3. The documentation penalty. An AED 10,000 penalty per violation applies to failure to maintain required records, independent of any tax adjustment.

My fix is specific and much cheaper done proactively. Draft a formal loan agreement setting out principal, rate, repayment schedule, currency and purpose, then benchmark the rate against market data, typically EIBOR adjusted for a credit spread. Calculate the interest that should have accrued since inception, and assess whether a voluntary disclosure fits the prior periods. The specific rate matters less than having a documented basis for it.

A mid-sized Abu Dhabi group operating a mainland LLC and a free zone subsidiary, combined revenue around AED 25 to 30 million. The client came for CT advisory ahead of the first filing deadline. The business was well run commercially. What the owner had not understood was that the financial arrangements between his two entities had been quietly accumulating a compliance problem since the entities were first connected.

Management fees flowed from the free zone entity to the mainland entity covering shared overheads and the owner’s split time. A loan of roughly AED 1.8 million had been extended in the early years when the free zone business needed working capital. None of it was unusual. All of it had happened informally, without documentation, and without anyone asking what the correct price should have been.

The free zone entity was claiming QFZP status, which is where the problem became consequential rather than merely technical. QFZP status requires that related-party transactions are priced at arm’s length. If they are not, the FTA has a basis to recharacterise them. For an entity whose income already carries the 0% versus 9% distinction, that is potentially a challenge to the qualifying income test itself.

What the Work Actually Involved

Three components, all completed before either CT return was filed.

First, I mapped every intercompany transaction across the prior three years. The management fee payments, the loan balance, and several smaller cost allocations that had been processed through the accounts without any formal characterisation.

Second, working from the accounting records, I established a pricing policy per transaction type. For the management fee I used cost-plus: identifying the actual costs the mainland entity incurred providing the services, adding a market-rate markup for a routine service provider, and benchmarking that markup against comparable UAE data. The resulting fee differed from what had been charged, higher in some periods and lower in others. A formal services agreement documented the scope, methodology and calculation basis.

Third, the loan. I applied an arm’s length rate to the outstanding balance, benchmarked against dirham interbank rates, then back-calculated the interest that should have accrued from inception. My adjustment was recognised in both entities’ accounts before the returns were prepared, with a formal agreement executed for the arrangement going forward.

Both entities ended with a defensible position: documented, benchmarked and supported by formal agreements. The QFZP position was protected because the management fee income was correctly priced and the loan interest correctly recognised.

The low-value-adding intra-group services safe harbour is the most consistently overlooked simplification in the UAE framework, and it applies to the single most common intercompany arrangement in the market.

Most intercompany service arrangements in Abu Dhabi groups are exactly what the name describes: finance and accounting support, HR administration, IT helpdesk, legal and compliance support, administrative and secretarial services. These are back-office functions that keep the group operational, not the activities that generate its profit.

Where the services qualify, a cost-plus markup of 5% is accepted as arm’s length without a benchmarking study. You identify the costs, apply the markup, charge that as the fee, and document the cost pool and the allocation basis. No database search, no comparable company analysis, no interquartile range calculation.

The criteria come from the OECD guidelines. The services must be ancillary and supportive, performed by one or more group members, available from independent parties, not the primary business activity, and not involving unique and valuable intangibles. For most shared administrative services in an Abu Dhabi family group, those are met without difficulty.

In my experience most mid-market businesses have never been told this exists. Their documentation either includes a full benchmarking study for transactions that would have qualified for the simplified approach, or has no analysis at all because the owner assumed documentation was only required at larger thresholds. Both are avoidable.

Before methods, before benchmarking, before engaging anyone to prepare documentation, I ask for one thing: every flow of value between your related entities and connected persons over the last twelve months. Not from memory. From the bank statements and the accounting records simultaneously.

Clients almost never think to ask for this, because they arrive with a narrative. “We have a management fee and an intercompany loan.” They expect the work to be about formalising those known arrangements.

My map almost always reveals that the narrative was incomplete. The director’s personal expenses paid by the company and never classified, the cost one entity absorbed for another because the invoice arrived at the wrong address, the staff member working across both entities whose salary sits entirely in one, the rent one entity absorbs for space the other uses.

All of those are potential transfer pricing transactions. Not all are material, but until the map exists nobody knows which are significant. It takes a day for a simple two-entity group and about a week for a complex structure.

The second benefit is less obvious. The map is the foundation for the disclosure schedules on the CT return. If it exists, populating them is a data exercise. If it does not, the schedules get built from the same incomplete narrative that created the exposure.

It is a mandatory schedule of everything you are required to disclose. I find the distinction sounds subtle and the practical difference is significant.

In my review of returns prepared without specialist support, the form captures what the owner thought of as transfer pricing relevant, usually the formal management fee and the intercompany loan, and omits everything else. The informal cost recoveries. The owner-director salary above the AED 500,000 threshold. The intercompany property use never formalised.

The consequence runs on two levels. Directly, an omitted disclosure that meets the threshold is a compliance failure attracting a penalty independent of any TP adjustment. Indirectly, it creates a visible gap. Accounts showing AED 800,000 of management fee payments against a form showing AED 600,000 leaves a AED 200,000 discrepancy in the FTA’s data, and the gap itself triggers a query.

The detail I see producing this most often is owner-director remuneration. In most mid-market Abu Dhabi businesses the total package of salary, housing, transport and bonus exceeds AED 500,000 and therefore meets the connected-person threshold. The packages are genuinely arm’s length in most cases. The issue is that nobody identified them as requiring disclosure, because the words “transfer pricing” do not intuitively connect to a salary. The FTA’s Transfer Pricing Guide sets out what the schedules require.

A management consulting firm, Abu Dhabi mainland LLC, single owner-director who is also the primary fee earner. Revenue around AED 3.8 million. The owner drew a combined package of salary, housing, transport and a discretionary bonus totalling roughly AED 720,000, which put aggregate connected-person payments above the AED 500,000 threshold and triggered the Connected Persons Schedule.

The question the disclosure forces into view is whether AED 720,000 is arm’s length for someone combining managing director, principal consultant and business development lead at that revenue level.

My analysis used two reference points. Published salary surveys for senior director-level roles in UAE professional services showed a range of roughly AED 550,000 to AED 850,000 for equivalent scope. Separately, the owner billed around AED 180 an hour for his consulting time at roughly 60% utilisation, giving about 1,200 billed hours and a market value near AED 216,000. The full package was defensible once the whole director role was benchmarked, not just the fee-earning part.

The bonus needed its own treatment. AED 120,000 paid in the final month of the year raises a fair question about whether it is remuneration or a year-end profit extraction. The documentation tied it to a specific client win in the final quarter, showed the amount was proportionate to the incremental revenue, and identified comparable bonus structures in the sector.

Without that, the schedule discloses AED 720,000 of connected-person payments with no supporting analysis, which is an invitation for a query. With it, the disclosure comes with contemporaneous documentation that answers the question before it is asked.

This is the tip I would give above all others, and it determines whether your documentation is contemporaneous or reconstructed. That distinction is the difference between a defensible position and an indefensible one.

The thirty-day window the FTA allows for producing documentation is not the production window. It is the delivery window for documentation that already exists. A report prepared in 2026 for the 2024 period, in response to a query, is not contemporaneous documentation. It is a reconstruction, and an experienced reviewer can often tell from the metadata, the version history and the dates on the intercompany agreements.

So every TP-relevant decision needs a dated document at the time it is made. The management fee rate agreed for the year, with the agreement executed and dated before the first invoice. The loan rate set, with the agreement signed before the first drawdown. The cost allocation methodology documented before the first allocation is processed, and the annual benchmarking review completed before the year-end statements are prepared.

None of these need to be elaborate. A one-page services agreement with a date and two signatures is more evidentially valuable than a twenty-page report prepared twelve months later. The first proves the arrangement was deliberate and contemporaneous; the second proves it was documented, but not when.

The FTA began accepting domestic unilateral APA applications in December 2025, under the APA Corporate Tax Guide, reference CTGAPA1. An APA is a binding agreement fixing the pricing methodology for specified controlled transactions over three to five tax periods. The FTA commits not to challenge the covered pricing provided you comply with the agreed terms.

The primary eligibility indicator is AED 100 million of controlled transactions per period, which puts it beyond most SMBs in absolute terms. I see it as directly relevant for family conglomerates with significant intercompany volumes, and for businesses with material domestic transactions between a mainland entity and a QFZP.

That last case is where an APA is most valuable in a way specific to the UAE. Take a QFZP whose status depends on the arm’s length pricing of its management fee income, where the fee is both the primary income and the test of whether that income qualifies. The uncertainty compounds every period the arrangement continues unresolved, and an APA removes it for the covered period. At AED 30,000 for a new application, that is modest against five years of certainty.

Processing currently runs somewhere in the range of twelve to twenty-four months, and the framework covers prospective periods only, with no rollback to prior years. A business starting in mid-2026 is looking at a concluded agreement in 2027 or later.

UAE transfer pricing requires a systematic approach to ensure arm’s length compliance and avoid penalties. This five-step process guides you from mapping related-party transactions to submitting your annual corporate tax filing uae. By establishing clear documentation and benchmarking, you can protect your business from the uae corporate tax penalty framework and maintain defensible financial practices year after year.

The second filing cycle created a consistency requirement. The FTA now holds two years of return data and compares them: revenue trajectories, margin profiles, related-party disclosure changes. A business that disclosed specific transactions in year one and changed the methodology in year two without documenting why has an unexplained variance in its history. The FTA does not need to challenge the second position directly; it can ask why the first changed.

E-invoicing makes intercompany flows visible in real time. Mandatory from January 2027 for businesses above AED 50 million. A management fee invoiced at AED 250,000 monthly for three quarters and then adjusted to AED 350,000 in the fourth will be visible with the adjustment evident. An interest invoice appearing for the first time in year two, when none was charged in year one, will be visible without an audit being necessary.

Benchmarking Is an Annual Job Now

Benchmarking needs an annual review, not a three-year cycle. The FTA uses the interquartile range as the arm’s length range and may adjust a result outside it to the median. Market conditions have moved materially since 2023, so a study prepared then and applied unchanged to 2026 transactions may be producing results outside the current range without anyone noticing.

One correction worth making, because in my reading it gets stated the wrong way round. The penalty rates applying to a Corporate Tax adjustment come from Cabinet Decision No. 75 of 2023, not from Cabinet Decision No. 129 of 2025, which harmonised the VAT and Excise framework toward the Corporate Tax rates that already existed. On the rates themselves: a proactive voluntary disclosure carries 1% per month on the additional tax, while an FTA-discovered adjustment carries 15% fixed plus 1% per month.

The voluntary disclosure advantage is real and it compounds with the size of the adjustment. That matters for transfer pricing specifically, because TP adjustments operate on the whole quantum of an intercompany transaction rather than on a single line item.

I should be straight about this. I have not yet sat through a formal FTA transfer pricing audit with a client. UAE Corporate Tax is young enough that TP-specific audit activity at the SMB level is only now developing in volume.

What shaped how I approach the work was not a single regulatory interaction. It was reviewing an existing TP report prepared by a previous consultant for a large group. The report was formally correct: functional analysis, method selection, benchmarking data, conclusions, with the markup sitting inside the interquartile range. On paper, defensible.

What was missing was substance behind the description. The functional analysis described functions generically, strategic oversight and financial management and procurement support, without documenting what they actually involved for that group in that year. How many people performed them, what their time split was, what evidence existed that the functions were performed at the level the fee implied.

An FTA reviewer would not need to challenge the benchmarking. They could challenge the factual basis. A functional analysis describing what an entity is supposed to do, rather than what it did, looks like TP documentation without performing the function it exists to perform.

So my question before any report is finalised is simple. If an inspector sat down with this document and the underlying financial records together, could they trace every claim in the analysis to something observable in the business? If not, it needs more work regardless of whether the benchmarking is correct.

In my experience most Abu Dhabi businesses with intercompany transactions are not pricing incorrectly. The management fee is roughly what the market would charge, the interest rate is within a defensible range if anyone looked, the cost allocation would survive benchmarking. The underlying transactions are commercially sensible, which is why they were structured that way.

The problem is that nobody wrote it down. And nobody wrote it down because for the entire history of the business before June 2023 there was no reason to. What I am usually looking at is the residue of rational behaviour in a pre-tax environment.

Which changes what the work actually looks like. For most mid-market businesses this is not complex analysis, it is retrospective documentation of arrangements that were commercially correct all along. My benchmarking usually confirms what the owner already knew, and the agreements formalise what was always the commercial intent. A proportionate, systematic exercise.

Doing it is not an admission of a problem, it is the opposite. A business that has built the map, the agreements, the benchmarking and a complete disclosure form responds to an information request by retrieving a document. A business that has not responds with thirty days and no starting point. Those two outcomes are not the same.

If you are carrying intercompany arrangements you have never tested against the arm’s length standard, I would look at them while the correction is still proactive. Book a free consultation with AH Chartered Accountants in Abu Dhabi.