From “I didn’t Know” to “You should have known”: UAE Article 54(bis) and Supply-Chain VAT Risk

The new UAE rule: structure and legal mechanics

1. The amendments introduce three related elements into Article 54.

1.1. Actual knowledge and mandatory denial

Pursuant to Clause 1, the FTA must reject deduction of recoverable input tax if it establishes that:

1) the supply for which input tax is claimed formed part of a supply or chain of supplies related to tax evasion, and
2) the taxable person was aware of this link at the time of deducting input tax.

1.2. Constructive knowledge and discretionary denial

According to Clause 2, the FTA may reject input tax deduction if it establishes that:

1) the supply formed part of a supply or chain of supplies related to tax evasion, and
2) the taxable person should have been aware, based on the circumstances of the supply, of this link.

1.3. Due-diligence presumption (failure to verify is deemed awareness)

The law further provides that a taxable person “shall be considered to have been required to be aware that the supply was part of a supply or a chain of supplies related to Tax Evasion, if he did not verify the validity and integrity of the supplies … before deduction of Input Tax, in accordance with the measures, procedures and conditions determined by the Authority in this regard”.

2. Several points follow from this structure.

2.1. The core right to deduct is now expressly conditional on risk management.

Before the amendment, the conditions for input tax recovery in the UAE were relatively orthodox: being a taxable person, having a valid tax invoice, using the supplies to make taxable or certain exempt supplies, and so on. The new wording turns risk management and supplier due diligence into a legal condition for input VAT recovery, at least in situations where the supply chain is tainted by tax evasion.

Practically, that means that input tax is no longer only about “what did you buy and how did you use it?” but also about “what did you know, and what did you check, about your counterparty and the wider chain?”

2.2. The expression “part of a supply or a chain of supplies related to Tax Evasion” introduces a deliberately broad scope, enabling the FTA to look beyond the immediate supplier–customer relationship and assess the integrity of the entire transactional chain.

The wording does not limit the relevant misconduct to the immediate supplier. The FTA may deny input tax where the supply forms part of a chain of supplies related to tax evasion. This mirrors the European experience with “carousel” and supply-chain fraud: it is enough that the chain in which the taxable person participates is tainted somewhere up- or downstream, “of which that taxable person had … knowledge” or  “means of knowledge”.

The amendment does not define “tax evasion” in Article 54(bis) itself, but the term sits against the background of the Tax Procedures Law and the general criminal-law concept of deliberate, unlawful underpayment of tax.[1]

2.3. The dual test (“was aware” / “should have been aware”) is directly recognizable from the case law of the Court of Justice of the EU (CJEU) in Kittel and its progeny. In Kittel, the Court held that where, in view of objective factors, the supply is to a taxable person who “knew or should have known” that by his purchase he was participating in a transaction connected with fraudulent VAT evasion, the national court must deny the right to deduct.[2]

The UAE legislator is thus not inventing a new anti-fraud doctrine but codifying that logic in the primary text of the VAT Law, with one important additional step discussed below.

2.4. The due diligence presumption: a step beyond the CJEU

The third limb (deeming a taxable person as having been “required to be aware” if they failed to verify the validity and integrity of supplies) is distinctive.

In the EU, the CJEU has repeatedly rejected the idea of strict liability that would make the right to deduct conditional on systematically carrying out a prescribed checklist of due-diligence steps. In Mahagében and Dávid the Court explicitly held that imposing a strict obligation to carry out checks as a precondition to input tax deduction is incompatible with the Directive, as long as the taxable person cannot reasonably suspect fraud.[3]

The UK’s HMRC manuals,[4] reflecting that jurisprudence, acknowledge that while failure to carry out due diligence may be evidence that a trader should have known of fraud, EU law does not permit a pure checklist-based strict liability regime.

By contrast, the UAE amendment:

• explicitly ties input tax denial to non-compliance with FTA-prescribed checks, and
• appears to treat failure to verify as a deeming condition (“shall be considered…”), not just as an evidentiary factor.

In other words, once the FTA issues its procedures, non-compliance will itself put the taxpayer into the “should have known / required to be aware” box, irrespective of whether, on the specific facts, the taxpayer could realistically have uncovered the fraud. That is a stricter formulation than the one prevailing in EU law.

[1] Flashback on ECJ Cases C-439/04 (Axel Kittel v Belgian State) & C-440/04 (Recolta Recycling SPRL) — “Right to deduct VAT may be refused if participating in VAT fraud”, VATUpdate, 3 July 2021.

[2] Axel Kittel v Belgian State (C-439/04) and Belgian State v Recolta Recycling SPRL (C-440/04). Judgment of the Court (Third Chamber) of 6 July 2006; Document No. 62004CJ0439, paras 56, 61.

[3] Mahagében Kft v Nemzeti Adó- és Vámhivatal Dél-dunántúli Regionális Adó Főigazgatósága (C-80/11) and Péter Dávid v Nemzeti Adó- és Vámhivatal Észak-alföldi Regionális Adó Főigazgatósága (C-142/11). Judgment of the Court (Third Chamber), 21 June 2012; Document No. 62011CJ0080.

[4] VATF53405 - The Kittel principle intervention: Kittel in more detail: What is meant by ‘knew or should have known’: Introduction

International practice: how close is the UAE to the EU and UK?

3. Having set out the mechanics of the new UAE rule, it is helpful to step back and ask how unusual this approach really is in comparative terms. Outside the GCC, the most mature reference point for dealing with VAT fraud in supply chains is the experience of the European Union and, by extension, the United Kingdom’s pre-Brexit practice. For nearly two decades, the EU courts and tax administrations have been refining a doctrine that allows the right to deduct input VAT to be refused where the purchaser knew or should have known that it was participating in a fraud-tainted chain.

4. The UAE’s new Article 54(bis) regime fits squarely into this international conversation. In substance, it adopts the same organizing idea as the EU’s so-called Kittel doctrine, but it does so in a more explicit and codified way. To understand the extent of that alignment, it is worth recalling the essentials of the EU case law and the balance it strikes between combating fraud and preserving VAT neutrality.

The EU (Kittel, Mahagében and the balancing act)

5. The EU right to deduct is anchored in Articles 167 and 168 of the VAT Directive. Those provisions do not mention fraud chains or due diligence.[1]

The CJEU filled that gap in Kittel, where it held that:

• if the supply is real and the formal conditions are met, the right to deduct normally cannot be refused,
• unless it is established, on objective factors, that the taxable person knew or should have known they were participating in a transaction connected with VAT fraud, in which case the right to deduct must be refused.

In the later cases of Mahagében and Dávid, the Court attempted to balance fraud prevention with the principle of VAT neutrality:

• tax authorities cannot refuse deduction solely because the supplier committed irregularities or failed to remit the tax;
• authorities may refuse deduction where, in light of objective circumstances, the taxable person knew or ought to have known they were participating in a fraudulent chain;
• Member States may not impose disproportionate due-diligence obligations that effectively make input tax deduction conditional on exhaustive investigations into suppliers if there is no indication of fraud.

The EU model therefore rests on three pillars:

1) A neutral right to deduct as the rule.
2) Kittel-type denial where knowledge or constructive knowledge of fraud is proven.
3) A prohibition on turning due diligence into strict liability.

National implementations

6. EU Member States implement this case law in various ways.

Belgium writes a “knew or should have known” test directly into provisions on joint and several liability for VAT, targeting buyers in fraud-tainted sectors.[2]

The UK in the pre-Brexit EU practice made the Kittel doctrine the centerpiece of its approach to VAT fraud. HMRC’s VAT Fraud Manual explains that HMRC may deny input VAT recovery where the trader knew or should have known that its transactions were connected with fraudulent VAT evasion anywhere in the chain, even if the fraud is by another party.[3]

7. In Section VATF53410 of the VAT Fraud Manual, HMRC introduces the “knew or should have known” test and signposts the sub-sections that deal with its components. For corporate entities, knowledge is not analyzed as the purely subjective state of mind of one individual, but as something that may be attributed to the company as a whole, taking into account its governance and decision-making structures. In this section HMRC explains whose knowledge – or ability to know – within the organization is relevant when applying the Kittel principle.

Drawing on the High Court judgment in Livewire Telecom Ltd & Olympia Technology Ltd ([2009] EWHC 15 (Ch)), HMRC notes that:

“the Kittel [principle] applies to the taxable person. The taxable person was Olympia (the company). The question therefore for the tribunal was not what a director of Olympia knew or ought to have known, but what the company itself knew or ought to have known. The knowledge of a director of the company may, to be sure, be attributed to a company, but there may be other knowledge (for example that of a senior employee) which, on the facts ought also to be attributed to the company”.[4]

“In applying the correct legal test, the tribunal must consider not only the knowledge that should be attributed to the company via its directors, but also the knowledge that should be attributed to the company via its senior employees. In this context, knowledge includes both knowledge of facts and the ability to evaluate those facts and to draw appropriate conclusions from them”.

On that basis, VATF53410 instructs that, that “when determining who ‘knew or should have known’ in relation to a corporate entity you should take into consideration the knowledge of the officials of that corporate entity (director(s) and company secretary(ies)) as well as employees (in any capacity) and third parties (agents, advisors etc) who might have conducted or assisted in the transaction(s)”.

8. In respect of the ‘third party’, HMRC in VATF53440 refers to the decision of the Upper Tier Tribunal (UTT) in the case of Greener Solutions Ltd (([2012] UKUT 18 (TCC)). The Tribunal considered whether the knowledge of a third-party “deal consultant” could be attributed to the company for Kittel purposes. Greener Solutions Ltd (GSL) had engaged Mr Murray, the sole director of another company, to run a “trial transaction” in mobile phones in return for a share of the profits. GSL characterised him as a consultant, not an employee or agent, and argued that it remained in overall control while Mr Murray handled the day-to-day execution. HMRC denied input VAT on the basis that the deal was connected with fraudulent evasion and that GSL knew or should have known of that fact. GSL’s defence was that any knowledge of fraud lay with Mr Murray, not with the company.

The First-tier Tribunal initially accepted that Mr Murray’s knowledge should not be attributed to GSL, but the Upper Tribunal reversed that finding. Mr Justice Warren held that Mr Murray’s knowledge was to be attributed to the company for the purposes of Kittel: in substance, the transactions were carried out on GSL’s behalf and the company stood to benefit significantly from them. So, it could not escape the consequences of knowledge held by the person it had chosen to conduct those transactions. To allow otherwise would undermine the fraud-prevention purpose of the Kittel doctrine.

HMRC distils this in VATF53440 as a general rule of attribution: when applying the “knew or should have known” test, the decision-maker should consider the knowledge of the taxable person’s officials, employees and relevant third parties (agents, advisers, consultants, intermediaries) who have conducted or assisted in the transaction(s), and that knowledge may, in appropriate circumstances, be attributed to the company itself.

9. The HMRC Manual then develops the “should have known” limb through a series of refinements:

• VATF53425 explains the “only reasonable explanation” test: if, on the facts, the only reasonable explanation for the transaction is that it is connected with VAT fraud, a trader “should have known” of that connection, even if some other theoretical explanation exists.

HMRC provides non-exhaustive list of examples to illustrate ‘indictors’ that would have suggested to the taxable person that the only reasonable explanation was that the transaction was connected with fraudulent evasion of VAT. This list includes:

i) “unsolicited approaches from or to organisations with little or no history in the market offering a guaranteed profit on high-value deals”;

ii) “repeat deals at the same or a lower price and consistent profit”;

iii) “unsecured loans with unrealistic rates of interest”;

iv) “instructions to make payments to third parties or off-shore, normally for less than the full price (and often less than the VAT invoiced)”.

• VATF53430 focuses on “general awareness”. This section explains that HMRC’s “should have known” analysis is informed by the taxable person’s awareness of VAT fraud in general and in the specific market in which they trade, and that this awareness must be evidenced and documented. The Manual gives non-exhaustive examples of what HMRC should look at:
  (i) press and trade coverage and wider public discussion of VAT fraud in the relevant sector;
  (ii) any warnings (written, telephone or visit-based) given to the trader about the risks and typical features of fraud;
  (iii) records showing the trader acknowledged those risks;
  (iv) membership of trade bodies that have highlighted VAT fraud or engaged with HMRC’s anti-fraud measures;
  (v) the frequency and extent of prior HMRC notifications that the trader’s transactions were traced to fraud; and
  (vi) the involvement of key personnel in other businesses previously found in fraud-tainted chains.

All of these elements are used to build an objective picture of whether the trader should have been aware that its own transactions carried obvious fraud indicators.

• VATF53435 deals with “contrivance”, indicating that certain contrived features of a transaction can demonstrate that the trader should have known that it was participating in a fraudulent chain.

Some of the contrivance features detailed in VATF60000 can also be used to show that the taxable person ‘should have known’. However, not all of the features will be relevant. For example, HMRC “would not expect a taxable person to know about the uniformity of the mark ups applied by taxable persons further up/down the transaction chain. However, they would of course know about such things as inadequate insurance, contacts etc. in relation to their own transactions. Factors such as these may help to establish that their transactions were ‘too good to be true’ and that the only reasonable explanation was that they were connected with fraud”.

What emerges from these sections is that, in UK practice, “knew or should have known” is fleshed out through a structured analysis of corporate knowledge, the objective commercial explanation (or lack of it), the trader’s general market awareness and the presence of contrived features in the transactions, i.e. not through a formal statutory list of mandatory checks.

10. So, even in robust anti-fraud jurisdictions, the emphasis is still on proving a degree of knowledge (actual or constructive) on the facts of the case and using “due diligence” as an evidentiary factor, not as an automatic trigger. HMRC’s guidance, for example, cites Mahagében to confirm that the EU courts consider “strict liability to carry out due diligence checks as a precondition to deduct input tax” unacceptable, and frames the question as whether the trader took reasonable steps in light of identified risks.[5]

11. It is important to stress that, properly read, neither the CJEU nor UK case law turns “due diligence” into a free-standing legal obligation on the taxable person. In Mahagében, the Court held that the deduction of VAT cannot be refused merely because the issuer of the invoice has committed irregularities, and that the tax authority may not shift the entire investigative burden onto the taxable person. According to the Court, VAT Directive precludes national practice that requires the trader to systematically verify its suppliers’ compliance or to prove that it has done so as a precondition for exercising the right to deduct.

12. HMRC’s own guidance acknowledges this limit. In VATF36180, HMRC expressly notes that, in Mahageben kft & Peter David, the CJEU found “the imposition of a strict liability to carry out due diligence checks as a precondition to deduct input tax was unacceptable”. In other words, EU law (and, by extension, orthodox UK practice) accepts that “due diligence” can be used as evidence when deciding whether a trader “knew or should have known”, but it does not support the idea that failure to follow a prescribed set of checks automatically destroys the right to deduct. That is precisely where the UAE’s Article 54(bis) (through its Clause 3 presumption) goes a step further, by treating non-compliance with FTA-mandated verification procedures as itself sufficient to place the taxable person in the “should have been aware” category.

[1] Mahagében

[2] VATUpdate, 3 July 2021.

[3] VATF53420 - The Kittel principle intervention: Kittel in more detail: What is meant by ‘knew or should have known’: Extent of what a taxable person 'should have known'.

[4] Livewire Telecom Ltd & Olympia Technology Ltd ([2009] EWHC 15 (Ch)), para 123.

[5] VATF36180 - What to consider prior to determining whether to use an intervention: matters to consider when looking at particular types of taxable person or activity: labour providers: approach to due diligence.

Where the UAE now sits on this spectrum

13. Compared with this international landscape, the UAE aligns with the EU on the core idea that participation (even indirect) in a fraud-tainted chain, with actual or constructive knowledge, justifies denial of input tax. But the UAE goes further than the EU by:

1)enshrining this doctrine explicitly in the statute, not leaving it to case law; and
2) introducing a formal due-diligence presumption, linked to FTA-prescribed verification measures.

From a policy perspective, the UAE is clearly signaling that “I didn’t know” will no longer be an adequate defense if a business cannot demonstrate that it operated a robust, documented KYC/KYT-style process in line with FTA guidance.

GCC Perspective

14. While the UAE has now codified a knew / should have known test with a formal due-diligence presumption, no other GCC VAT jurisdiction presently embeds an equivalent mechanism in its primary VAT legislation, implementing regulations or publicly issued guidance.

14.1. Saudi Arabia (KSA)

Saudi VAT Law and its Implementing Regulations impose orthodox conditions for input tax deduction (valid tax invoice, taxable use, registration, non-blocked categories) and provide for penalties where VAT is under-declared or evaded. ZATCA may certainly deny input tax where those conditions are not met or where fraudulent invoices are used, but there is no statutory test that links denial of input VAT to participation in a “supply or chain of supplies related to tax evasion”, nor any express wording that the taxable person “should have been aware” of fraud elsewhere in the chain.

14.2. Bahrain

Bahrain’s VAT Law and NBR guidance follow a similar pattern. Input tax may be adjusted or denied if the basic statutory conditions are not satisfied, and tax evasion carries significant penalties. However, there is no Bahraini analogue of a Kittel-style input-VAT denial rule: no explicit notion of a chain of supplies related to tax evasion, and no due-diligence obligation framed as a legal precondition for deduction. Due diligence appears primarily as a general compliance expectation, not as a codified “knew / should have known” standard.

14.3. Oman

Oman’s VAT framework provides the usual rules on eligibility and restriction of input tax and contains a general anti-avoidance rule (GAAR). Administrative penalties apply for overstated input tax and for tax evasion. But Omani law does not currently set out a statutory knew / should have known test and does not treat failure to perform specified supplier checks as an automatic ground to deny input VAT by reference to fraud elsewhere in the chain.

14.4. GCC VAT Framework Agreement

Finally, the GCC Unified VAT Agreement itself sets only high-level conditions for input tax deduction and mechanisms for proportional deduction and adjustment. It does not introduce:

• a concept of a “supply or chain of supplies related to tax evasion”;
• a knew or should have known test; or
• a statutory due-diligence obligation as a precondition to input VAT recovery.

The UAE has therefore acted unilaterally in this respect and has gone beyond what is required by the Framework Agreement.

14.5. Implications of the UAE’s divergence from the GCC norm

This regional divergence has several practical consequences for taxpayers:

1) Multinational groups operating across the GCC will need UAE-specific VAT compliance and onboarding procedures. A “one-size-fits-all” policy built around KSA, Bahrain or Oman will not be sufficient.

2) Since Article 54(bis) directly targets the UAE taxable person’s right to credit, UAE buyers are likely to place much greater weight on a supplier’s VAT profile, internal controls and transparency when choosing between competing offers. That pressure will primarily affect UAE-registered suppliers, whose transactions sit inside the domestic VAT chain. Where the supply is from a non-resident and subject to the reverse-charge mechanism (RCM), the UAE buyer is treated, for VAT purposes, as supplying the goods or services to itself. In such cases, the “supplier” for UAE VAT purposes is effectively the buyer, not the foreign counterparty.

For imports of goods under the RCM, the import VAT is calculated at customs and then pre-populated in the VAT return of the registered importer. Payment at customs is deferred and the same amount is, typically, is recoverable as input VAT (where input VAT recovery is allowed). For cross-border services, the buyer simply includes the reverse-charge VAT as output and, subject to the normal rules, the same amount as input in its VAT return. Economically, these flows net to zero for a fully taxable business.

In this structure, there is no reliance on a third-party supplier to charge and remit UAE VAT, and the focus of risk shifts to the buyer’s own accounting rather than the supplier’s tax integrity. As a result, UAE buyers may perceive that less supplier-level due diligence is required where RCM applies, compared with domestic chains in which UAE-registered suppliers sit inside the Article 54(bis) denial risk. So, the UAE buyers are implicitly encouraged to source more from non-resident (including GCC) suppliers rather than from UAE-registered suppliers.

14.6. Potential signaling effect within the GCC.

As VAT systems in the GCC mature, other Member States may ultimately take inspiration from the UAE’s approach and introduce guidance or rules reflecting elements of the Kittel doctrine, at least in sectors historically associated with missing-trader fraud. For the time being, however, the UAE remains the only GCC jurisdiction that has explicitly legislated such a mechanism.

Key legal and practical questions under the UAE model

15. The statutory text raises several questions that will only be fully resolved once the FTA publishes its detailed procedures and the first disputes emerge. But some issues can already be identified.

15.1. Burden and standard of proof

Formally, the FTA must still “establish” that the supply formed part of a tax-evasion chain. That will require evidence of underlying evasion somewhere in the chain.

However, for the knowledge component:

• for actual knowledge, the FTA will need to show that the taxable person was aware of the fraudulent link;
• for constructive knowledge, the FTA must support an inference that the taxable person should have been aware based on the circumstances.

The due-diligence presumption significantly lowers the evidentiary barrier for the FTA: failure to follow the prescribed checks becomes, in itself, enough to treat the person as having been “required to be aware”.

The practical risk is that disputes migrate from “was there fraud?” to “did you tick all the boxes in the FTA’s verification procedures?”. This is exactly the kind of dynamic that the CJEU tried to limit.

15.2. Content and proportionality of FTA verification measures

Because the law defers the crucial notion of “validity and integrity checks” to FTA measures and conditions, the technical design of those measures becomes central.

These are some questions to watch:

1) Will the FTA adopt a risk-based model, focused on high-risk sectors and patterns, or a uniform checklist for all supplies?
2) Will it recognize that due diligence obligations must be proportionate to the size and resources of the business, and to the nature of the transaction?
3) Will it align verification measures with upcoming e-invoicing requirements and existing electronic systems, or impose separate paper-based routines?

If the measures are designed sensibly, the UAE regime will look very much like a codified, modernized version of the Kittel principle. If they are too rigid, there is a real risk that legitimate businesses are denied input tax for procedural slips without any realistic connection to fraud.

15.3. Temporal application and historic chains

The amendments are effective from 1 January 2026. This immediately raises the question: what, exactly, is “new” from that date: the standard itself, or only the FTA’s power to rely on it in respect of future periods?

There are at least three layers to this.

15.3.1. Periods fully after 1 January 2026

For tax periods that begin on or after 1 January 2026, the position is relatively straightforward:
1) input VAT attributable to those periods will clearly be subject to Article 54(bis) in its amended form;
2) returns filed for those periods should already reflect the new risk environment and the taxpayer’s compliance with FTA verification measures once they are published.

In practice, this is the “cleanest” zone: going forward, the FTA can deny input tax under the new clauses without any serious temporal argument.

15.3.2. Periods that straddle 1 January 2026

More delicate are tax periods that start before and end after 1 January 2026 (for example, a quarterly period running from 1 November 2025 to 31 January 2026).

One can reasonably expect the FTA to argue along the following lines:

1) the right to deduct is normally linked to the moment when the taxpayer both receives the supply and holds a valid tax invoice (and, in some cases, when the supply is used in making taxable supplies);
2) if those conditions crystallise after 1 January 2026, the deduction is being exercised under the new law, even if part of the underlying commercial activity or contract runs from 2025;
3) therefore, input tax first recognised / claimed in a return covering a period after 1 January 2026 is exposed to the new Article 54(bis) regime, regardless of when the contract was signed or the chain was set up.

From the taxpayer’s perspective, there is a good argument that the anti-abuse mechanism should not retroactively re-characterise input tax that was properly deductible under the law in force when the transaction took place. In a straddling period, however, this becomes a question of timing of the right to deduct, not merely of contract dates. If the right to credit arises or is actually exercised after 1 January 2026, it is difficult to insist that only the “old” rules apply.

At the same time, the text of the amendment does not expressly resolve how straddling periods should be treated. In the absence of specific transitional guidance, the analogy-of-law principle (analogia legis) can be invoked in favor of the taxpayer. In a number of other UAE tax amendments, the legislator has expressly provided that new rules apply only to “tax periods commencing on or after” a given effective date.

That drafting technique is more than stylistic: it is a clear indication of how the legislator deals with temporal application when it turns its mind to the issue. By extending that express approach by analogy to Article 54(bis), one can argue that the new mechanism is intended to govern only tax periods that start on or after 1 January 2026, and that periods which merely straddle that date should continue to be governed, in their entirety, by the pre-existing rules. On this view, the new Article 54(bis) regime cannot be relied upon to re-characterize deductions relating to straddling periods, because those periods do not fall within the category of “new” periods that the legislator has chosen to subject to the amended standard.

A further element is Clause 3, which defers the key notion of “verification of the validity and integrity” of supplies to measures, procedures and conditions still to be issued by the FTA. If those measures were applied in such a way that taxpayers are criticized, after 1 January 2026, for having failed to perform checks in 2025 that were not yet prescribed and could not have been known, that would amount to a clear retrospective application of the law, effectively turning past, law-compliant behavior into non-compliance by reference to ex post procedural standards. Such an outcome would sit uneasily with the express choice to introduce Article 54(bis) only from 1 January 2026.

Different interpretative approaches are therefore possible. Pending explicit clarification, a prudent, defensible position is:

1) to treat any input tax first claimed in a tax period that includes or follows 1 January 2026 as being potentially subject to the new denial mechanism in dealings with the FTA; but

2) to be prepared, where appropriate, to argue that deductions linked to supplies and factual circumstances predating 1 January 2026 should not be re-assessed by reference to verification procedures and standards that did not yet exist when the taxpayer acted.

15.3.3. Historic periods and ongoing audits (the “retroactivity” question)

The most contentious question is whether the FTA could seek to apply the new clauses to:

• input tax claimed in periods entirely before 1 January 2026,
• but which remain within the statute of limitation and are currently (in 2026 and afterwords) under audit or reassessment.

In formal legal terms, there is a strong doctrinal distinction between:

• substantive rules that define when a right exists (e.g. the conditions under which input VAT is recoverable), and
• procedural rules that govern how the FTA audits, collects and enforces tax (deadlines, documentation powers, etc.).

As a matter of general legal principle, new substantive conditions on rights (like input-VAT recovery) are not supposed to apply retroactively to periods closed under the old law, unless the legislature has clearly expressed such an intention. A taxpayer is entitled to structure its affairs based on the law as it stood at the time.

On that view, one can argue that:

• for tax periods entirely before 1 January 2026, input VAT already deducted should be judged against the old conditions;
• the FTA should not be able to deny a pre-2026 deduction solely on the basis that the taxpayer did not carry out checks that were not yet prescribed or required.

However, the reality in practice is more nuanced because:

1)the concepts of tax evasion and abuse of rights existed before the amendment;
2) even under the old law, the FTA could argue (by analogy with international practice) that a completely complicit taxpayer had no genuine “right” to recover;
3) the new wording might therefore be presented as an interpretative clarification or “codification” of what was already possible, rather than a completely new condition.

This is exactly where the risk lies. If the FTA frames Article 54(bis) as a clarification rather than a genuine tightening, it may be tempted to use the new language (especially the notion of “should have been aware”) in disputes over historic periods, even though the law was formally different at the time.

From a risk-management perspective, that means businesses should assume:

a) pre-2026 periods are not completely immune from Kittel-style arguments in audits, especially where the facts look egregious;

b) the FTA may still scrutinise historic chains and invoke ideas of constructive knowledge, even if the technical hook is framed under the old general anti-abuse and evasion provisions rather than the new Article 54(bis) wording.

15.3.4.Practical takeaway

Putting this together, a cautious but realistic reading is:

1) Tax audits initiated from 2026 onwards will almost certainly apply the new Article 54(bis) test to any input tax first claimed in periods ending on or after 1 January 2026.
2) For earlier periods, the FTA might not have a clean legal basis to deny input tax purely by invoking the new text, but it can still pursue aggressive positions using pre-existing anti-evasion concepts (especially if the taxpayer’s behavior looks reckless).

This is why 2025 should be treated as a genuine transitional year:

a) implementing due-diligence processes and documentation before 1 January 2026 gives the business a story that “we anticipated the new regime and already acted in its spirit”;

b) revisiting significant historic input-VAT positions (especially large refunds not yet closed or chains with obvious risk indicators) allows the taxpayer to clean up weak positions and, where necessary, adjust and disclose before the FTA does.

From a strict legal standpoint, the new Article 54 clauses apply prospectively from 1 January 2026. However, in practical audit situations, any input VAT first claimed in returns covering periods after that date can expect to be tested against the new standard. Historic chains with obvious risk features may still be challenged under then-existing anti-evasion rules. Treating 2025 as a process-re-design and risk-clean-up year is therefore the safest course.

Practical recommendations for UAE businesses

16. From a client-advisory perspective, the key message is straightforward: input VAT is now contingent not only on the nature of your purchases, but on the quality of your due diligence. Below are practical steps that can be recommended.

16.1. Adopt a formal “VAT fraud risk and due-diligence policy”.

Taxable persons should consider adopting a written policy that:

• recognizes that the business may lose input VAT if it participates in a fraud-tainted chain;
• sets out the business’s risk appetite and tolerance for counterparties and sectors; and
• designates responsible persons (e.g. tax manager / CFO) for approving high-risk counterparties and transactions.

This policy should be approved at board or senior-management level to evidence governance.

16.2. Implement risk-based supplier onboarding and monitoring

A risk-based onboarding framework should be established, with more intensive checks for high-risk sectors (e.g. trading chains, commodities, electronics, high-value, low-margin goods) and for unknown or offshore counterparties.

Typical checks may include:

a) verifying the supplier’s TRN and registration status and, where relevant, their license and permitted activities;
b) confirming physical presence (office, warehouse, staff) and not merely a mailbox;
c) checking publicly available information on the business and its owners;
d) understanding the commercial rationale of the transaction (why this supplier, why these margins, why this routing?).

The important point is not to adopt an abstract “checklist”, but to design controls that genuinely prevent participation in fraud and, as a consequence, produce a documented trail showing that the business took reasonable, risk-based steps. The mere ability to show evidence of due diligence can always be challenged with the argument that “you did not do enough”. In the best case, good documentation helps you win a dispute, but a mechanism that is built to prevent problems rather than simply to “look good on paper” is far more likely to keep the dispute from arising in the first place.

16.3. Build transaction-level “red-flag” review

Article 54(bis) targets supplies that are part of a chain connected with tax evasion. This is often visible in the pattern of transactions, not just in the attributes of the supplier.

Taxable persons should therefore embed simple red-flag criteria into their Accounts Payable and tax review processes, for example:

1) unusually high or low margins compared to market;
2) frequent changes of supplier in the same trading chain;
3) circular flows of goods (or services) with no obvious added value;
4) pressure to settle quickly or use unusual payment arrangements (third-party accounts, offshore banks) inconsistent with the commercial context.

When such flags appear, the policy should require escalation, additional questioning, and (if unresolved) refusal from purchase or from claiming input tax.

16.4. Align with forthcoming FTA procedures

Once the FTA issues the measures, procedures and conditions for verifying the “validity and integrity” of supplies, those will become the minimum legal standard. Thus, prudent taxpayers should:

1) implement those measures as a baseline,
2) map them into existing procurement, Accounts Payable and tax workflows (including e-invoicing as it comes on line), and
3) go slightly beyond where the business is exposed to heightened risk (e.g. sectors historically associated with missing-trader fraud in other jurisdictions).

The documentation should clearly show that the business has internalized the FTA’s procedures, not treated them as a box-ticking exercise.

16.5. Strengthen contractual protections

Contracts with suppliers should reflect the new environment, for example by:
1) including warranties that the supplier is properly registered, will comply with VAT obligations and is not involved in tax evasion;
2) providing information and audit rights (e.g. to request supporting documents, evidence of VAT returns, etc., where concerns arise);
3) incorporating clear termination and indemnity clauses triggered by evidence or strong suspicion of VAT fraud.

However, international practice (and now the UAE amendment itself) makes it clear that contractual clauses alone do not protect input VAT if the business is found to have “should have known” of fraud. Contracts are one strand of a broader defence, not a substitute for independent checks.

16.6. Build and maintain a “due-diligence file”

In disputes under the Kittel principle in Europe, one of the decisive factors is often whether the taxpayer can produce a contemporaneous due-diligence file: evidence of checks made, questions asked, decisions taken.

Taxpayers should maintain, preferably in electronic form, a file that records for each high-risk supplier or transaction:

1) KYC / onboarding documents and licenses;
2) screenshots of registry and FTA database checks;
3) internal emails / memos explaining why the supplier and transaction are considered acceptable;
4) records of red flags identified and how they were mitigated.

This file should be capable of being presented quickly in an FTA audit to show that the business did not intentionally or negligently close its eyes to obvious fraud.

16.7. Training and culture

Finally, the new rule should be embedded into training for finance, procurement and sales staff:

1) explaining in plain language that careless selection of counterparties can lead to loss of input VAT and penalties;
2) giving concrete examples of red flags, drawn from international practice with Kittel-style fraud;
3) encouraging staff to escalate concerns without fear of being seen as “blocking business”.

A culture in which “everyone knows we must check” is itself one of the best protections.

Procurement reality: why “document collections” and indemnity clauses are not enough

17. Experience from other jurisdictions shows that the instinctive reaction to VAT-fraud risk is often to ask procurement to “collect more documents” from suppliers: extra certificates, letters of good standing, comfort letters, additional questionnaires and so on.

18. This approach has a clear advantage: sometimes, it does work as a defensive argument. But in many real disputes it fails, because it does not address the underlying economics of the transaction. When there is a shortfall in the budget and the tax authority needs to recover lost revenue, it is often easier to deny input VAT from the buyer for “lack of prudence” than to pursue the ultimate beneficiaries behind non-compliant counterparties.

19. Ever-expanding document packs can become an administrative burden without changing the basic fact that the authority is looking for someone in the chain who can realistically pay. The deeper problem frequently lies in the procurement formula itself. If the business implicitly evaluates suppliers on the basis that:

Supplier price = competitor’s price – internal cost – “tax if paid honestly”

then, in many heavily competed markets, this algebra ensue the following outcomes:

1) With that pricing pressure, paying all taxes honestly may be commercially impossible.
2) The cheaper the price selected by procurement, the higher the probability that someone in the chain is cutting corners on tax.

20. In such a setting, it is illusory to believe that VAT risk can be eliminated solely by collecting more paperwork. At some point, either:

• the supplier exits the market, or
• the supplier (or someone behind them) breaks the law, because otherwise they cannot meet the price that procurement is optimised to select.

The rational response, in light of Article 54(bis) and its due-diligence presumption, is to change the formula, not just the documentation:

• allow for a “tax-compliance premium” in price evaluations – recognizing that a supplier who can show sustainable margins and full tax compliance will not be the cheapest on paper;
• build explicit VAT-risk criteria into RFPs and scoring models (e.g. quality and transparency of tax profile, history of disputes, robustness of internal controls);
• treat unusually low prices in high-risk sectors as an inherent red flag, triggering escalation rather than being celebrated as a win.
  
  
  21. Many groups also experiment with compensatory clauses – contractual provisions stating that if the tax authority assesses additional VAT because of supplier non-compliance, the buyer can recover that amount from the supplier. These clauses are useful and should be part of the toolkit, but they are not a complete solution:
  
  1)their legal enforceability can be contested;
  2) practical recovery is uncertain if the supplier becomes insolvent or simply disappears;
  3) suppliers and internal management often resist strong indemnities as “too aggressive”;
  4) in practice, they may function more as a psychological comfort than as a realistic way to recover assessed VAT.
  
  22. Under the new UAE regime, this means that document packs and indemnities are helpful but not sufficient. The taxpayers need to re-design procurement so that “cheapest at any cost” is no longer an acceptable KPI.

Only when pricing, supplier selection and contract design all reflect the reality of VAT-fraud risk will the business be able to demonstrate, credibly, that it did what Article 54(bis) implicitly demands: to avoid being the party that “should have known” it was participating in a tax-evasion chain.

Conclusion: codified Kittel with a UAE twist

23. Viewed globally, the UAE’s new Article 54(bis) regime is broadly aligned with the EU’s Kittel principle and with UK practice under HMRC’s anti-fraud interventions: the underlying policy is to deny input VAT where a trader, knowingly or with wilful blindness, participates in a fraud-tainted chain. Regionally, however, the UAE now stands apart from its GCC peers as the only jurisdiction that has chosen to legislate this doctrine explicitly and to link it to a formal due-diligence presumption. What is distinctive about the UAE is the choice to:

1) write this doctrine directly into the VAT Law, rather than leaving it to case law and soft guidance; and
2) link it to a formal, prescriptive due-diligence regime, where failure to follow FTA-mandated checks can itself be enough to treat the business as having been “required to be aware” of tax evasion.

The practical message for Taxable Persons is clear and immediate: From 1 January 2026, input VAT is no longer a purely mechanical entitlement. It is conditional on demonstrable, risk-based due diligence over your supply chains.

Businesses that invest now in robust, documented suppliers and transaction-level controls will not only be better protected from fraud, but will be in a much stronger position to defend their input-tax claims when the first FTA audits under the new regime begin.

Disclaimer

Pursuant to the MoF’s press-release issued on 19 May 2023 “a number of posts circulating on social media and other platforms that are issued by private parties, contain inaccurate and unreliable interpretations and analyses of Corporate Tax”.

The Ministry issued a reminder that official sources of information on Federal Taxes in the UAE are the MoF and FTA only. Therefore, analyses that are not based on official publications by the MoF and FTA, or have not been commissioned by them, are unreliable and may contain misleading interpretations of the law. See the full press release here.

You should factor this in when dealing with this article as well. It is not commissioned by the MoF or FTA. The interpretation, conclusions, proposals, surmises, guesswork, etc., it comprises have the status of the author’s opinion only. Furthermore, it is not legal or tax advice. Like any human job, it may contain inaccuracies and mistakes that I have tried my best to avoid. If you find any inaccuracies or errors, please let me know so that I can make corrections.