When a multinational group carries out transactions between its own subsidiaries, it must set a price for these internal transactions: this is known as the transfer price. Tax authorities require that this price correspond to the price that two independent companies would have negotiated under comparable conditions. However, this is a complex and costly exercise, and is often a source of disputes, as two countries may have different views on what constitutes a ‘fair’ price, with the risk of double taxation as a result.
To address this challenge, the OECD introduced “Amount B” as part of its international tax reform. Instead of having to carry out complicated calculations every year, a matrix is applied – a table which, depending on the sector of activity, assets and expenses, directly determines the remuneration margin. The result: fewer disputes over how much the distributor should earn.
Amount B is optional for Member States, which may either require the companies concerned to include it, or undertake to recognise it when it is correctly applied, without calling it into question. Luxembourg has opted for recognition. This means that the Direct Taxation Administration (ACD) will not reclassify the price if the group has applied the matrix correctly. And if double taxation does occur, Luxembourg undertakes to eliminate it.
This recognition is, however, subject to three conditions set out in point 8 of the circular. Firstly, the other jurisdiction must be a covered jurisdiction appearing on the list drawn up as part of the political commitment under the OECD’s Inclusive Framework. Secondly, a double taxation agreement must be in force between that jurisdiction and Luxembourg. Finally, the covered jurisdiction must have adopted and effectively applied Amount B to distributors operating in its market.
An approach that prioritises legal certainty, but which also raises practical issues for multinational groups. Interview with Vanessa Ramos Ferrín, managing partner at Transfair Pricing Solutions.
Luxembourg has opted for the “recognition” approach rather than the “obligation” approach. What does this actually mean for your clients?
Vanessa Ramos Ferrín. – “Asymmetry is inherent in the very structure of Pillar B, which determines the remuneration of the local distributor – in other words, the entity being assessed – rather than that of the principal. The principal is the group entity that holds the goods, sets prices and bears the main commercial risks. It is precisely because it performs these functions and bears these risks that it is entitled to the residual profit, and not because it manufactures the products. If a subsidiary established in a covered jurisdiction – Morocco, for example – were to apply Amount B because its jurisdiction chose to apply it to distributors operating in its market, the distributor’s margin would be determined in Morocco, and the residual profit would accrue to the Luxembourg parent company, that is to say, the group entity that holds the goods, sets the prices and bears the main commercial risks.
The circular of 13 April 2026 specifically precludes ‘windfall gains’ in this scenario: where the three cumulative conditions for recognition are met, the ACD undertakes to respect the outcome of the correctly applied simplified framework; in other words, not to make any adjustment incompatible with that outcome and to take all reasonable measures to eliminate any double taxation through a corresponding adjustment or through the mutual agreement procedure. There is only one price, recognised by both parties provided the framework is correctly applied.
It has become a fully-fledged tool for tax governance.
Luxembourg’s decision is in line with those of the Netherlands, Germany and, more recently, Belgium, whereas the United States and, since 1 January 2026, Singapore [as a pilot scheme until 2028] opted instead to introduce an elective ‘safe harbour’ under domestic law. Japan, for its part, has explicitly rejected the simplified approach. This patchwork alone illustrates the asymmetries we are referring to. The real friction arises elsewhere: when the counterparty’s jurisdiction applies Amount B without meeting the conditions for recognition, for example in the absence of a tax treaty in force with Luxembourg, or when adoption is delayed.
In this case, the risk of double taxation becomes a real concern once again. This has become an explicit request: several groups have commissioned us to carry out mapping exercises that cross-reference their distribution footprint, the list of jurisdictions covered, Luxembourg’s network of double taxation agreements and the status of adoption on a country-by-country basis. It has become a tax governance tool in its own right.
In practice, how many situations actually meet all three conditions? Does this apply to many groups in Luxembourg today?
“Let’s be honest: at present, the population affected is still limited, but it is set to grow. The list of jurisdictions covered, drawn up by the OECD in June 2024 and annexed to the circular, comprises 66 countries, mainly low- and middle-income countries. It also includes major emerging economies, such as Brazil, Mexico, Argentina and South Africa, on the understanding that inclusion on this list does not oblige these jurisdictions to adopt the simplified approach. Five of these – Argentina, Brazil, Costa Rica, Mexico and South Africa – have expressed their intention to apply the approach to the Inclusive Framework, which is precisely why they appear on the list. The first condition is therefore broad in theory.
Secondly, the existence of a tax treaty in force with Luxembourg already significantly narrows the scope, as there is only a partial overlap between Luxembourg’s network of tax treaties and the list of jurisdictions covered.
The complete triptych remains rare at this stage, but now is precisely the time to prepare for it.
The third factor – this court’s actual decision to apply Amount B to distributors operating in its market – is now the real bottleneck: for financial years commencing on or after 1 January 2025, the number of jurisdictions covered that have formally implemented the simplified framework remains modest. In practice, we are already assisting with cases where the first two conditions have been met and where the third is in the process of being fulfilled, as the relevant legislation or local administrative guidelines have been announced or are currently under consultation.
The complete set of three documents is still rare at this stage, but now is precisely the time to prepare for it: the circular applies to financial years commencing on or after 1 January 2025, and those groups that plan ahead will avoid being subject to the measures as a result.
Have you ever agreed with a client that the general transfer pricing regime was more favourable than Amount B?
“Yes, and it’s actually a process we carry out systematically: comparing the results of the matrix with those of a conventional comparability analysis. Three scenarios typically arise. The first concerns sectors with structurally low margins, where a search for regional comparables may support a return on sales below the matrix’s lower limit; the simplified framework then assigns the distributor a higher margin, thereby reducing the residual amount payable to the principal by the same amount.
The second concerns distributors in the market penetration phase or experiencing justified losses: the matrix does not recognise losses, whereas the general regime allows for the reporting of temporary negative results where the circumstances justify it.
The third concerns entities situated close to the thresholds of the operating cost corridor, where mere cyclical fluctuations – despite the smoothing effect provided by the three-year weighted average stipulated by the scheme – cause the entity to move in and out of the scheme; these are what we call ‘false negatives’ – essentially routine distributors – excluded for purely arithmetical reasons. That said, it is important to bear in mind a reality of the Luxembourg model: in the ‘recognition’ variant, the choice often does not lie with the group as seen from Luxembourg. If the jurisdiction in question imposes Amount B on its distributor, the comparison becomes purely academic; it then serves to measure the impact, not to make a decision.
Many organisations in Luxembourg combine distribution with other functions. Where does the line lie in practice?
“The line is drawn in two stages. Firstly, through explicit exclusions: the distribution of intangible goods and services, as well as the marketing, trading or distribution of commodities, and any transaction that cannot be precisely defined due to additional activities going beyond the standard distribution.”
Secondly – and this is the crux of the matter for Luxembourg – through segmentation. The most common implementation challenge concerns entities with a dual role, which combine a principal function with residual sales or local support activities. Eligibility depends on the substance of the activities carried out, not on the legal classification of the entity; it is therefore necessary to demonstrate a reliable functional and financial separation, which often requires adapting internal systems, as ERP structures and cost centres are not always aligned with functional roles. The same challenge arises for ancillary activities integrated with the sale of goods, such as logistics coordination, customer onboarding or support services, which require careful demarcation.
Artificial restructuring aimed at taking advantage of the scheme will be subject to increased scrutiny.
As for reclassifications, the scheme is too new to allow for any observed cases. The risk is nevertheless evident in both directions: taxpayers might seek to bring entities within the scope of the scheme that actually assume economically significant risks and, conversely, tax authorities might be tempted to include entities that exceed these thresholds. The circular and the OECD report also emphasise this: artificial reorganisations designed to take advantage of the scheme will be subject to increased scrutiny. This, in my view, is where the litigation of the future will play out.
We’re simplifying the pricing, but the risk of double taxation remains. Isn’t that a contradiction?
“The paradox is real, but it must be properly contextualised. Within the scope of the commitment – namely, a covered and listed jurisdiction, a tax treaty in force and Amount B correctly applied – the risk is largely neutralised: the circular confirms that Luxembourg will eliminate double taxation through a corresponding adjustment or as part of a mutual agreement procedure.
The problem lies outside this scope, and it is twofold. On the one hand, certain covered jurisdictions – specifically those targeted by the mechanism – have a limited network of treaties and restricted administrative capacity in relation to amicable proceedings; without a treaty with Luxembourg, the commitment to recognition simply does not apply. On the other hand, the asymmetrical timing of adoption creates points of friction: inconsistent supervisory approaches between administrations and uncertainty in the design of group policies.
Conventional dispute resolution mechanisms – such as amicable settlement and arbitration where provided for in the agreement – offer useful, albeit imperfect, safeguards: the timeframes remain lengthy, and compulsory arbitration is rarely accepted by courts in developing countries. Two complementary instruments are worth mentioning. In the EU, the Directive on the settlement of tax disputes offers a binding procedure with a fixed timeframe; however, by its very nature, it does not apply to developing jurisdictions, which are precisely the target of Amount B. Furthermore, the OECD has published a model agreement between competent authorities designed to organise, on a bilateral basis, the recognition of the results of the ‘B’ amount and the elimination of double taxation – a tool that is particularly relevant where the network of tax treaties is incomplete.
Amount B significantly reduces the scope for dispute regarding the margin level, but it does not eliminate the risk of double taxation.
These concerns are not merely theoretical: the survey we conducted in 2025 within the Luxembourg Transfer Pricing Association [LTPA] working group amongst corporate tax specialists already identified, as their primary concern, the risk of double taxation arising from asymmetric adoption, ahead of inconsistencies in audits in key foreign markets and uncertainty regarding the design of the group’s pricing policies. My view is therefore nuanced: Amount B considerably reduces the scope for dispute regarding the level of the margin, but it does not eliminate the risk of double taxation; rather, it shifts it to cases outside the scope of the rules, which require proactive rather than reactive management.
In your view, is this a genuine step forward, or is it merely a passing fad driven more by rhetoric than by tangible results?
“Amount B represents a genuine, albeit targeted, step forward: for distributors whose business is genuinely routine and who are located in jurisdictions covered by agreements, it removes the most common source of transfer pricing disputes, namely the debate over the level of the margin. This is not just a passing fad: the mechanism is now incorporated into the OECD Principles and implemented in binding national legislation, including the Luxembourg circular of 13 April 2026. However, its success will depend on the extent to which it is adopted by the covered jurisdictions, which remains limited to date. The progress is therefore real within its scope, but that scope is still narrow.”
