Of the reporting regimes introduced over the past decade, DAC6 is the one that most reliably surprises groups that are not doing anything aggressive. It does not ask whether an arrangement is abusive. It asks whether the arrangement has certain features. A cross-border transaction becomes reportable because it displays a listed characteristic, not because an inspector has formed a view about it. That distinction is the whole of the regime, and failing to internalise it is why well-run groups discover, usually late in a transaction, that an unremarkable reorganisation carried a filing obligation nobody had allocated.
What the regime actually is
DAC6 is the European directive on mandatory disclosure of cross-border arrangements, implemented in the Netherlands within the domestic legislation on international assistance in tax matters. It applies to arrangements involving either more than one member state, or a member state and a third country, where at least one of a defined list of hallmarks is present. The report goes to a single tax authority and is then exchanged automatically with the others.
Two practical consequences follow from that architecture. The first is that a group files once but the information travels everywhere; a filing made in the Netherlands is read in the other jurisdictions touched by the same arrangement. The second is that inconsistency is visible. Where two intermediaries in two member states describe the same transaction differently, or where one files and another concludes there is nothing to file, the divergence sits in the exchanged data. Both points argue for the same discipline: settle the description of a transaction once, centrally.
Hallmarks, and the two tiers within them
The hallmarks are grouped into five categories, conventionally lettered A to E. Category A covers generic features associated with marketed arrangements: confidentiality conditions imposed on the client, fees linked to the tax result, substantially standardised documentation requiring little adaptation. Category B covers specific structural features such as the acquisition of a loss-making company to use its losses, the conversion of income into categories taxed more lightly, and circular transactions returning funds through interposed entities. Category C addresses cross-border payments between associated enterprises. Category D covers arrangements that undermine automatic exchange of financial account information or obscure beneficial ownership through opaque chains. Category E is the transfer pricing category: unilateral safe harbours, transfers of hard-to-value intangibles, and intra-group transfers of functions, risks or assets where the transferor’s projected earnings fall materially as a result.
The critical structural point is that these hallmarks operate on two tiers. Categories A and B, and part of category C, are only capable of triggering a report if the main benefit test is also satisfied, meaning that obtaining a tax advantage is one of the principal benefits a person may reasonably expect from the arrangement. Category D, category E, and the remainder of category C apply on their own terms, with no motive filter at all. A deductible cross-border payment to an associated enterprise resident in a jurisdiction that imposes no corporate tax, or that appears on the relevant list, can be reportable regardless of why the payment structure exists.
This is precisely why ordinary groups are caught. The hallmarks that require no motive test are the ones that describe routine corporate life: the migration of a manufacturing or distribution function, the licensing of intangibles whose value is genuinely uncertain at the time of transfer, an intercompany financing flow into a jurisdiction that happens to be listed. Groups that have already worked through their withholding tax position on dividends, interest and royalties tend to spot the payment hallmarks early, because the conditional withholding regime has already forced them to map where their deductible outbound flows land.
Who has to report, and what privilege changes
The primary obligation falls on the intermediary. The directive defines that term through two limbs. The first captures the person who designs, markets, organises or makes available for implementation a reportable arrangement, or who manages its implementation. The second captures anyone who provides aid, assistance or advice in relation to it, measured against what that person knew or could reasonably be expected to know. The second limb is the wider of the two, and it is the one that draws in advisers who did not conceive the transaction and would not describe themselves as its architects.
Where several intermediaries are involved, each is in principle obliged to file, and each is relieved only to the extent it holds proof that the same information has already been reported by another. That proof is a document, not an assumption. The common failure is a chain of parties each of whom believed a different party was filing.
Legal professional privilege does not extinguish the obligation; it relocates it. Where an intermediary is bound by professional secrecy under national law, the duty shifts to the other intermediaries and, failing them, to the relevant taxpayer, who must then report on its own account. The practical effect is that privilege makes it more likely, not less, that the obligation lands on the group itself. The same result follows where the only intermediary sits outside the European Union, or where an arrangement was developed in-house and there is no intermediary at all. In each of those cases the taxpayer reports.
Reporting is information, not confession
The most useful thing to say to a board is the simplest. A DAC6 report is a disclosure of facts within a defined window, running from the earliest of the arrangement being made available for implementation, being ready for implementation, or the first step in its implementation being taken. It is not an admission that anything is wrong, it does not create a presumption of abuse, and it does not concede any position on the merits. Nothing in the regime provides that a reported arrangement is thereby treated as ineffective.
A filing records what was done. It does not concede that it should not have been done.
The asymmetry is worth weighing. A report that turns out to have been unnecessary costs the effort of preparing it. A failure to report where an obligation existed exposes the group to penalties, and, more corrosively, invites an inspector to conclude that the group’s grasp of its own structure is weaker than its filings suggest. Where a hallmark is genuinely arguable, the disciplined answer is usually to report and to keep the reasoning that explains why the arrangement is nonetheless ordinary.
What an orderly group does about it
Very little of the work is technical. Most of it is allocation and record keeping.
- Map the intermediaries transaction by transaction. For each cross-border project, identify every person who could meet either limb of the definition, including advisers engaged by counterparties, and record who is expected to file. Do this at the outset, not at completion.
- Take the proof. Where reliance is placed on another party’s filing, obtain the reference number and hold it. Reliance without evidence is not an exemption.
- Keep a single register. One record covering arrangements reviewed, hallmarks considered, the conclusion reached and who reached it. The entries showing why something was not reportable matter as much as the ones showing that it was.
- Fix the trigger dates. The window runs from events that occur inside the deal timetable, not from signing or from closing, and it can open before the group considers the transaction settled.
- Put the obligation in the engagement terms. Advisory contracts should state expressly who files, who notifies whom, and what happens where privilege applies.
- Give it an owner. One named person in the group, usually the head of tax, with visibility over transactions initiated by operating subsidiaries rather than by the centre.
A reorganisation that turns out to be reportable
Consider a mid-sized industrial group consolidating its European distribution. Four national sales companies, each historically operating as a full-risk distributor, are converted into limited-risk distributors. Inventory ownership, credit risk and the customer contracts move to a Dutch BV, which becomes the European principal and holds the trademarks used in the region. The commercial rationale is real: one order book, one working capital pool, one set of terms. The group’s participations continue to fall within the participation exemption, transfer pricing documentation is prepared under article 8b, and remuneration for the converted entities is set by benchmarked routine returns. No tax advantage is being sought beyond the ordinary consequences of where functions now sit.
This is reportable, and on more than one basis. The conversion involves an intra-group cross-border transfer of functions and risks under which the transferors’ expected earnings fall materially, which is a category E hallmark carrying no main benefit test. The transfer of the regional trademarks may separately engage the hard-to-value intangibles hallmark, since no reliable comparable exists and the projections underpinning the valuation are uncertain. If any converted entity is remunerated by reference to a unilateral safe harbour in its own jurisdiction, that is a further category E hallmark. None of these depends on motive, so the group’s commercial reasons are, for reporting purposes, beside the point. They remain entirely relevant to whether the pricing survives audit, which is a separate question and the one that actually matters.
The practical failure in transactions of this shape is rarely analytical. It is that the tax function reviews the arrangement for substance, pricing and operating model, satisfies itself on all three, and never asks the separate question of whether anyone has filed.
The proportionate view
DAC6 is best treated as an administrative discipline rather than a risk. It rewards groups that know the shape of their own structure and can describe it consistently; it punishes groups whose cross-border projects are assembled by different advisers who each assume someone else is holding the file. Keep a register, allocate the obligation in writing, report where the answer is uncertain, and the regime becomes unremarkable.
Montclare structures and operates Dutch and cross-border holding platforms for international groups. Our services are set out on our services page.
This article is informational and does not constitute tax advice. Each engagement is subject to scope and applicable regulation.