A group with a Dutch principal and a Spanish operating subsidiary does not have two transfer pricing positions. It has one economic arrangement described twice, in two languages, to two administrations that increasingly read each other’s descriptions. Both the Netherlands and Spain apply the OECD Transfer Pricing Guidelines, so the substantive standard is common ground. What differs is the packaging: what must be documented, by when, in what form, and how the inspector on the other side of the table is disposed to read it. The practical exposure for a mid-sized group sits almost entirely in that gap.
One standard, two administrations
The Dutch obligation is set out in article 8b of the Wet op de vennootschapsbelasting 1969. It contains two distinct duties. The first is the arm’s length principle itself. The second, and the one groups underestimate, is the duty to keep in the administration the information showing how the prices were arrived at. That second duty has no turnover threshold. A Dutch BV with a single intra-group loan and EUR 4 million of revenue owes it in the same terms as a group with EUR 400 million. We have set out the mechanics of that obligation in more detail in our article on the article 8b duty.
Spain builds on the same OECD foundation but codifies its documentation requirements in a tiered structure, with the content of the local file scaling according to the size of the group and with a separate informative return sitting alongside the corporate income tax filing. The principle is identical. The administrative choreography is not.
Where the formalities diverge
Four differences matter operationally.
- Thresholds. Master File and Local File become mandatory in the Netherlands from EUR 50 million of consolidated group turnover, with Country-by-Country Reporting from EUR 750 million. Spain tiers its documentation by reference to net group turnover at a lower level, above which the simplified local file content falls away and the full content applies. A group can therefore sit in the full Spanish tier while remaining below the Dutch threshold, which is a common and awkward position.
- Form. The Dutch file is kept in the administration and produced on request. Spain requires an affirmative annual filing, Form 232, reporting related-party transactions above transaction-level thresholds that bite well below any group turnover test. Filing it is a positive act with a deadline; forgetting it is visible in a way that a thin Dutch file is not.
- Timing. Dutch documentation is expected to be in place when the transaction is entered into and priced, rather than reconstructed at audit. Spanish documentation must be complete by the corporate income tax filing deadline. In practice the Spanish deadline drives the calendar and the Dutch requirement drives the drafting discipline.
- Temperament. This is impression rather than law, but in our experience Dutch inspectors tend to open with the logic of the analysis, testing whether the functional characterisation supports the return being reported, while Spanish inspections more often open with the comparables set and the statistical range, and are more willing to challenge the search strategy itself. The same file can survive one line of attack and fail the other.
The exposure is the second narrative
Groups typically discover the problem in the following way. The Dutch file, drafted by the group’s Amsterdam adviser, describes the BV as the entrepreneur bearing market and inventory risk. The Spanish file, drafted locally two months later, describes the Spanish company as a distributor with meaningful local commercial autonomy, because that framing sat more comfortably with the Spanish team and with an old commercial agreement nobody re-read. Neither document is untrue on its face. Together they are indefensible, because between them they characterise the same risks twice and pay for them twice, or once and nowhere.
The exposure is rarely the price itself. It is the second description of the same facts, written eighteen months later, by someone who never read the first.
Under the exchange of information mechanisms in force across the European Union, and through CbC reporting where the group is in scope, both administrations have structural visibility into what the other has been told. An inconsistency that would once have remained buried in two filing cabinets is now a question in a first audit letter.
Building a single functional analysis
The remedy is architectural rather than clever. One functional analysis is prepared at group level, tested against both regimes, and then rendered into two local files that differ in presentation but not in substance.
That analysis has to do the work properly. It must identify who performs the significant people functions, who has the financial capacity to bear each risk, and who exercises control over it. Where intangibles are involved, it must allocate returns by reference to development, enhancement, maintenance, protection and exploitation, and it must do so on the evidence of who actually decides, not on the evidence of who holds the registration. It must then select a method, CUP, TNMM, resale price, cost plus or profit split, and explain why the alternatives were rejected. That rejection paragraph is the part most often omitted and the part an inspector in either country reaches for first.
Where the group also has Luxembourg or other European entities in the chain, the same discipline extends across the whole structure; we have discussed the interaction of the Dutch, Luxembourg and Spanish positions in a separate article on cross-border holding structures.
A worked example: a Spanish adjustment and its Dutch consequence
Take a group with a Dutch principal and a Spanish limited-risk distributor. The Spanish company records EUR 40 million of third-party turnover and is remunerated on a TNMM basis at a target operating margin of 1.8 per cent, giving EBIT of EUR 720,000.
The Spanish administration reviews the comparables set, rejects three companies as insufficiently comparable, and rebuilds the interquartile range. On the revised set the median sits at 3.2 per cent. The inspector adjusts to the median rather than the lower quartile, on the basis that the taxpayer’s own search was unreliable.
- Revised Spanish EBIT: EUR 1,280,000
- Upward adjustment: EUR 560,000
- Spanish corporate income tax at the general 25 per cent rate: EUR 140,000, before late payment interest and any documentation penalty
The group’s instinct is to book a corresponding downward adjustment of EUR 560,000 in the Netherlands, relieving tax at 25.8 per cent, or EUR 144,480, and to treat the matter as economically closed.
That instinct runs into two constraints. The first is procedural: a corresponding adjustment following a foreign assessment is claimed from the Dutch authorities on evidence, not booked unilaterally because a Spanish notice has been issued. The second is substantive. The Dutch rules against transfer pricing mismatches, in force since 1 January 2022, restrict a downward adjustment in the Netherlands where no corresponding amount is actually taken into account in the counterparty’s taxable base. Those rules brought informal capital planning to an end, and the same logic governs here: the group must be able to evidence the Spanish assessment, the amount actually brought into charge and the tax actually paid.
Two failure modes follow. If the Spanish assessment is later settled at a reduced figure without the Dutch position being revisited, the Dutch downward adjustment is overstated and unsupported. If the Dutch year in question is already closed, relief may be unavailable through the domestic route altogether, and the group carries EUR 140,000 of double taxation, plus interest on that amount, on profit it earned once and was taxed on twice.
When the mutual agreement procedure earns its cost
Where relief cannot be obtained cleanly through domestic correction, the mutual agreement procedure exists precisely for this position. Within the European Union, Council Directive (EU) 2017/1852 gives taxpayers a structured route: a complaint must be lodged within three years of the first notification of the action giving rise to the double taxation, the competent authorities have two years to reach agreement, extendable by one, and failing that an advisory commission is convened.
MAP repays the effort where the amounts justify a multi-year process and where the group’s two files tell a consistent story. It deserves more caution where they do not, because the procedure requires both administrations to look at the same documentation side by side. A group that enters MAP with contradictory local files is inviting a second assessment rather than relieving the first. That, more than anything, is the argument for consistency at the drafting stage.
What holds up under both readings
A file that survives in both jurisdictions shares certain features. The functional analysis is one document, not two. The intercompany agreements match the conduct described in it. The comparables search is documented step by step, including the rejections, so that a Spanish challenge to the search strategy meets a written answer rather than a spreadsheet. Adjustments made in one country are reflected in the other in the same financial year, with the evidence of the counterparty’s treatment retained. And the whole set is reviewed when the business changes, not when the audit letter arrives. Groups building or reorganising their European structure will find the same considerations addressed in our guide to Dutch holding structures.
Montclare has published a short self-assessment, the Transfer Pricing Readiness Check, which sets out ten questions that identify where a group’s exposure sits. It can be downloaded from our transfer pricing page.
This article is informational and does not constitute tax advice. Each engagement is subject to scope and applicable regulation.