The fifty million euro threshold is the most misread figure in Dutch transfer pricing. It is read as the point at which documentation begins, when in fact it is the point at which documentation acquires a prescribed form. Below it, a Dutch taxpayer still has to be able to show how its intercompany prices were set; above it, that showing has to be organized into two specific files, with specific contents, available at a specific moment. Groups that treat the threshold as an on switch tend to arrive at the crossing year with nothing to convert, and groups that treat it as a formatting exercise tend to arrive with a Local File that describes a company other than the one the inspector is looking at.
Where the threshold sits, and where it does not
Two obligations run in parallel and are frequently collapsed into one. The first is the general obligation under article 8b of the Wet op de vennootschapsbelasting 1969: where bodies participate directly or indirectly in each other’s management, supervision or capital and agree conditions between themselves that differ from what independent parties would have agreed, the profit is determined as if the arm’s length conditions had applied, and the third paragraph requires those bodies to hold in their administration the data showing how the prices came about and whether they meet that standard. That obligation carries no turnover threshold whatsoever. It applies to a BV with a single intercompany loan as much as to a Dutch entity in a group turning over billions. We set out its scope in the foundational article on the article 8b obligation, and everything below assumes it as the floor.
The second obligation is the standardized documentation package introduced with the Dutch implementation of BEPS Action 13, which sits in articles 29b to 29h of the same act. A group entity that is liable to tax in the Netherlands must take a Master File and a Local File into its administration, the two documents the statute calls the groepsdossier and the lokale dossier, and article 29g attaches that duty to group entities of a multinational group that achieved at least EUR 50,000,000 of consolidated group revenue in the reporting year immediately preceding the year to which the return relates.
Country-by-Country Reporting sits far higher and is a different instrument. Article 29c disapplies the country report for group entities of a multinational group that earned less than EUR 750,000,000 of consolidated group revenue in the reporting year immediately preceding the year the report would cover, and where the report is due it goes to the inspector within twelve months of the last day of the reporting year the report covers. The notification duty runs separately and falls earlier. Under article 29d, a group entity resident in the Netherlands tells the inspector, no later than the last day of the reporting year itself, whether it is the ultimate parent entity, the surrogate parent entity or the designated group entity, and if it is none of them, the identity and tax residence of the entity that will report. Article 29h attaches an administrative penalty to intentional or grossly negligent failures under those two articles; it does not name article 29g. None of this has any bearing on the fifty million question.
That the test looks to the preceding reporting year matters more than it appears to, and it is the source of the timing confusion addressed in the worked example below.
What the Master File is actually for
The Master File is a group document. It gives the Dutch inspector the context in which the Dutch entity’s results are supposed to make sense. Its contents are not left to judgment: article 29g instructs that the form and content be fixed by ministerial regulation, and the model in Annex D to the Regeling aanvullende documentatieverplichtingen verrekenprijzen follows the OECD template. It calls for a chart of the group’s legal and ownership structure and the geographical location of operating entities; the important drivers of business profit; the supply chain for the group’s five largest products or service offerings by turnover, plus any others amounting to more than five per cent of group turnover; a list and brief description of important service arrangements between group members other than research and development services; the main geographic markets for those products and services; a brief functional analysis describing the principal contributions to value creation by individual entities; and a description of important business restructurings, acquisitions and divestitures occurring during the year.
Beyond that, it covers the group’s intangibles: the overall strategy for development, ownership and exploitation, the list of intangibles important for transfer pricing purposes and which entities legally own them, the important intercompany agreements relating to them including cost contribution arrangements and licence agreements, the group’s transfer pricing policies on research and development and on intangibles, and a general description of any important transfers of interests in intangibles among associated enterprises during the year, including the entities, countries and compensation involved. It covers intercompany financial activity, including how the group is financed, the identification of any members performing a central financing function together with the country under whose laws each is organised and its place of effective management, and the general policy on financing arrangements between associated enterprises. Finally it covers the group’s financial and tax position: the consolidated annual financial statement for the year where one is otherwise prepared, and a list and brief description of existing unilateral advance pricing agreements and other tax rulings relating to the allocation of income among countries.
The Master File explains the system. It does not, and is not meant to, prove that the Dutch entity’s margin is defensible.
What the Local File has to do
The Local File is the document that carries the Dutch tax burden, and its model sits in Annex F to the same regulation. It describes the Dutch entity: its management structure, its local organization chart, the individuals to whom local management reports and the countries in which those individuals maintain their principal offices, a detailed description of the business and business strategy including an indication of whether the entity was involved in or affected by business restructurings or intangible transfers in the present or immediately past year, and the key competitors.
Everything after that is organized per material category of controlled transaction, and within each category the file moves from what happened to why the price stands. It begins with a description of the transactions and the context in which they take place, and the amount of intragroup payments and receipts for each category broken down by tax jurisdiction of the foreign payer or recipient, so that the flow being priced is measured before it is defended. It then identifies the associated enterprises involved and the relationship among them, and carries copies of all material intercompany agreements concluded by the local entity, which is where a file first fails in practice, since an agreement that was never signed cannot be annexed. On that footing comes a detailed comparability and functional analysis of the taxpayer and the relevant associated enterprises, including any changes compared to prior years, the change requirement being what stops a file from repeating last year’s description of a company that has since moved.
Only then does the pricing analysis proper begin. The file has to indicate the most appropriate transfer pricing method for the category of transaction and the reasons for selecting it, so that the choice reads as a decision rather than an inheritance. Chapter II of the OECD Transfer Pricing Guidelines describes the comparable uncontrolled price, resale price and cost plus methods and the transactional net margin and transactional profit split methods, and the reasoning is expected to engage with them. The file has to indicate which associated enterprise is selected as the tested party, and why, because the method means nothing until it is clear whose margin is being tested. It then has to give the list and description of selected comparable uncontrolled transactions with the financial indicators relied on and the source of that information, and a description of any comparability adjustments made, which is the evidence the tested margin is measured against. Only with all of that in place does the closing element carry any weight: the reasons for concluding that the relevant transactions were priced on an arm’s length basis.
It must also include the local entity’s annual financial accounts for the year, audited where audited accounts exist, information and allocation schedules showing how the financial data used in applying the method tie back to those accounts, and summary schedules of the relevant financial data for the comparables with the sources from which they were obtained. Article 29g adds that both files are drawn up in Dutch or in English. That is a genuine convenience, and it is also where the most common failure originates.
The translation error
A group head office produces a global Master File and a set of country files generated from a single template. The Dutch finance lead receives the Netherlands chapter, confirms the entity name and the year are correct, and files it. Formally the group now has two documents. Substantively it has one document and a cover sheet.
The failure patterns are consistent. The functional analysis describes a distributor model the Dutch entity outgrew two years ago, or omits that Dutch staff now negotiate directly with regional customers. The benchmarking study uses a pan-European comparable set with no explanation of why that set fits the Dutch entity. The agreements annexed are group templates rather than the versions actually signed, and sometimes nothing was signed at all. The reconciliation to the Dutch statutory accounts is absent, so the margin tested cannot be traced to the figures in the return. Where intangibles are involved, the file asserts legal ownership without addressing DEMPE at all: development, enhancement, maintenance, protection and exploitation, and specifically who performs and controls those functions and who bears the associated risk. Legal title in a low-substance entity is not an argument; it is the thing the inspector is testing. The Dutch transfer pricing decree of 14 June 2022 states the point plainly: a legal owner that does not perform the relevant functions in respect of the asset can be allocated only a relatively limited return, and among the DEMPE functions development and enhancement will generally weigh more heavily in assessing the contribution to an intangible’s value than maintenance, protection and exploitation.
A Local File that could be lifted and dropped into any jurisdiction by changing the entity name is not documentation. It is a statement that nobody looked at the Dutch company.
When the files have to exist
Article 29g requires the group entity to take the Master File and the Local File into its administration, for the year to which the return relates, within the period set for filing the Dutch corporate income tax return. They are not filed with the return. They sit in the administration and are produced on request.
The practical consequence is that the analysis has to be contemporaneous. A benchmarking study built after an inspector’s first letter, using data selected in hindsight, carries considerably less weight than the same study run before the return was filed. Where the file is absent or materially deficient, the exposure is evidential as much as arithmetical. Article 52 of the Algemene wet inzake rijksbelastingen requires an administration from which rights, obligations and everything else relevant to taxation are clear at all times, and its third paragraph brings within that administration whatever is kept, recorded or drawn up under other tax laws. Where it falls short, article 52a allows the inspector to record the shortfall in an information decision open to objection, and once such a decision has become irrevocable, article 25(3) on objection and article 27e on appeal require the assessment to be upheld unless the taxpayer shows that, and to what extent, it is wrong. The taxpayer then has nothing of its own to rely on and carries a burden it would not otherwise carry.
The same discipline applies to any downward adjustment. Under article 8bb, in force since 1 January 2022, a downward adjustment of profit arising from a legal relationship with an associated body is left out of account to the extent the taxpayer does not make it plausible that a corresponding upward adjustment is included in a profit tax at that body. A file that still assumes a downward adjustment with no counterpart anywhere is documenting a position the statute no longer supports.
A group that crosses mid-year
Take a group with a Dutch operating BV that reports consolidated group revenue of EUR 44.6 million for FY2024. In July 2025 it acquires a competitor, and FY2025 consolidated group revenue closes at EUR 58.3 million. The Dutch BV’s own turnover is EUR 26 million, earned as a limited-risk distributor buying from a related principal.
Because article 29g tests the reporting year immediately preceding the year to which the return relates, the Master File and Local File obligation attaches for FY2026, with both documents required in the administration by the FY2026 return deadline. For FY2025 itself, the year of the crossing, no standardized package is required. Article 8b is unaffected: the BV must still hold information showing how its distribution margin and the new post-acquisition intercompany flows were set.
Suppose the BV reports an EBIT margin of 1.4 per cent, or EUR 364,000. Suppose further that a properly constructed transactional net margin study of comparable limited-risk distributors were to produce an interquartile range of 2.6 to 5.1 per cent. An adjustment to the lower quartile would give EUR 676,000, an uplift of EUR 312,000, which at 25.8 per cent, the rate article 22 applies to the part of the taxable amount above EUR 200,000, is roughly EUR 80,500 of additional tax for a single year. Repeated across three years, the ordinary period within which article 11(3) of the Algemene wet inzake rijksbelastingen allows the inspector to fix an assessment, extended by any filing extension granted, that is in the order of EUR 241,500 before interest, on a company that was, in its own view, compliant. The figures are illustrative; an inspector proposing an adjustment within a range need not stop at the lower quartile, and the FY2025 exposure is not reduced by the absence of a Local File. It is simply harder to defend.
The year before the threshold
The year in which a group approaches EUR 50 million is the year in which the work is least constrained. Functional interviews, agreements, benchmarking and reconciliations can be built while the facts are current and before article 29g attaches a deadline, so that the first mandatory Local File documents a position already tested rather than one assembled under pressure. Where ownership is reorganized in the same period, the documentation consequences of the restructuring belong in the Local File for the year in which it occurs, which is the point at which that exercise and the considerations in our guide to setting up a holding structure in the Netherlands meet.
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.