A newly incorporated Dutch BV is usually treated internally as a project still in progress. The bank account is open, the first staff have been hired, the group is deciding which functions will migrate and when, and the tax function is reasonably occupied with registration, VAT identification and the first payroll filings. Transfer pricing, in that sequence, tends to be scheduled for later, once the entity has a full financial year behind it and there is something to benchmark. That sequencing is the single most common defect we see in first-year files. The obligation in article 8b of the Wet op de vennootschapsbelasting 1969 does not wait for a structure to settle. It attaches to the first intercompany invoice, and by the time a group is comfortable enough to look at pricing seriously, the facts that the documentation is supposed to describe have already happened.
the obligation attaches to the first transaction
Article 8b contains two limbs, and the second is the one that catches new entrants. The first limb is the arm’s length principle itself: related parties must deal on terms that independent parties would have agreed. The second is a documentation duty, requiring the taxpayer to hold in its administration the information showing how the conditions of its intercompany transactions were determined and why those conditions are at arm’s length. There is no turnover threshold attached to either limb, and no exemption for a stub year. A BV that issued a single management fee invoice of EUR 40,000 in its first four months of trading is within scope on exactly the same terms as one turning over EUR 200m. We set out the mechanics of that duty in more detail in the foundational article on the article 8b obligation.
The formal reporting layers sit above this and are threshold-driven. Master File and Local File become compulsory where the group’s consolidated turnover reaches EUR 50m, and Country-by-Country Reporting where it reaches EUR 750m. A group already operating at scale abroad will therefore cross the EUR 50m line in the Dutch entity’s very first year, notwithstanding that the entity itself has been trading for four months. The Local File has to describe transactions that were priced before anyone in the group had thought about how to describe them.
A first year is not a grace period. It is the year in which every fact that the documentation will later have to describe is being created, mostly by people who do not know they are creating it.
deciding what the entity actually is
Before any pricing question can be answered, the group has to reach a defensible view of what the new BV is in functional terms. Four characterisations cover most of what arrives in the Netherlands, and they carry materially different return profiles.
- Limited risk distributor. Buys finished goods from a group principal and resells locally, with no inventory obsolescence risk, no bad debt risk and no pricing autonomy. Remunerated on a modest, stable operating margin, typically tested with TNMM.
- Service provider. Renders marketing support, back office, technical or regional coordination services to affiliates. Priced on a cost plus basis, on a defined cost base, with a mark-up supported by a benchmark.
- Principal or entrepreneur. Holds inventory title and market risk, controls pricing and takes the residual. This characterisation only survives scrutiny if the people who exercise control over those risks are genuinely located in the Netherlands and have the authority and capacity to exercise it.
- Holding company. Holds participations, may on-lend, may hold intangibles. Here the DEMPE analysis becomes decisive: development, enhancement, maintenance, protection and exploitation functions determine where intangible-related return belongs, and legal title alone determines very little.
The characterisation cannot be chosen from a menu. It has to follow the substance the group is genuinely putting in place, which is why this analysis belongs alongside the structuring decision rather than after it. The considerations that drive entity selection in the first place are covered in our guide to setting up a holding structure in the Netherlands.
contracts signed before the entity trades, not after
Intercompany agreements executed retrospectively are a recurring weakness in first-year files. They are not worthless, but they invite the question the taxpayer least wants asked: if the allocation of risk was clear at the outset, why was it recorded eighteen months later, and does the document describe what the parties did or what they later wished they had done? Conduct governs in any event, so an agreement that diverges from observed behaviour weakens the file rather than supporting it.
For a first-year entity the minimum set is short. A distribution agreement or services agreement setting out scope, territory, term and the pricing mechanism. A clause dealing with periodic true-ups, because provisional prices set on budgeted volumes will almost always need correcting. A risk allocation that matches the functional analysis, so that a limited risk distributor is not left carrying inventory write-downs in the contract while the benchmark assumes it carries none. Where financing is involved, a loan agreement with a stated rate, term and security position. These documents should be dated before the first invoice, because that is when the parties actually assumed the positions they describe.
a worked first year: a Benelux distribution subsidiary
The figures that follow are illustrative. Take a group with consolidated turnover of EUR 310m that incorporates a Dutch BV in January to distribute finished goods in the Benelux, buying from an existing German principal. Master File and Local File obligations therefore apply from year one; CbCR does not.
The BV is characterised as a limited risk distributor. A benchmarking study prepared for the entity supports an arm’s length operating margin range of 1.8% to 3.4%, with a median of 2.5%. Intercompany prices are set in January on budgeted volumes, and nobody revisits them.
The first year closes as follows. Third party revenue EUR 14,200,000. Cost of goods purchased from the principal EUR 11,568,000. Local operating expenses EUR 1,780,000. Operating profit is therefore EUR 852,000, an operating margin of 6.0%. Volumes ran well ahead of budget and local costs came in below plan, so the entity has substantially outperformed its target return.
At 2.5%, the arm’s length operating profit would be EUR 355,000. The excess is EUR 497,000. The group’s instinct is to issue a year-end debit note from the principal to bring the BV back to the median. That would reduce the Dutch base by EUR 497,000 and, at the 25.8% top-bracket rate, some EUR 128,226 of Dutch corporate income tax with it.
Two problems follow. First, there is no contractual basis for the adjustment; the distribution agreement was never signed, so there is no true-up clause to invoke and no documented mechanism explaining why a payment is due. Second, and more seriously, the Dutch mismatch rules in force since 1 January 2022 restrict downward adjustments to Dutch taxable profit where there is no corresponding upward inclusion in the counterparty’s jurisdiction. If the German principal does not pick up the EUR 497,000, the Dutch deduction is not available. The regime that ended informal capital planning operates precisely here, on the year-end correction that only moves in one direction.
The point of the example is not the tax saved or lost. It is that the outcome was determined in January, by the decision not to document the pricing mechanism before invoicing began. Had the agreement contained a quarterly true-up clause, prices would have been corrected in instalments during the year, on invoices the German entity recognised as income as they were issued, and the result would have been symmetrical, contemporaneous and unremarkable on review.
what reconstruction costs in the second year
Groups that defer the analysis discover that the second-year exercise is not the same exercise done later. It is a different and worse one. The functional analysis has to be reconstructed from memory, from people who may have moved on, about decisions taken in a period when the operating model was still shifting. The choice of method has to be justified against transactions already invoiced, which means the method is being fitted to the outcome rather than the outcome derived from the method. Comparables have to be selected for a year that has closed, with the search inevitably informed by the answer already on the ledger.
None of this is fatal, and files are routinely rehabilitated. But the group has moved from a position where it selects the most appropriate of the recognised methods, CUP, resale price, cost plus, TNMM or profit split, on the facts, to one where it defends a single method against a Belastingdienst reviewer holding the same numbers and no attachment to the group’s preferred characterisation. Where the documentation duty has not been met, the practical consequence is a materially weaker evidential position on an assessment the taxpayer then has to displace.
a workable first-year sequence
The sequence that avoids all of this is short and belongs in the first quarter of trading: settle the functional characterisation before the first invoice; execute the intercompany agreements that reflect it; commission or refresh the benchmark supporting the target return; write a pricing policy that states the tested party, the method, the profit level indicator, the cost base and the true-up cadence; and instruct finance to monitor the realised margin quarterly rather than annually. Local File preparation then becomes a drafting exercise against an existing record, to be completed by the time the first corporate income tax return is due, rather than an investigation into the group’s own recent past.
None of this produces a lower effective rate, and it is not meant to. It produces a file in which the return reported in the Netherlands corresponds to the functions actually performed there, and in which that correspondence can be demonstrated from documents written at the time rather than reasoned backwards from the closing accounts.
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.