Intragroup financing is where transfer pricing stops being an academic exercise and starts producing cash consequences. A licence fee five per cent too high is an irritation; an interest rate three hundred basis points too high on a forty million euro loan produces a permanent, annual and entirely visible difference. It is also where groups are most tempted to treat pricing as internal policy rather than evidence, since the money never leaves the consolidated perimeter. The Dutch tax authorities do not share that view, and neither does the file they will ask to see.
the question before the price: is it debt at all
Before any spread is discussed, the instrument has to survive characterisation. The Netherlands respects the civil law form of a loan in most cases, but form is not conclusive. Where an advance carries no realistic prospect of repayment, no fixed term and no consequence for default, the Dutch courts have long been prepared to treat it as something other than a loan, whether as a capital contribution or under the domestic doctrine of the unbusinesslike loan, under which a lender that accepted a risk no independent party would have accepted cannot deduct the resulting loss.
The practical test is not whether the documentation says loan, but whether the borrower, on its own projected cash flows, could service and repay it. If it could not, and the group advanced the funds anyway, the arrangement is priced as debt but behaves as equity. Debt capacity analysis, sized against the borrower’s own EBITDA rather than the group’s, belongs at the front of the file.
what the borrower would have paid on its own
The starting point is the credit standing of the borrower as a separate entity, not the rating of the group. A subsidiary with thin equity, customer concentration and negative working capital does not borrow at the parent’s rate because it shares a logo. Establishing an implied rating from the borrower’s own financial ratios, using published rating agency methodologies, is the conventional first move, and Chapter X of the OECD Transfer Pricing Guidelines proceeds on the same basis.
The rating then has to be adjusted for implicit support. A subsidiary whose failure would damage the group’s reputation or its access to funding is more creditworthy than its own balance sheet suggests, and lenders price that in. The usual approach is to notch the stand-alone rating upwards towards the group rating, the size of the adjustment driven by how integrated and strategically important the borrower is. A dormant holding company gets little; the operating company generating a third of group turnover gets a good deal more.
Once the rating is set, the comparables follow. The comparable uncontrolled price method dominates here, because market data is genuinely available: corporate bond yields by rating and tenor, syndicated loan pricing, and the group’s own third party facilities, often the most persuasive internal comparable in the file. The other recognised methods, resale price, cost plus, the transactional net margin method and profit split, have their place elsewhere, but a financing transaction benchmarked on anything else needs an explanation.
the terms that actually move the rate
A benchmark is useful only if the comparables share the economically relevant characteristics of the tested loan. Four terms do most of the work.
- Tenor. A ten year bullet loan is not comparable to a revolving facility, and a loan formally repayable on demand but outstanding for eight years is priced on the economics, not the label.
- Currency. The rate is built from the risk free reference rate of the currency of denomination. A euro benchmark applied to a Turkish lira loan is not a benchmark.
- Security and seniority. Unsecured and subordinated debt costs more. If the loan is in fact structurally senior and effectively secured on the borrower’s only asset, pricing it as mezzanine is unsupportable.
- Covenants and options. Prepayment rights, step ups, payment in kind features and financial covenants all have a price; a file that ignores them compares instruments that only look alike.
Two further points are specifically Dutch. First, article 8b of the Wet Vpb 1969 imposes both the arm’s length principle and a standing duty to hold documentation showing how prices were determined, with no turnover threshold; the smallest Dutch entity with a single intragroup loan is inside it, as set out in our note on the article 8b documentation obligation. Second, since 1 January 2022 the Dutch mismatch rules restrict downward adjustments where there is no corresponding upward inclusion in the counterparty’s jurisdiction. That change ended informal capital planning: a Dutch borrower on an interest free shareholder loan can no longer claim a deemed arm’s length interest deduction that nobody anywhere picks up as income.
guarantees: when a fee is due and when it is not
The distinction that matters is between an explicit guarantee and passive association. If the Dutch parent signs a legally enforceable guarantee that a third party lender relies on, the subsidiary has received something a stranger would have charged for, and a fee is due. If it simply benefits from being known to belong to a solid group, that is implicit support, and it is not a chargeable service. Groups get this wrong in both directions: charging nothing where a signed guarantee exists, or charging a fee that captures the entire interest saving, including the part attributable to implicit support.
The workable method is the yield approach with an implicit support correction: take the rate the borrower would pay stand alone, adjust its creditworthiness upwards for implicit support, and compare that adjusted rate with the guaranteed rate. Only the remaining differential is attributable to the explicit guarantee, and the fee sits inside that band, bounded by the guarantor’s expected cost and the borrower’s saving. Charging the full raw differential is the common overreach.
A treasury company that borrows at four per cent and on-lends at nine has not priced risk; it has relocated profit, and the file will have to explain why that is not what happened.
cash pooling: the leader rarely earns the spread
The benefit of a cash pool is a synergy created by the participants collectively: balances net off, the group borrows less externally and the spread between debit and credit positions narrows. The question is who is entitled to it.
In most structures the pool leader coordinates: it operates the accounts, arranges the netting and manages the banking relationship. Where it does not meaningfully bear the participants’ credit risk and does not control the risks it nominally assumes, it is a service provider and should earn a routine reward, not the spread. The reasoning is familiar from the DEMPE framework applied to intangibles, development, enhancement, maintenance, protection and exploitation: what matters is not who signs the contract but who performs the functions, controls the risk and can bear it financially.
Two further points recur in reviews. Balances that never reverse are not pool positions; a participant in credit for four consecutive years has made a term deposit and should be priced as one. And the synergy benefit should be allocated among participants, typically in proportion to their contributions, rather than accumulating in the leader by default. Groups running treasury through a European holding company, a structure covered in our guide to holding structures in the Netherlands, should expect that allocation to be examined directly.
a worked example: three hundred basis points with nothing behind them
Take a Dutch operating company that borrows 40 million euros from a group finance company under a five year unsecured term loan at 8.5 per cent fixed, the stated rationale being that the debt is unsecured and subordinated to bank facilities. The benchmarking memo in the file is a one page table of unrelated corporate bond yields with no rating analysis. The figures that follow are illustrative and assume no other deduction limitation applies.
On review, the borrower’s own ratios support an implied stand-alone rating in the BB range, notched to BB+ for implicit support given its weight within the group. On the facts assumed, comparable five year euro debt for that profile prices at 5.5 per cent. The differential is 300 basis points, or 1.2 million euros of interest a year.
- Annual disallowed interest: 40,000,000 x 3.0% = 1,200,000 euros
- Additional Dutch corporate income tax at 25.8%: 309,600 euros per year
- Over the five year term, before interest on underpaid tax: 1,548,000 euros
The correction increases Dutch taxable profit, so the 2022 mismatch rules, which bite on downward adjustments, do not stand in its way. The harder problem is the other side: the lender’s jurisdiction has already taxed the full 8.5 per cent, and no corresponding downward adjustment follows automatically. Unless relief is obtained through the mutual agreement procedure, which takes time, the group is taxed twice on the same 1.2 million euros. Benchmarking at inception would have cost a fraction of one year’s exposure.
what the file has to contain
Groups with consolidated revenue of 50 million euros or more must maintain a Master File and a Local File; consolidated revenue of 750 million euros or more brings Country-by-Country Reporting. Below those thresholds the article 8b duty still applies in full, and for financing transactions the substantive content is much the same either way: the loan agreement, a debt capacity assessment, a credit rating analysis with the implicit support adjustment reasoned rather than asserted, a benchmarking study with dated and reproducible comparables, and, for pools and guarantees, a functional analysis that says plainly who controls which risk.
None of this requires exotic structuring, and groups should be sceptical of anything promising a lower effective rate as its principal output. What it requires is that the numbers in the intercompany agreements trace back to evidence that existed when the rate was set. Our transfer pricing practice works on that basis: a defensible file is cheaper than a contested one.
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.