Every few years someone declares the Dutch holding company finished. The declarations follow a pattern: a directive is implemented, a conduit route closes, a headline rate fails to fall, and the conclusion is drawn that the jurisdiction has lost its purpose. The conclusion is wrong, but the premise is not. What the Netherlands offered fifteen years ago and what it offers now are different propositions, and a group working from a file assembled before 2019 is reasoning about a country that no longer exists.
What the participation exemption actually does
The deelnemingsvrijstelling exempts dividends and capital gains on qualifying shareholdings. It requires a minimum shareholding percentage and that the holding is not a low-taxed portfolio investment; a participation that would otherwise fail is admitted through the motive test, the subject-to-reasonable-tax test or the asset test. Those are the mechanics, and they are widely reproduced elsewhere. The mechanics are not where the value sits.
Two features distinguish it. First, it is mandatory rather than elective. There is no annual claim and no risk that relief is lost because a form went unfiled in a year when the group was distracted by an acquisition; where the conditions are met, the exemption applies as a matter of law. For a group planning a disposal several years out, that is worth more than a marginally wider definition of a qualifying holding somewhere else.
Second, it is symmetric. If gains are exempt, losses are not deductible. This is the part omitted from the promotional literature and it is the part that occasionally decides the structuring question. A group with a portfolio of venture positions, some of which will be written off, is not obviously well served by a regime that exempts the winners and disallows the losers. The exemption is a feature for a holder of durable operating subsidiaries and an irrelevance, sometimes a cost, for a holder of speculative ones. The conditions and the tests are set out in more detail in our note on how the participation exemption works in practice.
The treaty network, and what it no longer buys
The Dutch treaty network is broad, and for a group with subsidiaries spread across many jurisdictions the practical effect is a reduced and reasonably predictable withholding cost on inbound flows. That is a genuine operating advantage. It is also a smaller advantage than it was, and honesty requires saying why.
Treaty access is now conditioned. Anti-abuse conditions in the treaties themselves and in domestic law have between them removed the arrangement in which a Dutch entity with a mailbox and a part-time director sat between an investor and an operating company for the sole purpose of improving a withholding rate. Since 2021 a conditional withholding tax applies to interest and royalties paid to low-taxed or listed jurisdictions. The general dividend withholding rate remains 15%, reduced under treaties and with exemptions available within the EU, and those exemptions are themselves subject to anti-abuse conditions. The interaction between the withholding regimes is set out in our note on Dutch withholding tax on dividends, interest and royalties.
The network is therefore now useful to groups that would have existed anyway and useless to those constructed to exploit it. From the perspective of a genuine investor that is an improvement, since conduit structures were always a shared reputational liability.
Company law that does what shareholders need
The corporate law argument is underrated because it is unglamorous. The BV permits share classes with tailored voting and economic rights, allowing a family to retain control while distributing return, or a founder group to keep board appointment rights through successive funding rounds. The stichting administratiekantoor certifies shares and separates control from economic entitlement, which addresses the employee participation problem and the succession problem through a single instrument. The cooperatie retains its uses for certain investor groups. The NV, the CV and the VOF each have narrower but real applications.
Incorporation runs through a Dutch civil law notary and registration with the KVK. The notarial requirement is often described as friction; it is better understood as insurance, since the deed is passed by a professional with statutory duties and the resulting title is difficult to challenge.
The UBO register is maintained by the KVK. General public access was restricted following the Court of Justice judgment of November 2022, and access now runs to competent authorities and obliged entities. Foreign shareholders should assume registration, not anonymity, and plan on that basis.
What the regime demands in return
The ruling practice is where the change of character is clearest. Since July 2019, advance certainty requires genuine economic nexus with the Netherlands. Rulings are not granted where the decisive motive is tax saving, and not granted in relation to entities in listed jurisdictions. The consequence is that a group with real people, real decisions and real risk in the Netherlands can obtain a documented position from the tax authority on a defined question, and a group without those things cannot.
The Netherlands stopped competing on price some time ago. It competes on the predictability of the answer, and predictability is only sold to those with something real to be predictable about.
Around that sits the ordinary compliance architecture. Article 8b imposes the arm’s length principle and a documentation duty with no threshold, which means every intercompany charge in every Dutch entity needs support, however small the group. Master File and Local File obligations begin at 50 million of consolidated turnover, country by country reporting at 750 million, and the Pillar Two minimum of 15% applies to groups at or above the same 750 million threshold. Earnings stripping limits interest deduction to a percentage of tax EBITDA subject to a floor; both parameters have moved more than once, which is itself a planning consideration for any structure relying on leverage. DAC6 requires reporting of cross-border arrangements bearing defined hallmarks, with the obligation falling on the intermediary or, failing that, on the taxpayer. What operational substance looks like is covered in our note on Dutch substance requirements.
VAT deserves a line because it is routinely missed at the modelling stage. A pure holding company that only holds shares is generally not a taxable person and recovers no input VAT, so its advisory and audit costs are borne gross. A holding company supplying management services to its subsidiaries for consideration carries on an economic activity, and the position changes. The VAT group requires financial, economic and organisational links; a common shareholder alone does not create one.
The rate is not the argument
Dutch corporate income tax stands at 25.8% in the upper bracket, with a reduced rate on the first tranche of profit. Nobody arrives for that. Any adviser presenting the Netherlands as a low-tax destination is either uninformed or selling something, and a CFO who hears that pitch should treat it as a signal about the adviser rather than about the jurisdiction.
The case rests instead on the treatment of the specific flows a holding company actually experiences: exempt subsidiary dividends, exempt disposal proceeds where the conditions are met, treaty-reduced inbound withholding, and a deductible cost base. For a holding entity, taxable profit is often modest precisely because the principal receipts are exempt. The headline rate applies to a base that a holding platform of this kind largely does not generate.
A worked comparison
Consider a mid-market industrial group, family owned, with manufacturing in two EU member states, a distribution arm in Asia and an acquisition pipeline in central Europe. It is choosing between a Dutch holding platform and an alternative EU seat offering a narrower treaty network but a lighter administrative burden and a lower quoted running cost.
The alternative wins on annual cost, speed of incorporation and the volume of local documentation. Those are real advantages. The Dutch platform wins on four others. Exit treatment on the acquisition pipeline, where the exemption conditions are met, is exempt and mandatory, with no election to preserve across a decade. Inbound withholding from the distribution arm may be lower depending on the applicable treaty, and any differential compounds annually rather than arising once. Governance instruments exist to hold the family’s control intact through a generational transfer without fragmenting economic ownership. And on the eventual sale of a division, the acquirer’s diligence team encounters a jurisdiction whose holding regime it has priced before, which shortens negotiation on tax indemnities.
The decision turns on whether the group has, or is willing to build, genuine decision-making presence in the Netherlands. If the answer is no, the alternative seat is the better answer and the Dutch platform is an expensive way to acquire the same exposure with more filing. If the answer is yes, the cost differential is recovered on the first material disposal.
Substance as a competitive position
The Netherlands did not become a substance jurisdiction by choice; it was pushed there by international standard setting, by the European Commission and by its own political reckoning with conduit flows. The outcome, however, favours the groups that were never the problem. A regime that admits everyone offers certainty to no one, because its positions are perpetually under external attack. A regime that admits only those with real activity can afford to defend the positions it grants.
For a group with operations, employees and capital genuinely deployed, the Netherlands today offers a mandatory participation exemption, a broad treaty network that survives scrutiny, corporate instruments equal to most shareholder problems, and an administration that will discuss a position before it is taken. It does not offer a low rate, and it no longer offers a shortcut. Those who mourn the shortcut were never the intended clientele.
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.