The question arrives in the same form every few months: London or Amsterdam. It is usually asked by a group that already has a presence in both, or is about to, and wants to know which of the two should be treated as the centre of gravity. Framed as a contest, the question has no useful answer. Framed as a question about function, it resolves fairly quickly. The two cities do different jobs, they were doing different jobs long before the referendum, and the withdrawal of the United Kingdom from the European Union sharpened that division rather than creating it.
What London does well
London remains a deep pool of capital in the European time zone. That matters in the ordinary, unglamorous sense: when a group needs to place debt, syndicate a facility, run an equity process or find a buyer for a business with cross-border operations, the counterparties, the intermediaries and the pricing benchmarks are there. Depth of market is not a legal attribute and cannot be legislated into existence elsewhere.
English law is the second asset, and arguably the more durable one. Investors read English law documentation without translation, in both the linguistic and the conceptual sense. Facility agreements, shareholder arrangements, warranty packages and security documents follow patterns that credit committees in other markets recognise on sight. Dispute resolution follows the same logic: parties with no other connection to the United Kingdom continue to choose English law because the outcomes are reasonably predictable and the body of precedent is thick.
The third asset is the advisory ecosystem. Corporate finance, restructuring, insurance, specialist tax and listed-market compliance sit in one city at a level of specialisation that few markets sustain. For a group running a genuinely global platform, that concentration has practical value which survives any change in the treaty position.
None of this was affected by Brexit. Capital markets, contract law and advisory depth are not functions of EU membership, and the argument that they would migrate wholesale was never persuasive.
What Amsterdam does well
The Dutch proposition is narrower and more technical, and it is worth stating without embellishment. A Dutch company is an EU company. That is the first attribute, and it is the one the United Kingdom no longer has.
On top of that sit four things. The participation exemption exempts dividends and capital gains on qualifying shareholdings, subject to a minimum holding percentage and to the requirement that the participation is not a low-taxed portfolio investment, tested through the motive test, the reasonable subject-to-tax test or the asset test. It is mandatory and symmetrical: where a gain is exempt, the corresponding loss is not deductible. That symmetry is routinely overlooked when the regime is described as a benefit, and it is one reason the treatment of a holding should be settled before acquisition rather than after. The mechanics are set out in our note on the Dutch participation exemption.
The second is the treaty network, which is broad and, more relevantly, long established, so the administrative practice around it is settled. The third is the corporate tax system itself, with a headline rate of 25.8 per cent in the upper bracket and a reduced rate in the first bracket, applied within a framework that has absorbed the ATAD earnings stripping limitation on interest deduction, expressed as a percentage of fiscal EBITDA with a minimum threshold, and the Pillar Two minimum of 15 per cent for groups above 750 million in consolidated revenue.
The fourth is administrative: a tax authority with a functioning rulings practice, which since July 2019 has been conditioned on real economic nexus. No ruling is granted where the decisive motive is tax saving, and none is granted involving entities in listed jurisdictions. The value of that practice is not that it produces favourable answers. It is that it produces answers, in writing, before the transaction closes.
What Brexit changed in practice
Strip away the constitutional commentary and the operative change for corporate groups is narrow and specific. A UK company no longer has access to the EU directives. The regimes that removed withholding tax on qualifying intra-group dividend, interest and royalty flows between member states, more or less mechanically, no longer apply to a UK entity in either direction.
Brexit did not make London worse at what London does. It made the United Kingdom a third country for the purposes of a body of law that most groups had stopped noticing, precisely because it worked.
The fallback is the bilateral treaty network and domestic law, and that is not a like-for-like substitute. Directive relief was multilateral, uniform and largely self-executing. Treaty relief is bilateral, varies by counterparty and carries its own conditions: beneficial ownership, in many cases holding period and shareholding thresholds, frequently a limitation-on-benefits or principal-purpose test, and often a procedural burden. Relief by refund and relief at source are not the same product when the group is managing working capital.
The effect is felt in three flows. Dividends moving up a chain that crosses the Channel now depend on treaty entitlement rather than directive entitlement. Intra-group interest faces the same analysis, with the added complication that the deduction side is separately constrained by the earnings stripping rule. Royalties, typically the most mobile flow and the most closely scrutinised, are exposed on both the entitlement axis and the transfer pricing axis at once.
The Dutch side of the same flows
Symmetry matters here, and it is worth being explicit that the Netherlands is not a zero-withholding jurisdiction. Dutch dividend withholding tax applies at a general rate of 15 per cent, reduced under treaties and, within the EU, capable of exemption, but always subject to anti-abuse conditions that are applied rather than recited. Since 2021 there has also been a conditional withholding tax on interest and royalties paid to low-taxed or listed jurisdictions, which operates independently of the general regime and is aimed at conduit arrangements.
The point for a group weighing London against Amsterdam is that neither jurisdiction offers an automatic result. What Amsterdam offers is a defined and documented path through the EU framework, with an administration prepared to confirm the treatment in advance where the underlying facts support it.
Substance is what carries the structure
Everything above assumes something that cannot be assumed. Directive access, treaty entitlement and ruling certainty rest on the same foundation: that the entity in question does something real. Not an address, not a service agreement with a corporate services provider, not a board that meets by written resolution drafted in another country. Real activity means decisions taken by people with the authority and the competence to take them, in the place where the entity is established, with the financial capacity to bear the risks the structure allocates to it.
The direction of travel is one-way. The anti-abuse tests in the directives and in the treaties, the rulings policy in force since July 2019 and the transfer pricing framework all converge on the same enquiry. Article 8b imposes the arm’s length principle and a documentation duty with no threshold at all; Master File and Local File obligations begin at 50 million in consolidated revenue and country-by-country reporting at 750 million. A structure that cannot answer the substance question will not survive the transfer pricing question either. The point is developed in our note on Dutch substance requirements.
A worked example
Consider a group with three components: an operating business across several EU member states, a capital markets and treasury function that needs London, and shareholders in a mix of treaty and non-treaty jurisdictions.
Placing the EU holding function in a Dutch BV, properly capitalised and managed by a board resident in the Netherlands with genuine decision-making authority over the subsidiaries, allows dividends from the operating companies to reach the holding within the EU framework and, on distribution or disposal, to fall within the participation exemption where the qualifying conditions are met. The same entity sits inside the Dutch treaty network for distributions to shareholders outside the EU, subject to the anti-abuse conditions being genuinely satisfied rather than merely asserted.
The London component is not displaced. It continues to run financing, investor relations and the transactional work for which the market is there, contracting under English law where counterparties expect it. What changes is that the UK company is no longer the natural conduit for intra-group dividend, interest and royalty flows inside the EU perimeter, because it no longer has the legal access that made it convenient. Transactions between the two arms are priced at arm’s length and documented as such, which is a requirement rather than a refinement.
Two hubs, two jobs
The honest conclusion is that London and Amsterdam are not competing for the same mandate. London is a market: capital, law, expertise, counterparties. Amsterdam is a seat, the place from which an EU holding and treaty position is held and defended. A group that treats the first as a substitute for the second, or the second as a cheaper version of the first, tends to end up with a structure that does neither job properly.
Most groups of any size end up using both. The design question is not which city wins, but which functions belong where, and whether each entity has the people and the decisions to support the role it has been given.
Montclare structures and operates Dutch and cross-border platforms for international groups and private clients. 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.