A client you have advised for years tells you they are expanding into the Netherlands. Nothing about the instruction looks exotic. You have handled their acquisitions, their reorganizations and their disputes, and you know their business better than any outside adviser could. The Dutch piece appears to be an extension of work you already do well. So you read the relevant statute, you speak to a colleague who has seen a similar structure, and you form a view.
That is the moment the exposure begins, and it does not announce itself. The advice you give will look competent, because it is given competently. The problem is not the quality of the reasoning. It is that the reasoning is applied to a system whose written rules you can read and whose actual practice you cannot see. What follows is where that gap shows up, what it costs, and why handing the Dutch layer to someone who lives inside it is a decision about risk rather than a concession about capability.
The part of competence that does not cross the border
A statute is the smallest part of what a good adviser knows. The larger part is judgment about how that statute is applied: which authority reads it strictly and which reads it purposively, which arguments are accepted in practice, how much documentation is enough, what a registrar returns and what a registrar lets pass. None of that is written down. It is acquired by working inside the system repeatedly, over years, and watching where files succeed and where they are sent back.
That knowledge is precisely the part that does not travel. A translated text of Dutch company law is accurate and almost useless on its own, because it tells you what the rule says and not what the rule requires of you. The foreign adviser reading it is not ignorant. They are working with a partial instrument, and the incompleteness is invisible from where they are standing, because the text reads as though it were self-sufficient.
This is why the error pattern here is so specific. Foreign advisers rarely get Dutch law spectacularly wrong. They get it plausibly wrong. The structure they design is defensible on the face of the legislation and misaligned with how the legislation is administered, and that misalignment surfaces later, in front of a bank or an inspector or a buyer’s due diligence team, when it is expensive to correct.
Substance is where it shows first
Substance is the clearest example. The requirements read like a checklist, and in their written form they are one. For a company whose activities consist mainly of receiving and paying interest, royalties, rent or lease payments from or to group entities established outside the Netherlands, with activities connected to holding participations left out of that assessment, article 3a, seventh paragraph, of the Uitvoeringsbesluit internationale bijstandsverlening bij de heffing van belastingen sets out the requirements on its presence in the Netherlands, among them that at least half of its statutory, decision-making directors live or are actually established in the Netherlands, that board decisions are taken there and that the bookkeeping is kept there. For that company, where it invokes or can invoke a tax treaty or other arrangement for the avoidance of double taxation in force in relation to the Netherlands, Council Directive 2003/49/EC on interest and royalties or a national provision implementing that directive, the list is applied item by item: the same article requires it to state in its tax return whether it met every one of those requirements throughout the year, and, if it failed any of them at any moment in the year, to identify which ones and to supply the further information the article lists; under article 8, fifth paragraph, of the Wet op de internationale bijstandsverlening bij de heffing van belastingen, that information is provided to the Minister of Finance with a view to implementing directives of the Council of the European Union or other arrangements of international and interregional law on mutual assistance in the levying of taxes. A competent foreign adviser will see a list of this kind and satisfy each item on paper, although some items, such as the requirement that board decisions are taken in the Netherlands, describe what must actually happen there. What a list satisfied on paper cannot show is where the company’s decisions are genuinely taken, which is the question underneath, and the listed items are evidence of that, not a substitute for it.
The consequence is a structure that answers every listed condition on paper and still fails the question the conditions were written to test. Directors are appointed and never actually decide anything. Board meetings are recorded in minutes drafted abroad and adopted without discussion. An office exists at an address that serves the entity and nothing else. Every element is present on paper and the whole is unpersuasive, which is exactly the outcome the rules were designed to produce.
A foreign adviser has no reliable way to calibrate this from outside. The list does contain figures, including a wage cost of at least 100,000 euro for the activities concerned and an office in the Netherlands, at the company’s disposal for at least 24 months, in which those activities are actually carried on. No figure measures whether decisions are genuinely taken where the minutes say they are. What there is instead is an accumulated sense of what an assessment actually looks like when it happens, what an inspector reads first, and which arrangements have held up. That is local knowledge, and its absence is not a defect in the adviser.
Powers, and who can actually bind the entity
Corporate representation is the second recurring failure, and it is quieter. Every jurisdiction has its own logic about who may commit a company, how that authority is recorded, how it may be limited, and how much a counterparty is entitled to rely on the public record. Advisers carry the logic of their own system with them without noticing, because it feels like a general principle rather than a local rule. In the Netherlands the last of those questions turns on article 25 of the Handelsregisterwet 2007: subject to the exceptions that article itself lists, a fact that must be made public by registration or filing cannot be invoked against a third party who was unaware of it until the registration or filing, and where it applies the announcement referred to in article 24, has taken place, and the persons that article names, among them the registered legal entity, cannot hold the incorrectness or incompleteness of the registration or filing, or of that announcement, against a third party who was unaware of it.
So a foreign adviser drafts a power of attorney that would be unimpeachable at home, and it arrives at a Dutch notary in a form that cannot be used for the act in question. Or a commitment is given by a person whose authority was jointly held, in a way the register would have shown to anyone who looked. The articles of a BV may provide that a director can represent the company only together with one or more others, as article 2:240 of the Dutch Civil Code allows, and an adviser used to a different default may not think to check. Or the scope of the power omits the one faculty the transaction turns on, and the discovery is made on the day of signing.
These are not catastrophic errors in isolation. They are delays, and delays in a transaction have a price the client can calculate precisely. What the client cannot see is that the delay was avoidable, which is fortunate for the adviser, until the next one.
The deeper issue is that representation questions are checked by the parties least willing to be flexible about them. A bank’s lawyers read a power against the facility they are about to advance. A notary reads it against the deed they are about to execute and their own liability for having done so. A registry reads it once more before anything is recorded. Each reads for a different purpose, and a document that satisfies one can be returned by the next. A foreign adviser drafting for the first of those readers, without knowing that the other two exist in this form, will produce something that passes and then stops.
The corporate calendar nobody put in a diary
There is a phase after a structure is set up in which nothing appears to happen. Accounts must be prepared, adopted and filed; for a BV, article 2:210 and article 2:394 of the Civil Code set that sequence, subject to the exemptions Book 2 itself provides and to the rule in article 2:394 on making public accounts that have not yet been adopted, and article 2:394 fixes an outer limit of twelve months after the end of the financial year for making the accounts public. Resolutions must be passed. Changes must be notified: article 19 of the Handelsregisterwet 2007 requires those obliged to do so to make the filings needed for the registered data to be correct and complete at all times. Registrations must be maintained. Each obligation is administrative and none of them is intellectually demanding, which is why they fall between advisers so easily.
The foreign firm assumes the work ended when the structure was delivered. The client assumes their adviser is watching. Nobody is watching. The obligations accrue silently and the entity drifts out of good standing without any single moment at which someone made a mistake, which is the characteristic of this failure and the reason it is so common.
What makes it dangerous is that the consequences are not proportionate to the effort involved. Unfiled accounts can bear on directors personally. In a bankruptcy of the company, article 2:248 of the Civil Code treats a board that has not met its filing obligation under article 2:394 as having performed its task improperly and presumes that this was an important cause of the bankruptcy, which opens the directors to joint and several liability towards the estate for the deficit, subject to the three-year window before the bankruptcy, the individual defence, the disregard of unimportant omissions and the court’s power to reduce the amount that the same article contains. An entity whose register entries are out of date can be a problem for a bank review, a counterparty check or a sale process. The cost is not incurred at the moment of the omission. It is incurred later, all at once, at whatever moment happens to require the entity to be clean.
What the bank asks, which company law does not answer
A structure can be perfectly lawful and unbankable. Banks apply their own onboarding standards, and those standards are not derived from company law or tax law. Much of their statutory frame comes from the Dutch anti-money laundering act: article 3 of the Wet ter voorkoming van witwassen en financieren van terrorisme requires a bank to carry out customer due diligence, among other cases when it enters into a business relationship, and that due diligence must enable it, among other things, to identify the ultimate beneficial owner, to take reasonable measures to understand the ownership and control structure of a corporate client, to establish the purpose and intended nature of the business relationship and to monitor that relationship on an ongoing basis, examining the source of the funds where necessary. How each bank applies that frame is its own policy. The standards concern the beneficial ownership chain and how clearly it can be evidenced, the commercial rationale for a Dutch entity, the source of the funds, the countries involved, and whether the entity has any real connection to the place where it is asking to bank.
A foreign adviser can read that statutory frame, but not the appetite that each bank applies on top of it, because the appetite is not in the legislation. It is in the practice of institutions that publish little and change their appetite without notice. So the structure is incorporated and then it cannot open an account, and the client, reasonably, does not distinguish between a legal question and a banking question. They asked for a working Dutch company and they were given one that does not work.
The remedy at that stage is usually a partial rebuild, which is more expensive than doing it in the right order and harder to explain, because the explanation involves telling a client that the advice was correct and the outcome was still wrong.
What the firm is actually exposed to
The professional exposure is worth naming directly. Advising on a jurisdiction in which the firm is not qualified sits uncomfortably with most regulatory frameworks governing competence, whatever the disclaimer at the foot of the memorandum says. Professional indemnity cover frequently excludes or qualifies advice on foreign law, and that exclusion is examined at the moment of a claim, when it is too late to restructure the engagement.
There is a softer exposure that matters more in practice. The client does not parse the boundary between what you advised on and what merely happened while you were their adviser. If the Dutch entity fails a bank review or is challenged on substance, the firm that has held the relationship for years is the firm the client turns to, and the conversation is not about scope of engagement. It is about whether they were well served.
That is the real cost. Not litigation, which is rare, but the erosion of confidence in a relationship that took a decade to build, over a piece of work that represented a small fraction of the fees and none of the firm’s actual strength.
Delegation as a decision, not a concession
Firms bring in specialist counsel constantly and think nothing of it. Tax counsel is instructed on a tax question, litigation counsel on a dispute, competition counsel on a filing. Nobody treats those as commercial defeats, because the client understands that a serious firm knows the limits of its own desk and manages them openly.
The Dutch layer is the same category of decision, and it is treated differently only because a foreign jurisdiction feels like something a capable generalist should be able to absorb. It is not. It is a specialism that happens to sit behind a border, and the sensible response is the one the profession already applies to every other specialism: instruct someone who does it every day, define the boundary in writing, keep the client relationship, and stay in control of the file.
Knowing where your competence ends is not the absence of competence. It is the most reliable evidence of it. The firm that delegates the Dutch layer keeps its client, keeps its judgment on the commercial question, and removes from its own balance sheet a risk it was never paid to carry.
This article is informational and does not constitute tax, legal or investment advice. Each engagement is subject to scope and applicable regulation.