Not every structure arrives through careful planning. Many are inherited from a previous adviser, copied from someone else’s arrangement, or entered quickly under commercial pressure, and the client only later asks whether it was right. Sometimes the answer is that it was fine. Often it is that it works today but carries a risk that has not yet surfaced. The useful question is not how it happened, but what can be done about it now, and the answer is usually more than the client fears and less than a clean start would have given.
Signature is where that stops being theoretical. A contract is executed, a deal closed, and only afterwards does anyone ask which entity signed and whether the signatory could bind it. What is in doubt is not the document but whether it binds the party everyone assumed. If it binds, the problem is the internal breach and the signatory’s own exposure; if it does not, the counterparty holds a document that may be unenforceable. Diagnosis comes first.
Who could actually bind the company
In a Dutch BV the board represents the company, and under article 2:240 of Book 2 of the Burgerlijk Wetboek that power belongs to each director individually unless the articles reserve it to named directors or require joint action. Once it exists the power is unlimited and unconditional so far as the law does not provide otherwise, and a permitted restriction may be invoked only by the company. A group mandate table is real internally and empty outside it.
What outsiders may rely on is fixed by statute. Article 22 of the Handelsregisterbesluit 2008 requires the register to show every director and whether he may represent the company alone or only jointly, and Directive (EU) 2017/1132 requires that disclosure across the Union. Article 25 of the Handelsregisterwet 2007 then stops a company invoking an outstanding or incomplete entry against a third party who did not know of it, as our note on signing powers and authority across borders develops.
First, assess honestly
The starting point is a clear-eyed assessment of what actually exists: what the structure does, what it was meant to do, where it is exposed, and how urgent each exposure is. This is not an exercise in blame, and a defensiveness about how the structure came to be only delays the diagnosis. Some problems are cosmetic, some are serious, and some are latent, fine until a bank, an investor or a tax authority looks closely. Sorting them by severity is the first real step.
Applied to a signature already given, the order is fixed. Which entity is named as party. Who signed for it, and whether he appeared in the register with sufficient authority on the signature date rather than today. Whether that authority was sole or joint, and if joint, whether the second signature arrived. Whether the approvals were minuted at the time.
Two findings recur. The register lags, since filings fall due within a week of the event, so a person can hold office before it shows and can have resigned while it still lists him. And the entity named is often not the one intended, because groups run companies separated by a single word. See our note on legal due diligence.
When the signature was not the one that counted
Conflict of interest is most often misread. Article 2:239 bars a director with a direct or indirect conflicting personal interest from the deliberation and the decision-making, passing the decision to the supervisory board or the general meeting. What it no longer does is touch the contract: the former article 2:256, under which a conflict affected the power to represent, lapsed on 1 January 2013.
Missing shareholder approval works the same way. In an NV, article 2:107a submits certain decisions to the general meeting and says expressly that its absence does not affect the representative authority of the board or its directors. For a BV no statutory list exists at all, and approval by another organ exists only where the articles create it.
Two provisions bite where one person sits on both sides. Article 2:247 requires an act between the company and the holder of all its shares, represented by him, to be recorded in writing, failing which it may be annulled for the company’s benefit. Article 2:7 makes an act exceeding the objects voidable, but only where the counterparty knew or should have known.
What can usually be fixed
A great deal can be corrected without unwinding everything. Missing substance can be built. Absent documentation, intercompany agreements, transfer pricing files, board minutes, can be prepared, though it is always better to document contemporaneously than retrospectively. Governance can be tightened, filings brought current, and a UBO register corrected. Much of what makes a structure fragile is fixable housekeeping, and doing it converts a risky structure into a defensible one, as we describe in our note on keeping a company in good standing.
Where the signature itself lacked authority, the repair is ratification. Article 3:69 of Book 3 of the Burgerlijk Wetboek lets the person in whose name an unauthorised agent acted ratify it, giving it the effect of an act done under a power of attorney. It fails once the counterparty has declared the act invalid for want of authority.
Most bad structures are not beyond repair. They are structures that were never finished, and finishing them is cheaper than the client fears and far cheaper than being caught.
Nor is the timing in the company’s hands, since any interested party may set a reasonable period and need not accept a partial or conditional ratification. Where an approval was never taken, adopt it now and record the conflict and the abstention rather than backdate a minute, as our note on governance design for Dutch holding companies sets out.
What is harder to undo
Some things resist correction. A transaction already completed on the wrong basis may have crystallized a tax cost that cannot be reversed. A structure built around a hybrid mismatch or an aggressive position that the rules have since closed may need genuine restructuring rather than repair, and restructuring has its own tax consequences that have to be modelled. The honest advice sometimes is that part of the structure has to be dismantled and rebuilt, and that doing so now, deliberately, is better than doing it later, under examination.
Form defects are the clearest example. Article 2:196 requires the issue and transfer of a BV share to be effected by deed before a civil-law notary holding office in the Netherlands. A privately signed agreement moves nothing, so a group that thought it had owned a subsidiary for years may find title and the register disagreeing. Hence the Dutch notary asks for everything.
Moving a contract to the entity that should have signed depends on the counterparty. Article 6:159 of Book 6 of the Burgerlijk Wetboek transfers a contractual relationship by deed with the cooperation of the other party, and article 6:155 makes a debt assumption effective against the creditor only once he consents. There is no unilateral novation.
Where the director is personally exposed
The signatory carries the first risk. Article 3:70 of Book 3 makes anyone acting as an agent warrant to the counterparty the existence and extent of his power of attorney, unless that counterparty knows or ought to understand it is lacking. Where ratification fails, the counterparty still has a debtor, and the measure of the exposure is the transaction.
Inside the company, article 2:9 makes every director liable in full for improper management unless no serious reproach attaches to him and he was not negligent in averting the consequences. Article 2:11 removes the shield a corporate director is assumed to give: that liability rests hoofdelijk, jointly and severally, on whoever was its own director when it arose.
Insolvency and tax run on shorter clocks. Article 2:248 makes every director jointly and severally liable for the deficit where the board manifestly failed and that plausibly caused the bankruptcy, and turns missing records or unpublished accounts into a presumption of causation. Article 36 of the Invorderingswet 1990 adds the same liability for wage tax and turnover tax, with article 7 of the Uitvoeringsbesluit Invorderingswet 1990 setting the notice of inability to pay at two weeks after the tax fell due. Our note on directors’ liability in a Dutch BV has the rest.
The substance question runs through everything
Very often the core defect is the same one: the structure was built for a tax outcome without the substance to support it. That is now the most dangerous single flaw a structure can have, because the rules that target it, from the principal purpose test to the Unshell directive, have all moved in the same direction. We deal with them in our notes on Dutch substance requirements and ATAD 3 and Unshell. Building real substance into a thin structure is frequently the single most valuable fix available.
A signature is where that abstraction becomes evidence, because the contract records which entity took a risk and therefore which should be paid for bearing it. Where a company signed an agreement its own people never negotiated, never priced and cannot monitor, the file says the function sits elsewhere. Article 8b of the Wet op de vennootschapsbelasting 1969 requires the data behind transfer prices to be kept.
The documentation duty on top of that is threshold-based. Article 29g requires a master file and a local file, in Dutch or English, within the period allowed for the corporate income tax return, for group entities of a multinational group with consolidated group revenue of at least 50,000,000 euro in the preceding reporting year. Below it, article 8b still applies.
Fix it before it is examined
The decisive factor is timing. A structure corrected on the client’s initiative, before a tax authority, a bank or a counterparty examines it, is a structure that has been put right. The same correction made after the examination has started is a concession under pressure, worth far less and read far worse. A client who suspects their structure is not sound has a window, and the value of acting within it, rather than hoping the question never comes, is the difference between fixing a problem and being caught with one.
The bank is usually the first reader, and it looks at a narrower set of documents than clients expect. Article 3 of the Wet ter voorkoming van witwassen en financieren van terrorisme requires client due diligence establishing whether the person representing the client is authorised to do so and, for a legal person, insight into its ownership and control structure. None of that is the agreement, as our note on what the KVK, the bank and the tax authority expect describes.
If you are facing this, we can help. Montclare handles exactly these situations for international businesses and families. Our services are set out on our services page.
This article is informational and does not constitute tax, legal or financial advice. The right course depends on the facts. Each engagement is subject to scope and applicable regulation.