Dutch Supreme Court clarifies anti-abuse rule in dividend tax cases involving international holding companies

Context

Building on the Dutch Supreme Court rulings of 25 April 2025 (see our earlier website post), the decisions of 18 July 2025 offer further substantive guidance on the interpretation and application of Dutch anti-abuse provisions. They shed light on how Dutch tax courts should assess the EU concept of abuse and allocate the burden of proof.

The anti-abuse rule plays a central role in both the Dutch dividend withholding tax and the non-resident corporate income tax regime. These latest rulings are particularly relevant for foreign personal and family holding companies receiving dividends or realising capital gains from Dutch shareholdings. However, the Supreme Court’s reasoning may have broader implications for corporate structures more generally, especially where questions of beneficial ownership arise.

Relevant facts (simplified)

Both cases concern Belgian companies owned by members of respective families, each of which invested in a Dutch ‘feeder entity’ linked to a single Dutch private equity fund. In both instances, the Belgian companies and their shareholders had no involvement in the fund beyond the passive investment of capital.

One of the Belgian companies was largely inactive: it had no office, paid no directors’ fees, and held no assets other than two classic cars of modest value. The other company, by contrast, maintained its own office, paid substantial management fees (although it did not have own employees), and actively managed a portfolio of investments, including a small brewery.

The Dutch tax authorities (DTA) argued in both cases that the shares in the Dutch feeder entity were not held as business assets and that the structures constituted abuse, meaning that Dutch dividend withholding tax exemption should be denied. The District Court ruled in favour of the more active Belgian holding company, while siding with the DTA in the case of the passive holding company. On appeal, the Court of Appeal ruled in favour of the DTA in both cases.

Relevant Dutch anti-abuse rule

Under the Dutch implementation of the EU Parent-Subsidiary Directive (PSD), profit distributions by Dutch companies to qualifying non-resident corporate shareholders may be exempt from dividend withholding tax. The Dutch exemption is subject to the following basic conditions: (i) the shareholder must be resident in an EU/EEA or treaty jurisdiction; and (ii) the shareholder must be eligible to apply the Dutch participation exemption to the shareholding if it were subject to Dutch corporate income tax. The exemption is further subject to an anti-abuse provision, which must be interpreted in accordance with the EU concept of abuse as developed in EU case law.

In practice, the DTA applies this anti-abuse rule by (i) comparing the Dutch dividend tax outcome with a hypothetical scenario in which the non-resident holding company (or companies) is disregarded; and (ii) assessing the substance and activities of the non-resident holding company in relation to its Dutch shareholding.

Decisions of 18 July 2025

In the decisions of 18 July 2025, the Supreme Court upheld earlier rulings and confirmed the application of the doctrine as applied in the CJEU’s Nordcurrent case (see our earlier website post). It also clarified the allocation of the burden of proof in establishing whether an abusive structure exists, aligning it with principles established under EU law.

The Supreme Court clarified that even if an investment structure was not initially established with abusive intent, its continued maintenance – through certain developments or the addition of specific elements – may render it (partially) artificial over time. Importantly, the Supreme Court confirmed in light of the PSD that a structure need not be artificial in its entirety to fall within the scope of the anti-abuse provision; individual steps or elements may be deemed artificial, even if the structure as a whole was set up for legitimate commercial purposes.

In this context, the Supreme Court emphasised that the mere existence of a material business operation at the level of the non-resident shareholder is not, in itself, sufficient to rule out abuse. Rather, the shareholding in the Dutch company must be functionally attributable to those business operations in order to benefit from the dividend tax withholding exemption.

The Supreme Court upheld the Court of Appeal’s conclusion that the DTA had sufficiently demonstrated abuse and that the taxpayers’ counterarguments were insufficient to support business reasons for the investment structure as it relates to the particular Dutch shareholding.

Crucially, the Supreme Court explicitly accepted the Court of Appeal’s notion that the families behind the holding companies retained full discretion over whether to distribute or reinvest the dividends received from the Dutch entity. Disregarding the corporate governance structure, the courts consider this lack of obligation to reinvest, combined with the absence of final control over the use of the dividends, to contribute to the case of abuse.

Finally, it was raised by the claimants that non-resident personal holding companies were placed at a disadvantage compared to Dutch personal holding companies, which generally benefit from dividend tax exemption in otherwise comparable structures. The Supreme Court dismissed these arguments based on EU law.

Relevance and key take-aways

The Supreme Court decisions provide important clarification on the application of Dutch anti-abuse rules, reinforcing the need for a holistic, substance-over-form assessment in line with the EU abuse of law doctrine. These rulings underscore that the presence or absence of abuse must be evaluated in light of the broader factual and historical context of the structure.

Several key considerations emerge from the Supreme Court’s reasoning:

• Even where an investment structure is established and maintained for non-tax reasons, the presence of passive investments within such structures – particularly in family or personal holding companies – may still give rise to abuse concerns. The Supreme Court’s approach suggests that such structures are not immune from challenge solely due to there also being non-tax related considerations behind a structure.
• The Supreme Court placed significant weight on the fact that the ultimate decision-making power over the use of dividends rested with the family UBOs, rather than the holding company itself. This raises broader questions about whether corporate holding structures may be viewed as conduits, and how this aligns with the concept of beneficial ownership.
• Also important is that the Supreme Court ruled that a structure that was initially set up for legitimate business reasons may, as from a certain moment, be considered artificial.
• The Supreme Court upheld the Court of Appeal’s consideration that the use of intragroup management undermines the substance of a holding company. If adopted more broadly, this line of reasoning could have far-reaching implications for multinational structures that rely on shared service models or centralised management.

In conclusion, the implications of these rulings are potentially far-reaching, extending beyond family or personal holding companies. The rulings highlight the importance of a case-by-case analysis and may affect a wide range of cross-border investment structures investing into the Netherlands. Investors should carefully assess the functional substance and genuine link between their holding companies and Dutch passive investments in light of evolving Dutch case law.

Should you have any questions on this topic, please contact your regular trusted Loyens & Loeff adviser or one of the contact persons mentioned below.