Given the breadth of the directive, the carve-outs set out in Article 6(2) play a crucial role in determining whether an undertaking is within the scope of the rules. In an ideal world, this first filter should take out genuine holding vehicles.

• If it remains, the regulated financial undertakings carve-out will do a lot of the heavy-lifting for the industry (taking out AIFMs, UCITS and AIFs amongst others), however, under the current drafting the carve-out only applies on an entity- by-entity basis. This means holding vehicles or entities that may not themselves be regulated in a fund structure (even though they may be under common management with a regulated entity) are potentially exposed to the effects of the Directive.
  
  We understand the Commission is reviewing its position on this aspect and may adopt the approach settled on by the OECD Pillar Two proposals which employs a broader definition of investment funds and their holding structures. The European Parliament amendments first published in May 2022 offered some potential respite with an amendment proposed that would mean entities owned by regulated financial undertakings with a purpose of holding assets or investments would also be exempt (Amendment 15). The September 2022 edition of amendments were less encouraging with proposals to remove the regulated financial undertakings exemption entirely (Amendments 105 and 106) or restrict the carve-outs to undertakings not taking advantage of directive tax benefits (Amendment 104). The final amendments endorsed by the Parliament in November 2022 left the exemption unchanged. The spectrum of amendments is indicative of the political camps within the European Parliament. It should be emphasised that the amendments are advisory for the Council rather than binding.
  
  
• The directive makes some attempt to carve-out domestic holding companies, however its operation, as the name suggests, has limitations. Under Article 6(2)(d) only undertakings with holding activities that are resident in the same Member State as the undertaking’s shareholders or ultimate parent entity are excluded. 
  
  That said, as it currently stands most commentators expect this should allow a double tier of entities resident in the same location to qualify for the domestic holding company exemption where the top company is not itself caught by the directive i.e. it is a regulated financial undertaking or it satisfies the minimum substance requirements. A typical Master Lux Co with sufficient substance along with a Luxembourg subsidiary may therefore satisfy this test.
  
  
• Another carve-out from the directive relates to full time employees. Under Article 6(2)(e) undertakings with at least five full-time equivalent employees or members of staff exclusively carrying out the activities generating the income are not subject to the requirements of the directive. 
  
  Although the current draft of the directive does not impose any requirements around the location of those employees or what role they play, the reality is there will be few entities below the fund that employ staff. Often staff will be employed by a local service company providing services across a number of different entities and different funds and whilst commercially this makes sense - allowing employees to roll-on and off funds without having to continually change employment contracts - the current directive takes a much more limited view by testing employees on an entity-by-entity basis. 
  
  The number of staff required (5) under the current directive is arbitrary and bears no relation to the size or industry. It will be interesting to see whether this requirement changes.

If it is not possible to satisfy one of the carve-outs the next filter to work through is the gateway test. The gateway test set out under Article 6(1) aims to identify entities that are at risk of lacking substance and imposes a reporting requirement if all three criteria are met (irrespective of whether the undertaking is ultimately determined to be a shell entity).

The crucial test (under Article 6(1)(c)) is whether the entity has outsourced the “administration of day-to-day operations and the decision-making on significant functions”. As it is currently drafted the test requires both the administration and decision-making to be outsourced, so where decision- making is retained and performed by directors in-house this test might not be the hazard it first appears to be. This anomaly has not been picked up by amendments. If this gateway remains unchanged a disproportionate focus will be placed on the meaning of "decision-making" and significant functions. 

There is however an amendment proposed to clarify that outsourcing should be focused on third parties. This was a point made in multiple responses to the consultation therefore we expect the treatment of outsourcing to evolve in the compromise text. 

The amendments proposed in relation to Article 6 do not offer any clarity on what “outsourcing” means. In one of the earlier rounds of amendments (in relation to a different point about reporting) outsourced activity is defined as “accounting, sales, marketing, logistics, consulting etc”. This would appear to be a very broad range of services, much wider than the use of third-party professional directors or domiciliation and entity management services that might be typically associated with entities having minimal levels of substance.

Undertakings that are caught by the gateway test will need to report in their annual tax return whether they meet certain indicators of minimum substance. Meeting all three of the indicators means the undertaking is presumed to have the requisite level of substance. However, if not, the undertaking is considered a “shell” and will face the tax consequences unless it can rebut the presumption or avail itself of one of the final exemptions offered. 

In practice the bank account and qualified people tests should be straightforward to satisfy. However, the lack of clarity around the premises test has caused some uncertainty around its potential reach, for example, would the use of rented office space meet the definition of “own” premises? Furthermore, where the premises are used by a number of group entities there is a risk that the “exclusive use” test is not met by the entity. The amendments do not provide much clarity at this stage. In our view, if the property is rented, by say a Master LuxCo and other group entities use the property under a licence, the licensees i.e., other group entities using the property could be said to have the non-exclusive possession of the property, leaving the Master LuxCo with “exclusive use”. It is worth noting that where a company does own its own premises, other group companies can use the property as the “exclusive use” condition appears to be drafted as an alternative test. Building on this, there is an amendment that seeks to include “premises shared with entities of the same group”. 

Where the minimum substance requirements are not met, there are two further options to consider. The first is the rebuttal of presumption which involves providing evidence around the commercial rationale and operations to prove that the undertaking has control and operations to prove that the undertaking has control over and borne the risks of the business activities. Although not ideal, this may in practice be possible to satisfy due to the number of commercial non-tax reasons why a holding company might be used in a fund structure. 

The second option is a request for an exemption where it can be proved that the undertaking does not reduce the tax can be proved that the undertaking does not reduce the tax might be more difficult to satisfy as it requires comparing the amount of tax due with and without the interposition of the undertaking. While the main objective of the holding company is to ensure a fund structure does not create an additional layer of tax, there are circumstances where it can also create a more favourable withholding tax outcome for investors than if they had invested directly. 

The directive targets EU resident undertakings so a non-EU holding company could offer a viable alternative. For certain managers the introduction of the UK Qualified Asset Holding Company regime provides a well-timed alternative, however, the EC has made it clear that it intends to introduce further regulations to target non-EU entities.

Work is currently underway to design these mechanisms, but it will clearly be constrained. The EC will not be able to override bilateral tax treaties so it will be forced to adopt a soft-law approach. The first measure the Commission has put forward is a regulatory framework for tax intermediaries that are involved in the creation of arrangements outside of the EU that facilitate tax evasion or aggressive tax planning in the EU the so-called SAFE (Securing the Activity Framework of Enablers) initiative.

The EU list of non-cooperative jurisdictions for tax purposes is also under a separate review. These could be used as a mechanism to encourage third countries to adopt similar standards to ATAD 3. Although the UK is likely to resist such pressure, standards may eventually converge towards those set by ATAD 3.