Designing A Way Forward For Trusts

At present, jurisdictions within the European Union (EU) are in the process of implementing two new pieces of legislation. Since 2018, Member States have been in discussion about how to transpose the Fifth Anti-Money Laundering Directive (5MLD). The legislation follows soon after the Fourth Anti-Money Laundering Directive and, amongst other aspects, mandates the use of  publicly accessible registers of beneficial ownership for trusts. At the same time, jurisdictions are also dealing with the latest round of amendments to the Directive on Administrative Cooperation (DAC6). These new measures require the reporting of so-called aggressive cross-border tax planning schemes. Complicating the implementation of DAC6 is the existence of the similar, but unrelated, Mandatory Disclosure Rules (MDRs). Published by the Organisation for Economic Co-operation and Development (OECD), the MDRs also target aggressive cross-border tax planning and outline a reporting regime.

Clearly, it is important that legislation aimed at tackling serious issues such as money laundering and tax evasion should be implemented, but it is equally important that it is proportionate. Unfortunately, it is clear that neither 5MLD nor DAC6 have been designed with trusts in mind.

To this end, STEP has engaged with both national governments and the EU to try to ensure the implementation of both 5MLD and DAC6 is balanced and properly recognises the particular positon of trusts.

As trusts are not widely used in many EU Member States, it is, perhaps, understandable that DAC6 doesn’t always properly account for them. But this also means those jurisdictions with large trust sectors need to ensure the rules are implemented in a way that doesn’t unduly burden practitioners in the industry. It has been STEP’s contention that where DAC6 is being implemented, it should be accompanied by clear and exhaustive guidance to assist practitioners in complying, particularly given the burden it places on intermediaries and the consequences of failing to make a report. There is still work to be done in this regard.

DAC6 includes a number of hallmarks which outline the specified characteristics a cross-border scheme must meet in order to trigger a mandatory report. The majority of Member States appear to be following closely the hallmarks contained in the legislation  (a notable outlier in this regard is Poland which has increased the number of hallmarks from 15 to 24, along with a penalty regime that appears to be higher than most other jurisdictions). When considering trusts, the most relevant hallmarks are Hallmark A (in the context of standardised documentation given its relevance to the creation of trusts) and Hallmark D.

One area that has largely been ignored in discussions around the implementation of DAC6 is that Hallmark D is very different to the other hallmarks in the legislation. The majority of the hallmarks attempt to tackle aggressive tax avoidance, an act which requires forethought, but Hallmark D will be triggered if an arrangement may have the effect of undermining automatic exchange information reporting obligations. This means it could apply to perfectly innocent transactions, where none of the participants has given thought to whether a proposed arrangement may trigger fewer reporting obligations than an alternative.  

Applying the legislation in a trust context brings with it a range of problems. The term ‘cross-border arrangement’ may be, in many ways, self-explanatory in a non-trust context but it becomes more complex when a trust is involved. Clearly if there is no cross-border component, an arrangement will not be reportable making it important to identify any grey areas prior to implementation. The Directive defines a ‘cross-border arrangement’ in the following way:

An arrangement concerning either more than one Member State or a Member State and a third country.[i]

At present, the term ‘concern’ is ambiguous and has been left undefined. In the UK, a key jurisdiction for the use of trusts in the EU, it has been suggested that for an arrangement to ‘concern’ a country, that country must be of ‘some material relevance’ to the arrangement. STEP’s view is that in relation to a trust, an arrangement should be deemed to ‘concern’ a Member State if any of the following categories are tax resident in the Member State in question at the time the arrangement is made available:

• A corporate trustee or an individual trustee.

• The settlor.

• The protector.

• Any named beneficiary or beneficiary with a fixed interest.

• Any natural persons exercising effective control over the trust.

The term ‘intermediary’, as found in the legislation, also brings with it ambiguity. It is possible that the current definition risks imposing additional obligations on practitioners not involved in setting up an arrangement. This is most noticeable in the case of arrangements that are designed but not subsequently implemented. It is possible for a trust that may trigger Hallmark D to have been prepared for a client, but ahead of it being formally put in place the reporting implications are noticed and it is halted. Despite not being implemented, it was still made available to clients meaning that, in theory, it would bring with it the burden of reporting.  More clarity on exactly when a report should take place in these instances would be welcome.

Ambiguity in relation to trusts is also present in 5MLD. This is despite the implementation, for the most part, being more advanced than in the case of DAC6. 

One core issue which is still to be resolved is what actually constitutes a trust.  When the Directive was agreed it was mandated that:

The Commission shall publish the consolidated list of such trusts and similar legal arrangements in the Official Journal of the European Union by 10 September 2019.[ii]

Although the list has now been published, it contains little specific information about trusts. This has the potential to create an unfavourable situation for the trust industry, with jurisdictions taking a cautious approach by making more varieties of trust than necessary reportable. It would be disproportionate to mandate the registration of several types of trust, such as trusts of life insurance policies, trusts holding pension or death benefits, and pilot trusts (trusts which only hold nominal values until they actually receive funds).

An important point that has been left open to the interpretation of Member States is the definition of ‘legitimate interest’. This term governs who will be permitted access to the various national registers of beneficial ownership. It has been STEP’s contention that the term should be interpreted narrowly. This would have the advantage of protecting the privacy of those individuals on the register, as well as potentially deterring a large number of spurious requests which could otherwise consume resources to respond to.

The term ‘business relationship’ may also have very real implications for the trust industries of EU Member States.  Under one interpretation, the requirement could lead to any non-EU trustee using a service from within a Member State being obliged to enter their details on the beneficial ownership register of that Member State. There is a risk that the additional compliance work, costs, and privacy implications could make EU jurisdictions unattractive and uncompetitive regions for this trust-related work. This would seem to be a needlessly damaging provision for Member States to implement.

STEP has argued that a trustee should be required to enter into a business relationship ‘in the name of the trust’ before any registration requirements are triggered. This would ensure registration of a trust takes place only when a trust can be deemed to be an ‘entity’ in the same way as a company. It would also deliver the objectives of 5MLD in a proportionate way.

Both directives leave much to be desired in terms of legislating effectively for trusts and it is incumbent upon Member States with trust industries to transpose them in such a way that practitioners are not negatively impacted.

Whilst practitioners are awaiting the final implementation of both 5MLD and DAC6 another issue is just around the corner.

The idea of a minimum rate of taxation has been discussed by a number of bodies. After some attempts to reach a consensus view, the EU abandoned its proposals for a minimum rate of tax in March 2019; however, the OECD has now announced plans to design an approach. Work on the issue is also being undertaken a national level. The French government has implemented a 3 per cent tax on sales by multinational technology companies with revenues of over €750m (of which at least €25m must have been made in France), whilst the UK has also announced plans for a Digital Services Tax. 

How these proposals will interact with one another, and to what extent they will impact on the use of trusts, remains to be seen; but it is important that the lessons of 5MLD and DAC6 are learned.

[i] Council Directive (EU) 2018/822 of 25 May 2018: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L0822&from=EN

[ii] Directive (EU) 2018/843 of the European Parliament and of the Council of 30 May 2018: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L0843&from=EN

Daniel Nesbitt
Daniel Nesbitt is a Policy Executive for STEP (Society of Trust and Estate Practitioners), London, UK. Prior to joining STEP he worked in various policy roles in and around Westminster for a number of years. Most recently he worked at the civil liberties group Big Brother Watch as Research Director; overseeing the preparation of reports, consultation responses and inquiry submissions as well as representing the organisation externally in meetings with key stakeholders and in the media.