In a seminal case regarding trust matters, the Court of Justice of the European Free Trade Association (‘EFTA Court’) has ruled in Olsen (7 July 2014) that a trust, as a form of establishment, may fall within the scope of articles 31 and 40 of the Agreement on the European Economic Area (‘EEA Agreement’). These articles of the EEA Agreement form the legal basis of the so-called freedom of establishment and the free movement of capital within the EEA.
As the case was decided on the basis of these fundamental freedoms, the Olsen-case is also of particular significance for European Union (‘EU’) law. After all, both the Court of Justice of the EU (‘CJEU’) and the EFTA Court have held that substantially identical provisions of the EEA Agreement and the Treaty on the Functioning of the European Union (‘TFEU’), which have the same legal scope, should be given a uniform interpretation. (See e.g. CJEU, 1 April 2004, C-286/02, Bellio Fratelli, para. 34.) Currently, a similar case is pending before the CJEU (Case C-646/15).
The Olsen-judgment primarily concerned Norwegian tax legislation. However, it may have important implications for other branches of the law as well. The Olsen-judgment raises first and foremost one fundamental question: can trusts be considered to be ‘entities’ on the same footing as ‘legal persons’?
‘The trust’ does not exist
A complicating factor in any form of research concerning trusts, is that one has to realize that the prototypical trust does not exist. For example, Black’s Law Dictionary mentions more than 160 types and categories of trusts, most of which can be found in English law alone. Also, today, ‘trusts’ or ‘fiducies’ can be found in many (civil law-) jurisdictions, such as France, Luxembourg, Liechtenstein, the Czech Republic, Romania, China and Curacao. To top it all off, different trust forms can be found in different so-called ‘mixed jurisdictions’, such as Scotland, South Africa and Québec.
This blog post will focus only on traditional common law trusts, and especially on private, express trusts. Charitable trusts, purpose trusts and trusts which are settled involuntarily are therefore excluded from this discussion.
What is an ‘entity’?
To answer the question whether the trust should be considered to be an ‘entity’, we should first understand what it means to be an ‘entity’.
Black’s Law Dictionary defines the term ‘entity’ as: “[a]n organization (such as a business or governmental unit) that has a legal identity apart from its members.” (B.A. GARNER (ed.), Black’s Law Dictionary, Eighth Edition, St. Paul, Thompson Business, 2004, 573).
Starting out from the observation that in economic theory, ‘firms’ are generally considered to be the central counterparty in a complex set of legal relationships, and therefore a nexus of contracts (see: M.C. JENSEN, W.H. MECKLING, “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure”, Journal of Financial Economics 1976, 310 – 11), Hansmann, Kraakman and Mattei argue that, in itself, this explanation does not seem to suffice. Instead, these authors contend that, to effectively serve as a ‘nexus of contracts’, any entity should display two essential characteristics:
The first is a well-defined decision-making authority. In a trust, this decision-making authority rests primarily with the trustee. However, in many jurisdictions, it is possible for the settlor to reserve some powers for himself, or grant certain decision-making powers to a protector. In this way, it becomes possible to police the trustee’s behavior.
The second is that the entity needs to have a dedicated pool of assets which are available to satisfy the claims of entity’s creditors. In other words, it is essential for any entity to possess a dedicated pool of assets to which the entity’s creditors have a priority claim. Inversely, this means that the personal creditors of the entity’s owners, managers, shareholders or beneficiaries are not on the same footing as the entity’s creditors, when it comes to satisfying their claims by taking entity assets. The authors call this phenomenon ‘affirmative asset partitioning’ or ‘entity shielding’ and contend that this characteristic is essential to any ‘legal entity’. Entity shielding may be contrasted with the non-essential characteristic of ‘limited liability’ or ‘defensive asset partitioning’, a characteristic typically associated with entities that enjoy legal personality (see: H. HANSMANN, R. KRAAKMAN, “The Essential Role of Organizational Law”, Yale Law Journal 2000, 392).
The trust is an ‘entity’ …
Interestingly, Hansmann, KRAAKMAN and mattei note that trusts also display the essential characteristic of ‘entity shielding’ because the trustee’s personal creditors cannot satisfy their claims with the trust assets. In other words, the integrity of the trust fund is assured, even in the case where the trustee becomes insolvent.
In addition, trustees have legal title to the trust assets, which gives them the primary decision making authority regarding the segregated pool of trust assets.
In conclusion, the trust indeed appears to be, at least prima facie, a legal entity as it displays the two essential characteristics (H. Hansmann, U. Mattei, “The Functions of Trust Law: A Comparative Legal and Economic Analysis”, N.Y.U. Law Review 1998, 454 et seq.) Moreover, in practice, trusts are often spoken of as if they were actual entities, blessed with an existence, separate from the individuals who fulfill the roles of settlor, beneficiary, trustee or protector.
… or not ?
If a trust can be considered to be a ‘legal entity’, the decision of the EFTA Court in Olsen seems logical. However, this decision becomes less straightforward when one takes a closer look at the many particularities of, for example, English trust law.
As noted above, a typical characteristic of an ‘entity’ is that a distinction can be made between (i) the personal creditors of the owners, managers, shareholders and other persons involved with the entity and (ii) the entity’s creditors. We saw that the trust assets are protected against the trustee’s insolvency. This looks like an ‘entity’: personal creditors of the ‘manager’ have no claim against the entity assets.
However, the classification of trusts as a type of ‘entity’ seems to break down upon closer inspection. While personal creditors cannot satisfy their claims against the trust assets, there are no ‘trust creditors’ with a (direct) priority claim over the trust assets. Take as an example the trustee who contracts with a third party as trustee, or the trustee who commits a tort while exercising his duties as trustee. One would expect these specific creditors to be ‘trust creditors’. Under English law, this is not the case. All these creditors are in fact personal creditors of the trustee. The trust itself does not have creditors who can attach directly to the trust assets. Correspondingly, in English law, only assets can be held on trust and never liabilities. The latter are attributed to the trustee personally. (See: A.J. Oakley, Parker and Mellows: “The modern law of trusts”, 8th Edition, Butterworths, London, 2003, 22 – 23; C.E. Rounds, C.E. Rounds, Loring and Rounds: A Trustee’s Handbook, New York, Kluwer, 2015, 800.)
When acting intra vires, the trustee does have the possibility to pay creditors with trust funds or to take any such funds to repay himself. Should the trustee fail to do so, these creditors have at their disposal a so-called ‘subrogation claim’, which allows them to step into the trustee’s shoes and attach to the trust fund in the place of the trustee. Crucially, this implies that any claim of the ‘trust creditors’ with regard to the trust fund can never be stronger than the one the trustee himself has.
If the trustee acted in breach of trust, and thus ultra vires, any possibility for the trust creditors to reach the trust fund directly, may – and probably will – be cut off. Moreover, when contracting with third parties, the trustee is allowed to stipulate that his personal goods are unavailable for trust creditors. This is however, a feature of contract law and not trust law. Although, taken together, these rules may operate to the detriment of ‘trust creditors’. On the other hand, an altruistic trustee may also contract with third parties without any contractual limitation of liability.
In the absence of genuine trust creditors however, one may doubt whether common law trusts may actually considered to be entities. At most, there exists a special class of creditors who may step into the shoes of the trustee in order to claim trust assets. But on a fundamental level, these creditors are still personal creditors of the trustee, as there are no liabilities which are to attributed to the trust itself.
Question moot after Brexit?
One final note: as long as the procedure for the United Kingdom to leave the European Union is not finalized, the UK remains part of the EU and so does the trust. However, even in the worst-case scenario where the UK would leave the internal market completely, this would not mean that the common law trust would disappear from the internal market. After all, common law trusts may also be found in Irish law.
This blogpost is based on: N. APPERMONT, “Adrift between Scylla and Charybdis? The trust caught between a civil law rock and a fiscal hard place”, Trusts & Trustees 2016, issue 10, 31p.