Journal of International Trust and Cor****ate Planning
[2006] JTCP 179
1 December 2006
The Future of the Trust Part I
Donovan WM Waters1
QC, FRSC
Counsel Horne Coupar, Barristers and Solicitors 300-612 View Street
Victoria, BC Canada V8W 1J5
Introduction
In 2001, I was invited to speak to a STEP London conference on the
subject of 'The Future of the Trust from a Worldwide Perspective'.2 It
was my thesis that the common law trust has a span of possible
structures that comes close to agency at one pole and trust fund
personification at the other, but is a relation****p of trustee and
beneficiary that has drawn its significance from being centred midway
between agency and personification. Since the 1970s, I argued,
innovatory legislation in a number of offshore common law
jurisdictions has moved this trust away from its traditional central
position between these two poles to quasi-agency, where the settlor is
dominant, while onshore legislation has been introduced making some
types of trust, for example, in public investment, effectively
cor****ate. It was my opinion that, if the common law jurisdictions
abandon the classical characteristics of this property model of trust,
and ape agency or cor****ate form, the future of the trust is what any
particular jurisdiction wishes to make of it. Instead of an
international concept it will be seen across international borders
merely as a jumble of ideas. Its reception inter-jurisdictionally will
be a matter of conjecture. Lost will be the present classic and clear
alternative to the other property management vehicles, namely,
incor****ation and agency (including partner****p), that the common law
trust concept now provides.
On this occasion, when I am asked to project a future for the trust
over the next 50 years, I would like to return to the phenomenon of
the trust that arises from the intent of a person, the express trust,3
and reflect on two practical issues. The first concerns the 'models'
of trust that will be in use in the coming half century?4 Fifty-three
jurisdictions worldwide possess statute or a code expressly or
impliedly providing for a 'trust', and this list excludes the world's
main common law jurisdictions -- those of the USA, England and Wales,
Canada, Australia and New Zealand.5 The second issue concerns what has
been done with the trust concept. What usage is being made worldwide
of the trust, and where will use of the trust during the next half
century concentrate? Well under 10% of the total dollar value of US
trust funds is under management in private client (ie family wealth
management) trusts.6 Will we see a continuing marginalisation of the
dollars involved with trusts in wills and family wealth management?
The trust has already shown its paces as a financial investment tool;
it is utilised widely as an alternative to the cor****ation as an
engine of economic growth. Some speak of the trust 'fighting for turf'
with the cor****ation, and in that regard the thoughts of anyone who is
aware of economic growth trends and capital flows in this first decade
of the twenty-first century will be on Asia.
What is a 'trust'?
We all know what a contract is -- it is an agreement between parties
each of whom is legally bound as a consequence to perform such
obligations, for the other party or a third person, as the promisor
has undertaken. A tort or delict, terminological obfuscation aside, is
obviously an involuntary injury causing physical harm or economic loss
to the victim. Even the notion of 'property' is not too challenging.
It is a tangible or intangible asset (like a debt owed) over which, if
it is mine, I have rights and, as a result of it being mine, I have
obligations. I can sell it or gift it to another, I can maintain and
repair it, destroy or waste it, and I can enjoy it fully according to
the nature of the particular property. Nor is it difficult to imagine
a cor****ation, that is, a legal persona.7 A natural person has rights
and obligations, not only through the contracts he has made, the
injury that has been done by him or to him, or the property that he
owns, but because he is a member of human society. His rights are
vested in him, and his obligations are his. We all understand that. It
is therefore not a far stretch for us to conceive of the cor****ation;
we understand it as another legal person, or, as old time lawyers
called it, an artificial legal person. It was described as artificial
because it does not walk and talk; and therefore, needs human beings
to make decisions for it and otherwise act on its behalf. This process
provides the decisions and acts of the cor****ation. A cor****ation (or
'company' in the historic English usage) may or may not have
shareholders, but in any event, except in the liquidation of the
cor****ation once all creditor rights are met, no shareholder has a
claim to its assets. The foundation, the stiftung or stichting of
civil law jurisdictions is another form of artificial legal person. It
too is not subject to the claims of 'members', except where the
formation do***ents of the persona confer such rights. In effect --
and this is the point to be emphasised -- both the cor****ation and the
private foundation occupy an interposed position between the legal
person who conveys assets to the cor****ation or foundation as legal
persona, and those persons who look for conferred benefits flowing
from the persona.
A 'trust' on the other hand, by way of contrast with the interposed
legal persona, can be described simply as property administered by A
for the benefit of B. That language, however, describes agency, and a
trust is more than that. We would agree too that the trust is not a
legal person.8 So, if it is a concept with more features than one
party being in a special relation****p with another because he is
acting for the benefit of that other, can we say that an indispensable
feature -- beyond one acting for another -- is that the assets being
administered constitute a segregated fund? We know that only those
creditors whose claims arise out of the trustee administration of the
trust fund may sue the trustee for his discharge of the asserted
obligation, and that, if the trustee is a human person, the spouse and
children of the trustee, and the trustee's heirs, have no rights in
this segregated fund. Such an isolated fund is indeed a fundamental
element in the 'trust'.
Another element is the degree of final authority the trustee must have
in the decisions he makes with regard to the management or, if so
required, the disposition of the trust property. The agent acts at all
times in accordance with his principal's orders or counter-orders; the
trustee receives his instructions, powers and discretions from the
trust instrument, and thereafter makes such decisions, disciplined by
his fiduciary status, as he thinks appropriate. Once the terms of the
trust are in place, therefore, the distinction between agency and
trust is that the trustee has at least significant powers and
discretions whose exercise is not subject to the veto or approval of
another. This is usually described as the control element. Only ultra
vires acts on the part of the trustee, or improperly exercised powers
or discretions, entitle a settlor,9 'protector' or beneficiary to
intervene and seek remedy against the consequences of the exercise of
the power or discretion.10
An agent as a bare trustee may find himself managing a segregated
fund, and his management decisions may be more than ministerial in
character, but he acts nevertheless subject to the ultimate control
rights of his principal.
Two models of a trust
All 'trusts' conceived as such in jurisdictions around the world share
certain characteristics -- a segregated fund, trustee control over
management of the fund, remedies of some kind for the enforcement of
trustee duties, and the protection of the fund from misappropriation
and abuse of confidence. If segregation does not exist, whatever the
jurisdiction's concept may be called, it is not a trust. Trustee
control is regarded by all mainland jurisdictions as a basic element.
11
Different models of trust can be said to exist because between
jurisdictions there are different conclusions of what is the single
most fundamental building block of a trust. Is a trust the imposition
of an obligation upon a trustee who is to hold or administer
('manage') property, and empowered to discharge the obligation? Or is
asset holding together with management powers a property concept that
is comparable in the effects it produces to the cor****ation or the
civil law foundation? This difference of approach creates different
doctrinal perceptions about the essentials of a trust and how it is
constructed. Whether the remedies of trustee and beneficiary can be in
personam only, or must include the in rem, is controversial between
proponents of one set or another of doctrinal perceptions. 'Models' is
a word used in this paper to reflect different ideas about the basic
building blocks of a trust.12
With a common law trust the trustee receives a transfer of the title
to each asset of a segregated fund of assets. Segregation of the trust
property is the first crucial element. The trustee manages that fund
for beneficiaries (or for the furtherance of a purpose) and accounts
as a trusted individual (ie a fiduciary) to those beneficiaries, or to
the 'enforcer' of the purpose. It is a relation****p between a trustee
and a beneficiary (or a trustee and the enforcer if the trust object
is a purpose). There are obligations to which the 'fiduciary' is
expected to adhere, notably to avoid being in a position of a conflict
of interest and duty, and the trust is policed by exacting standards
of conduct and practice. These are the so-called fiduciary duties.
That relation****p in the writer's view is at the heart of the common
law trust concept. The trustee has the right of disposition and
management, the beneficiary the right of enjoyment. Both manager and
beneficiary in the common law model of a trust have property interests
in the trust assets. The common law splits the rights of so-called
owner****p between trustee and beneficiary.13 The common law trust is
known as the property model, because both trustee and beneficiary have
proprietary rights. These rights entitle each to recover trust
property wrongfully withheld from the trust fund. The trust is
therefore a bringing together of obligations owed by the trustee to
the beneficiary, and property rights vested in both trustee and
beneficiary whereby each can recover specific trust assets wrongfully
withheld from the segregated trust fund. The in rem remedy is a fund
protection device.
The common law has a singular approach to the concept of property. It
interposes the notion of 'interest' or, to use the older term,
'estate', between the entitled person and the specific asset itself.
The person owns his or her interest in the property, not the property
itself.14 The common law speaks of the rights that owner****p involves.
In equity the trustee has the rights of title holding, disposition and
management, the beneficiary has the right of enjoyment. Remedies to
rectify harm to the trust fund are extended to both trustee and
beneficiary. The trustee may bring a personal action for
restitutionary compensation against any non-performing contractual
party, tortfeasor or fraudster. The beneficiary may bring a personal
action against the trustee for breach of trust, and, if successful,
recover out of the trustee's own pocket the making good of the
depreciated or damaged trust fund. However, the trustee, holding legal
title, may bring a property action for the recovery of specific trust
assets, and the beneficiary has also a property right, entitling him
to recover from the trustee the trustee's illicit withholding of trust
property. Moreover, if the third party transferee knew or ought to
have known of the trustee's wrongdoing, the beneficiary can also claim
trust property in specie from that third party wrongdoer.15 Property
rights include the right to trace identifiable trust property in the
hands of the wrongdoing trustee, and beyond the trustee into the hands
of the wrongdoing third party. The property right has the considerable
advantage that it prevails against the creditors of the wrongdoer in
the event of the latter's insolvency or bankruptcy; the right is an
asset the beneficiary owns, and can assign. This model makes the trust
a self-contained management concept independent of its mode of
creation. Once created it operates under its own substantive law
rules.
The common law, including the property model of the trust, went from
its origin in England to the 13 American colonies, to what later
became Canada, Australia and New Zealand, to a number of African
territories, to India, Sri Lanka and Malaya, and to numerous outposts
of the British Empire, as it then was. British overseas ****ts included
Singa****e, Hong Kong, Bermuda and many of the islands of the Caribbean
and the South Pacific.16
Though the property or common law trust is the trust model that as the
'traditional trust', to borrow Lupoi's term, is best known, a civil
law model has been developed.
The trust in Scotland is said by scholars of that ancient civil law
jurisdiction to have been in existence since the seventeenth century,
which is certainly before English legal influences were experienced
there. But in places other than Scotland it was brought from England
by settlers in overseas Crown-acquired lands. These included existing
civil law jurisdictions in various parts of the world. There the
common law trust idea, moulded into a compatible form, thus became
part in each case of the local legal fabric. These jurisdictions, part
civil law, part common law, are now the so-called 'mixed'
jurisdictions. The French civil law scholar, Pierre Lepaulle, in the
first half of the twentieth century was fascinated by the challenge of
reconciling the common law trust idea with Roman civil law concepts,17
and his ideas have been persuasive in Quebec and at least two other
jurisdictions of the 'mixed' civil law and common law tradition.
However, in the present writer's view it is fair to say that until the
1930s the Anglo-Saxon trust, or rather the trust of the USA and of the
British Empire (later in the days of independence 'the Commonwealth'),
was the trust. Frederick Maitland, the late nineteenth and early
twentieth century English scholar, took great pleasure in his
Cambridge University lectures in speaking of the trust that
generations of English judges and practitioners had worked on, each
generation expanding upon and imaginatively applying a concept and
structure received from its predecessors. It is essentially the
post-1945 age of the European community of nations, plus the
globalisation of commerce and peoples, that have brought us to speak
of models of the trust, thereby recognising the number of non-common
law jurisdictions today that possess a 'trust' of some form. Indeed,
in order to give the trust historical perspective one has to be aware
that the conception of trust 'models' is a product of learning which
evolved only in the last 20 years of the twentieth century.18
The civil law has a natural reluctance to recognise the property or
common law trust in conflict of law terms, let alone adopt it into
local domestic law. That trust model may regard both trustee and
beneficiary as having property interests, but the civil law cannot
take that position. The civil law whose Roman roots are long pre-
feudal has no doctrine of an estate or interest interposed between the
person and the thing owned by the person; and therefore, there can be
no division of the rights of owner****p between two or more persons.
Unlike the common law, the civil law relates persons directly to
things (immovable and movable). Property as a legal expression
generates rights, the rights in things (disposition, maintenance, use
or abuse) constitute owner****p, and owner****p cannot be divided
between persons. Only the Roman concept of co-owner****p can do that.
With its roots in received Roman law there is a clear conceptual
divide in the civil law between the law of property and the law of
obligations, and this distinction is especially fundamental in all
Latin civil law jurisdictions. Moreover, and probably of even more
im****tance than 'owner****p' in the trusts law context, the numerus
clausus doctrine of the civil law is to the effect that there can be
only a limited number of real (or in rem) rights over or in property
owned by another, and no others. This adds to the civilian's
difficulties in contemplating any domestic adoption of the Anglo-Saxon
or property trust.
The countries of mainland Western Europe, the cradle of the civil law,
have traditionally been the most hesitant about, if not hostile to,
the trust.19
How then have the 'pure' civil law jurisdictions handled the situation
if they have introduced, or are interested in introducing, a domestic
'trust'? How have the 'mixed' jurisdictions devised a
conceptualisation for a trust, a working concept elsewhere that simply
arrived in each of these jurisdictions with the Anglo-Saxons? Common
law trusts and civil law trusts are now worldwide.20
Personification of a fund21 and contractual agency of the principal's
assets undertaken by the manager are well-established property
management concepts, but agency is not a trust, and nor in the present
writer's view is a personified structure, though the 'statutory
business trust' in the USA may be organised along unitholding/
shareholding cor****ate lines as 'a juridical entity'. The civil law
has developed an obligation model of trust.22 This model conceives of
the trust as an obligation to manage property, undertaken by one who
will discharge that task. The obligation may be created by agreement
between a fund provider and trustee, or as provided by statutory or
code authorisation. Civil laws today are almost invariably codified,23
and whether it is created by private act or authorised by statute or
code, the resultant 'trust' must in principle be compatible with the
basic doctrines and other provisions of the civil code in question.
The two models for a trust are therefore the property model and the
obligation model. There is a third structure which is reminiscent of
the US 'juridical entity', though it predates the US phenomenon. It is
Pierre Lepaulle's theory which poses as the trust's central
characteristic a fund of assets that is dedicated to expenditure upon
a particular object (person or purpose), but which as property is
owned by no one. It is classified by its proponents as an entity but
not a legal persona. It might be seen as a third model of trust, but
it is in fact a variant on the obligation model.24 In the property
model, as we have seen, both the trustee and the beneficiary have
proprietary interests, and each can vindicate that interest by actions
in rem. In the obligation model the trust is an obligation between the
creator of the trust and the trustee. The remedies for interference
with that obligation, or for its non-performance, are therefore in
principle personal. The beneficiary being not a party to the
obligation may or may not have remedy, though the usual statute or
code provision is to confer rights of some kind upon the beneficiary.
However, there is a range between two approaches in civil law
jurisdictions. At one pole the trust is thought of by analogy with the
existing civil law concept of contract. It is sometimes described as
an innominate contract. The Luxembourg 'fiduciary contract', and the
various French draft statutes of the last 15 years, fall into this
group. At the other pole are those jurisdictions that are concerned to
play down the obligation root. These jurisdictions fuse obligation and
distinct statutory provisions to create a stand-alone concept that is
independent of contract or trustee obligation to the fund provider.
The outcome is comparable with the concept of an incor****ated body
that is independent of the original promoters, and the partner****p
that is separate from the originally founding partners. Japan might be
considered the most obvious instance of jurisdictions in this group.
For these jurisdictions the challenge is the reconciliation of the
trust law with the tenets and rules of the local civil code.25
These jurisdictions like Japan in their civil code or statute
providing for the trust do not refer to contract. A latent basis of
obligation exists, but the stress is on the functional issues:
obligations and powers of the trustee, the appointment, retirement and
removal of trustees, who may enforce proper performance by the
trustee, and trust termination. In an inter vivos trust the creator of
the trust may be the principal enforcer, as in the People's Republic
of China, something which points to the contractual obligatory
character of the concept, but this connection with contract is not
spelled out. The effort of such a code or statute seems to be to
originate a concept that is functionally independent of any mode of
its creation.26 In each of these jurisdictions one has to look beyond
the opening sections of the statute or articles of the code, where
'trust' is described for the purposes of the code or statute, to the
totality of the provisions that are introduced. In any analysis of
models of trust these jurisdictions are of particular interest
because, while belonging to the civil law obligation tradition, they
are conscious of the characteristics of the property model that,
fa****oned by the courts over the centuries, has sought conceptual
independence from agency and personification.
The creator of the civil law inter vivos trust as the promisee will
usually enforce the trustee's duties and proper exercise of its
powers; the beneficiary (if a different person from the creator) may
or may not have enforcement rights of his own. In the dedicated fund
structure the entified fund stands between the creator of the trust
and the beneficiary; the trustee has a personal right to administer
the fund, and the beneficiary has a personal right to enjoy the
dedicated fund. In order to be recognised as a trust a segregated fund
is acknowledged as necessary for each model. In the property model
title to the fund is in the trustee, in the obligation model the fund
owner****p may be in the creator of the trust, the trustee, the
beneficiaries, or as we have seen in no one. The property model and
the obligation model both assume a relation****p. With the property
model it is the manager and the beneficiary. With the obligation model
it is the manager and the creator of the trust (who most frequently is
also the beneficiary), though statute or code may admit a third party
beneficiary to one or more remedies protecting the fund. An obligation
jurisdiction that seeks a conceptual independence for the trust is
drawn towards permitting any enforcement remedy given to the trust
creator to be extended also to the beneficiary. Even where the
jurisdiction does not refer to contract or agreement, there is never a
****fting of the concept to a clear trustee and beneficiary
relation****p. As between obligation jurisdictions there are merely in
effect degrees of movement towards that relation****p.
Fundamental elements of a trust, whatever the model
The concept of a trust that embraces both these models might be this:
a trust is a segregated fund managed or administered by one person, an
individual or a personified body of persons, for the benefit of the
creator of the trust, another person or the furtherance of a purpose,
where the manager is not an agent of, or recipient of a mandate from,
the trust creator or beneficiary, but has independence from others in
his or its decision-making.
There are three elements that are implicit in that definition. The
first is the segregation of the fund, and the second is the protective
process for that fund so long as the trust endures. Protection
involves the wrongdoer making good loss to the trust fund, and it also
involves specific recovery for the trust fund of wrongfully
appropriated assets. If protection is to be total, the trustee needs
to be able to acquire compensation from a wrongdoer, and to recover
trust property wrongfully in the hands of an insolvent or bankrupt
wrongdoer. The beneficiary (and/or the settlor) needs to have a remedy
in order that compensation can be had either from a wrongdoing trustee
or from a complicit third party wrongdoer. Again it enhances
protection of the trust fund that specific recovery can be obtained
from the insolvent or bankrupt trustee or third party. The third
element is the control over management of the fund assets that the
trustee has, and the agent does not have.
A word of explanation is advantageous as to what is meant by each of
these elements.
Segregation
The question is how far the assets comprising the trust fund are free
of claims brought against the asset manager in respect of matters that
are not connected with management of the trust assets. The claimant
against the manager on the basis of matrimonial rights, of forced
share rights, or of inheritance entitlement clearly cannot reach a
segregated fund. Creditors not connected with the management device
include those who are claiming with regard to an investment or
property management business that is not associated with the manager
as trustee, or the handling by the manager of another trust ****tfolio;
and this is so even if the creator of the trust is permitted to be,
and is also, the trustee. Assets, such as land or trade equipment,
that are simply in fact held in physical separation from the trustee's
personal assets are not 'segregated', as was stated earlier; nor are
assets that for commercial or accounting convenience are held in a
separate bank account or listing of securities. What is essential is
intention. In the act of creation of the alleged trust, or in the
later determination to transfer further property into the trust fund,
there must be evidence of an intention that the designated property be
segregated and maintained while under management in a segregated state.
27 It is clear that in jurisdictions that permit the settlor to retain
the owner****p of the trust assets, while the trustee has only rights
to deal with the assets, the evidence of an allegedly intended
'segregation' may well be insufficient.28 The same absence of evidence
will not exist where the jurisdiction in question allows transfer of
'owner****p' to the beneficiary or beneficiaries instead of to the
trustee.29
Protection of the fund from wrongdoers
The manager's remedies
No legal right exists if no remedy is available when the alleged right
is violated. The right to enforce the trustee's obligations may be in
the trust creator as the creator, or in the beneficiary as
beneficiary. A segregated fund needs protection, or assets may be lost
to the fund. The manager will surely have a personal action against
the wrongdoer, and in that action he will claim compensation for the
damage, destruction or loss that was caused by the act of the
wrongdoer, whether trust creator, beneficiary or third party stranger.
Has the manager the power to challenge the third party who claims that
in the cir***stances of the third party acquisition the party has
acquired a superior title?
Has the manager (the trustee) the power to reclaim specific trust
assets from a wrongdoer who has those assets and is a thief or
defrauder? If the wrongdoer is insolvent, but has the assets, can the
manager claim priority over the wrongdoer's creditors?
The beneficiary's remedies
The manager as an individual, or a cor****ation through its employees,
may itself be a wrongdoer. Does the beneficiary as trust enforcer have
the power to sue (1) requiring the manager to make good the trust fund
because of loss caused by the manager's acts or omissions? (2)
requiring the manager to surrender to the fund illicit gain made in
competition with the trust or in abusing his office of the trustee?
Does the beneficiary have the power not only to sue the manager
personally, but, if the manager is insolvent or bankrupt, to recover
misappropriated fund assets still in the manager's personal possession
ahead of the manager's unsecured creditors?
Then there is the wrongdoing third party. Does the beneficiary as
trust fund protector have the right (and power) to sue personally the
third party who was complicit in acting with the faithless manager in
harming the trust fund? Can the beneficiary recover in specie trust
assets still in the possession of the wrongdoing third party who is
insolvent or bankrupt?
An observation
Common law and civil law doctrine differ as to whether specific
restitution may be had by either the trustee from the wrongdoing
stranger, or the beneficiary from the wrongdoing trustee or complicit
third party.30 Civil law possesses a doctrine of real subrogation, but
it is not generalised to the extent of the tracing remedy which allows
the common law plaintiff to follow the trust property into the
wrongdoer's hands so long as the specific property is identifiable,
albeit in converted form. It is this remedy that allows the property
model claimant on behalf of the trust fund to take back the particular
trust asset, even if the asset is by that time converted, and to do so
in priority to the wrongdoer's (and even the donee's) creditors.
In the present writer's view the absence of in rem remedies exposes
the trust fund to an undesirable degree, but in rem remedies are not
an essential element of a trust. In policy terms with regard to
business and commercial trusts, excepting perhaps the pension trust,
the protection given by in rem remedies is in any event harder to
justify.
Independence of the manager
An agent must obey the principal's instructions, whenever originated
during the course of the agency. A manager (trustee) derives his
duties and powers from the law and the trust instrument (or oral
instructions). This marks his independence in decision-making.
However, the settlor may reserve certain rights and powers for
himself, or confer them upon the beneficiaries or a non-trustee third
party. At what point does this practice turn the manager into a
nominee, bound to act in the exercise of his powers as others think
fit? It is commonly said to be a matter of degree, but whatever that
phrase means there must surely be what a court could regard as a
realistic independence of the trustee to control trust affairs.
It is with these questions in mind, the writer would suggest, that we
can best look at the question of which various property management
arrangements around the world meet this internationally adjusted
concept of what is a trust. Where does the particular jurisdiction's
management by one for the benefit of another, whatever the degree of
'trust and confidence' placed in the manager, fade into agency? At the
other end of the trust spectrum when does a quasi-personification
structure (an entity, but not a persona) fade into a situation where
there is an indubitable legal person interposed between the funding
person and the person who is to enjoy the fund?31
Jurisdictional variations on the fiduciary administration theme
The so-called offshore jurisdictions whose trust is the property model
have departed in several respects from the 'traditional trust' of
England, common law Canada, Australia and New Zealand. Indeed, as we
have seen, Lupoi classifies the offshore trust as a different trust
model from the 'traditional trust'. In one offshore instance the
beneficiary may no longer be the enforcer of the trust, in three
instances trustee control can be reduced, if the settlor chooses,
almost to vani****ng point, and in yet another instance the trustee's
fiduciary responsibilities vis-=E0-vis certain trust assets have been
eliminated. Nor are the variations offshore only. The USA considers a
revocable inter vivos trust to be a trust, though the settlor as
beneficiary during his lifetime has complete control over the
investment of the assets and in that period of time the beneficiaries
whose interests vest in possession on his death have no right to an
accounting. Turning to the 'statutory business trust', this, described
as a juridical entity, must now be considered another creature when
compared with the wealth management trust of the individual. There are
'donatory trusts', and then there are commercial trusts.
The 'mixed' jurisdictions of common law and civil law have themselves
been working at their respective rationalisations of the trust, each
within the historic confines dictated by its past, and difference of
approach between them is inevitable. How each copes with the absolute
owner****p principle of civil law, other doctrines apart, is enough to
distinguish them. Each has gone in a different direction.
The pure civil law jurisdictions, whose entry into the domestic trust
area commenced in the 1930s among Central and South American states,
inevitably demonstrate different conclusions on how to meet the
absolute owner****p and numerus clausus problems. Some choose to remain
precisely at all points within the ambit of the local civil code,
others will rationalise an apparent departure from the code but in
one, or perhaps two, respects only, and yet others which seek a trust
concept independent in its structure and operation, like any other
mainstream code concept, have somehow to rationalise symmetry with the
code or just accept and brazen it out that the trust devised is 'a
drop of oil floating on a pool of water'.32
All civil law jurisdictions recognise the necessity that the fund
assets, whether original or added later, constitute a segregated mass,
whether the assets in question are to be retained for the
beneficiary's enjoyment or are to be invested and reinvested as a
fund. Given the civil law principle that all assets owned by the
individual (ie within his patrimony) must be available to his
creditors, the problem for the civilian may be that of finding
justification in the local civil code for the formation of a mass of
assets that is segregated from some creditors' claims. Such
segregation as is sanctioned by the code must ensure that the debtor-
protection principle is not violated. It is also interesting that,
unlike several property model jurisdictions, no civil law jurisdiction
appears to challenge the fundamental nature of the principle that the
trustee exercises the powers of disposition and management without the
say-so or interference of others.
Almost all jurisdictions that adopt the obligation model of the trust
accept the notion of fiduciary owner****p. That is, the trustee owns
the segregated mass of trust assets but in a fiduciary capacity. In
the last few years the civil law language of patrimony has been used
to explain this rationale. The trustee has his personal patrimony, and
also a segregated fiduciary patrimony. Proponents of this opinion
evidently consider that a trustee of the fiduciary patrimony can have
title to the rights of disposition and management, but another person
have the right of enjoyment. For some scholars of this opinion it does
not matter for trust purposes whether 'owner****p' (as opposed to
title, which is only one right comprised in the 'owner****p' bundle of
rights) is retained by the trust creator, or is granted to the trustee
or the beneficiary. Where the trustee is not to be the fiduciary
owner, conferred by the fund provider upon the trustee will be the
rights of disposition and management; 'owner****p' is retained by the
trust creator or transferred to the beneficiary. One would have
thought that so long as the trust is in being this conferment upon the
trustee solely of the rights of disposition and management must be
irrevocable by the trust creator or the beneficiary, but while one or
two jurisdictions require this others are silent on the subject.
There is a significant body of opinion on the other hand that
considers 'fiduciary owner****p' to be a contradictory term. Owner****p
involves all the rights in any property; even within the so-called
fiduciary owner****p patrimony the trustee will pur****tedly have rights
of disposition and management and the beneficiary the right of
enjoyment. This school of thought therefore follows the thesis of
Pierre Lepaulle and conceives of the segregated property as a fund
owned by no one and dedicated (or appropriated) to an object of
expenditure. Critics of this no-owner analysis say that it is
essentially personification of the fund, and because it is an entity
short of legal persona how and in what manner the entity protects its
interests from wrongdoers, and should be understood for such
requirements as property registration, has to be solved by each
jurisdiction that adopts this analysis. Whatever the criticisms,
however, it is the only approach that totally honours the
indivisibility of owner****p principle; fiduciary owner****p on the
other hand can claim to be more easily understood. The debate will
continue.
Is the trust an innominate contract or a stand-alone property concept?
Within the civil law world where the trust has never been other than a
foreign concept there are sup****ters of each approach. For those who
are anxious to fit the trust wholly within code principles and rules,
contract is the only conceivable approximation. Those who are more
concerned to draft a structure that is comprehensive of all practical,
working aspects of a property management vehicle will aim to create an
independent, self-contained concept. There are other differences
between jurisdictions that do not concern the core elements of a
trust. For instance, some jurisdictions limit the types of property
that may be held in trust. Others do not, permitting such assets as
intellectual property rights and derivatives. Many jurisdictions
expressly restrict the trust model to investment and financial usage;
others, much the minority, generalise permitted usage. A jurisdiction
that has both common law and civil law roots will have been
historically propelled into a generalised usage. The task now is to
rationalise.
Remedies given to the trust creator, the trustee and the beneficiary,
vary considerably from jurisdiction to jurisdiction. Personal actions
are likely to be given to each -- the trust creator and also the
beneficiary vis-=E0-vis the trustee and/or the third party, the trustee
vis-=E0-vis the third party. But proprietary actions are unlikely to be
made available to any of the three would-be plaintiffs, except that in
several jurisdictions the court is empowered to set aside a contract
between the trustee and a third party if there has been a breach of
trust. A 'setting aside' of any transaction involves the restitution
to the plaintiff of the specific property that was the subject-matter
of the transaction, if of course that property is still in the
defendant's hands. In civil law jurisdictions, as previously noted,
real subrogation has not led to the development of a generalised right
to trace as exists in common law jurisdictions.33
The essential difficulty the common law trust presents for the
civilian lawyer is that the trust as a property management device can
be applied in every area of private and commercial law. A system of
law other than the common law may break up into separate categories
areas of law that would be treated as one by the common law trust. As
a Hague Secretariat publication for the benefit of delegates said in
1982 on the eve of the first Trusts Convention sessions, the civil law
has a 'bouquet' of institutions with which to meet the variety of uses
to which the common law puts the generalised common law trust.34 Some
of the civil law's categories may allow segregation; other categories
do not, or do so but to a lesser extent.
All these differences of approach, of enacted law and of categories,
can be found running through the following account.
Trusts of the Civil Law in the Western world
One of the most interesting instances of a fiduciary property
administration vehicle that appears to have the 'trust'
characteristics discussed above, but which the jurisdiction in
question does not regard as being a 'trust', is the fiduziarische
Treuhand of Germany. The eminent German scholar, Hein K=F6tz, has
described this concept for his English speaking readers as a transfer
of assets to a fiduciary (the Treuh=E4nder), and a contract between the
trust creator and the fiduciary that the latter will manage the assets
for the benefit of beneficiaries in accordance with the creator's
instructions. The fiduciary acquires an absolute owner****p, ie
unrestricted rights in the assets, and the trust creator and
beneficiary have therefore only in personam rights to enforce the
proper conduct of Treuhand administration. However, German courts have
interpreted the Civil Code as excluding the personal creditors of the
fiduciary from claiming against Treuhand assets, and as establi****ng
the beneficiary's right to secure the release of Treuhand assets from
the estate of the bankrupt fiduciary. All the same, these beneficiary
advantages attach only when it is the creator of the Treuhand who
funds the Treuhand; transfers to the fund by third parties are
categorised differently. Even creator-transferred assets remain
excluded from creditor claims only so long as they are in their
original form; once there has been sale and re-investment, creditor-
protection is gone. Professor K=F6tz observes that it would be contrary
to German policy to extend 'real subrogation' (a German Civil Code
doctrine) from the administration of deceaseds' estates -- where the
'fund' concept is established -- to the Treuhand. Germans would not
give beneficiaries of fiduciaries in general such an 'imbalanced'
proprietary interest over the insolvent fiduciary's creditors.35
The 'pure' civil law jurisdictions
Central and South America
We begin with those jurisdictions that were never recipients of the
common law, the so-called 'pure' civil law jurisdictions.
Prior to the Second World War the only pure civil law jurisdictions
that possessed a concept domestically that resembles a common law
trust were a number of countries of Central and South America. Mexico
in Central America is probably the best known. In this instance it
seems to be accepted that this was a response in the 1930s to needs
resulting from trade and commerce with the USA. But earlier forms of a
fideicomiso were in existence in other South American republics from
the nineteenth century. The influence of trade and financial dealings
with the USA was existent already in that century. However, a flurry
of interest in the 1930s faded thereafter, and the fideicomisos then
existent, some embryonic, others more copious, were simply left in
being. There are often severe restrictions on the cir***stances in
which the particular fideicomiso may be employed, and in the majority
of Latin jurisdictions to this day financial institutions alone may be
trustees. This conferment upon licensed financial cor****ations of the
exclusive right to assume the office of trustee was and is no doubt
seen as a protection to beneficiaries from fiduciary breach, but it
also reflects an interest in the trust essentially as a commercial
vehicle for investment and the holding of security for loans.
Latin civil law, originally introduced from Spain, does not generalise
the segregation or 'fund' idea, and throughout the Latin Americas
where it does occur segregation, as opposed to the barring of some
claimants, is not frequently found. The emphasis seems rather to be
upon the high standards of conduct expected of the fiduciary.
Segregation or 'ring-fencing', as Professor Hayton has described it,36
can present problems in a civil law jurisdiction because 'ring-
fencing' is generalised by the common law trust. Segregation applies
whatever the cir***stances in which the common law trust is invoked,
whether private client or commercial purpose.37 Nevertheless, all
Latin American states clearly recognise in their fideicomiso codes the
necessity of a particularised fund and the principle of its protection
from the generality of claimants.
The ubiquitous issue of who owns, or should be owner of, the trust
property has, of course, been a constant source of debate in the
drafting of all Latin American fideicomisos. The then contem****ary
Lepaulle theory of an unowned segregated mass of trust assets38 had
particular appeal in Mexico in 1932, though it was later abandoned,
and it continues to attract the interest of legislators in this Latin
part of the world.
Since the early 1970s six jurisdictions have amended their codes or
otherwise introduced new provisions,39 though each of these Latin
jurisdictions continues to have its own fideicomiso structure and
elements. The fideicomiso is not stated to be a contract; it is
usually provided that it may be created by contract or will, and its
character is described as one party administering property for the
benefit of another.40 The promisor is the fiduciary (fiduciario) and
agrees in an inter vivos arrangement with the fideicomiso creator
(fideicomitente) as promisee that the former will act for the benefit
of the object(s) of the fideicomiso. If the vehicle of creation is a
will, the testator creates the fideicomiso unilaterally by force of
the code provision. The objects may be persons or purposes, or both.
The majority of South American jurisdictions continue to permit
financial inter vivos fideicomisos only; in these instances A funds
for the benefit of A (or another funds for A's benefit), and T, a
financial institution, acts as the fiduciary. This is also the most
frequently employed fideicomiso. The fideicomiso has made entry into
individual and family wealth distribution in very few states; the
civil code is too distinct and comprehensive to permit this entry to
be legislated. The Uruguay Ley de Fideicomiso of November 4, 2003,41
is the latest of this legislation. It has been described as being the
most in-depth legislation enacted in any part of Latin America.
In terms of the analysis above, segregation, protection of assets and
trustee control, the segregation in the Uruguay Law is subject to the
prohibition of certain assets from being fideicomiso property.
However, neither the fiduciary's personal creditors nor the
fiduciary's spouse may make claim on the fideicomiso property (Arts 6
and 7),42 and, save for the actio Pauliana to prevent creditors of the
fideicomiso creator from being defeated whose claims arose before the
creation of the fideicomiso, the creator's creditors are also barred.
The fideicomiso which is created by will is subject to the claims both
of the deceased testator's creditors, and of his or her family for
forced shares (Arts 7 and 10).43
Protection of fund is a matter of more complex provision, and there
are some unexpected silences. The settlor or the beneficiary may sue
the fiduciary 'for any loss' the wrongdoing fiduciary has caused (Art
16). But there is no provision in the Law as to whether the settlor
and beneficiary are confined to in personam action only against the
wrongdoing fiduciary, even if the fiduciary retains misappropriated
trust assets. So far as third party wrongdoers are concerned, it is
provided (Art 15) that 'the trustee is obliged to bring all actions
relating to the defence of the assets of the fideicomiso against third
parties ...'. Presumably this covers actions against those with whom the
fiduciary has contracted while carrying out fideicomiso duties, as
well as tortfeasors. And the court may consent to the settlor or
beneficiary suing third parties when the fiduciary, in breach of his
duties, refuses to bring action (Art 15). If the fideicomiso is
registered, the fiduciary who acts with fideicomiso property
improperly or ultra vires does not bind the settlor or beneficiary
with any contract. With the court's consent either can have such
contracts set aside (Art 17).44 Does 'setting aside' enable the
settlor beneficiary to recover specific trust funds still in the
wrongdoer's hands? Is this what 'binding the settlor or
beneficiary' (Art 17) means? Not necessarily, if compensation is
regarded by the jurisdiction as adequate recompense. It would appear
that neither the beneficiary nor the settlor has any direct action
without court consent against the third party wrongdoer who has caused
loss to the trust fund, for example, theft or damage, or who, being
solvent or insolvent, withholds possession of trust property. However,
in practice this lack of action will be relieved by the fact that, if
the fiduciary 'in breach of his obligations' refuses to bring action,
the settlor or beneficiary with court consent can sue in personam in
the fiduciary's place. A fiduciary, who acted in some degree of
complicity with the third party when trust assets were improperly
transferred to the third party, will surely refuse to bring action
against that party.
An interesting feature of this Uruguay legislation is that the trust
fund is neither transferred to the fiduciary (or the beneficiary), nor
retained by the fideicomiso creator. It is legislated to be a
patrimonio de afectaci=F3n. It is a fund that is 'separate and
independent from the trustee's own property, and from that of the
creator and the beneficiary' (Art 6). However, it appears that the
fiduciary has sole administration and disposition powers with regard
to this fund.45 More of this concept will be said later in discussing
the Quebec fiducie of the 1994 codification of the province's civil
law.46
Europe
In mainland Europe no country, except possibly Luxembourg, or any
jurisdiction within a European country, has a domestic trust concept
that satisfies Art 2 of the Hague Convention on the Law applicable to
Trusts and their Recognition of 1986 (the Trusts Convention),47 though
both The Netherlands and Italy have ratified the Trusts Convention and
therefore, if the terms of the Trusts Convention on any particular
occasion are met, their courts are required to recognise the foreign
trust.48 The Netherlands continue to examine how a domestic trust can
be fitted within the Dutch Civil Code.49 On the other hand, Italy has
made dramatic progress with a trust interno. However, though this is a
trust where the trust creator, trust property, beneficiaries, and
trust objects are usually all Italian, the trust itself is expressly
governed by the law of a foreign trust jurisdiction. Recognition of
the foreign law takes place, it is argued, by authority of the Trusts
Convention, and the validity of the trust is based upon Italy's
ratification of the Trusts Convention, an interpretation that in both
respects has been upheld by the Italian courts. Since the trust rules
of the selected governing law must be based upon either a property or
an obligation approach, and the governing laws selected to date appear
to favour the property trust model, it would seem the structure of the
trust interno has to be the same.
A measure that would introduce a domestic trust was put before the
Italian legislature in November 1999. This proposal would have made
the trust assets a distinct fund, fully segregated from unconnected
creditors' claims (Art 2.1). No powers are expressly conferred upon
the trustee enabling the trustee to sue others who cause loss or
damage to, or appropriate, the trust fund, but beneficiaries 'may be
substituted for the trustee for the purpose of exercising rights
belonging to the trustee against third parties' (Art 3.2.1). It is not
clear whether the trustee is to have a 'fiduciary owner****p' of the
trust assets, though this seems to be implied from reference to the
trustee's own estate distinct from that of the trust. However, the
proposal was put on hold in January 2000, in response to the Italian
government's stated intention to come forward with proposals of its
own.50
A civil law jurisdiction that rejects the common law trust model, as a
property-based concept incompatible with fundamental civil law
principles, has two choices. Either the trust is an obligation to
administer segregated property for the benefit of one of the parties
or a third party, or the trust concept is abandoned, the segregated
fund is personified, and so a legal person is created. The latter
choice is easily understood; everyone appreciates the legal nature of
a human being. Having legal personality the legal person can, like a
natural person, own its own assets, and these are the assets to be
administered. In the civil law, incor****ation aside, the charitable
organisation (the fondation or stiftung) can be traced back to late
mediaeval times. It is a legal person. The contem****ary private
fondation or stiftung, administering and distributing family wealth,
is a descendant of the charitable organisation, and is readily
understood by notary and advocate.
Can the obligation be described as a contract (or an enforceable
agreement) between trust creator and trustee, and the contract itself
be understood as a trust? One might look at the common law or property
trust for a moment. Though this trust can be created by contract, the
trust relation****p of trustee and beneficiary, once created, is
independent of the instrument of creation. With the common law model
the contract, as trust instrument, will probably name successor
trustees or the person who is to appoint successor trustees. Most
frequently the appointors will be the trustees for time being,
whatever time that is. The accent here is already upon providing
managers for the segregated fund as a fund under administration,
regardless of whether the parties (trust creator and trustee) to the
trust instrument are alive or dead (or, if a cor****ate person, it is
liquidated, merged with another, or taken over). But, even if there is
no process set up on the creation of the trust, the relation****p and
the management will not grind to a halt. The common law system
historically confers upon the court the power to sup****t, as well as
police, the trust administration. One of the common law court's powers
-- and they are several -- is to appoint new trustee(s) where there is
no trustee and no one with the power, or who is willing, to appoint.
The relation****p of trustee and beneficiary, and the segregated fund,
are elements that can long outlive the parties to the trust-creating
contract. As the booster rocket falls away once the spacecraft is in
space, so is it that the initiating act of creation, for example, the
contract, becomes irrelevant. Its work is done.
This could not be so when the trust is the contract. Any legislation
introducing a domestic trust into a 'pure' civil law jurisdiction must
take account of the possibility that a trust may long outlive the
parties to the contract, or their capacity. Of necessity, therefore,
the trustees or the beneficiaries must be provided with the means of
securing a continuing management. Specific provision must be made in
the code or statute introducing the contract of trust, permitting the
required longevity of the trust, regardless of the length of any human
life. And in designing those added elements, the drafters must
remember that civil law courts have little or no familiarity with the
care and supervisory role the common law courts have historically
exercised vis-=E0-vis trusts. Civil law legislation can deal with the
problem, but it cannot be overlooked or dealt with in a generalised
line or two.
Of course, the contract model is unlikely to provide continuity of
management problems when the trust is not used as the vehicle for
successive beneficiary interests stretching over the span of two (or
more) human generations, but is to be relatively short-lived. The
financial or commercial trust is such a trust. Situations where the
trustee is an investment manager for the settlor's benefit, is holding
security for the settlor's loan to a third party, or is holding an
insurance policy on the settlor's life in favour of a designated
beneficiary are probably going to fall into the short-lived category.
This is where the obligatory (or contract) model of trust is as its
most comfortable. Nevertheless, if contract is largely limited by the
length of the participatory lives of the parties to that contract, the
unknown duration of that period means that provision is vital if the
trust is to continue whose terms are not yet carried out.
Luxembourg
Luxembourg's 'fiduciary contract' (contrat fiduciaire) -- as amended
in 2003, is aimed to meet the requirements of Art 2 of the Trusts
Convention as an obligation model of the trust. The name aptly
describes the property management scheme created. The contract,
necessarily in writing, is between a provider of a fund and a
fiduciary (fiduciaire) in which the fiduciary is appointed, and the
duties and powers of the fiduciary, and the object of the contract,
are set out. The object of the benefit will be a beneficiary, who is
either the initiating contractor or a third party. Evidently this
legislation is not intended to extend to the trust object that is a
purpose. The 2003 amendments have introduced the feature of a
segregated fund, which is a requirement of the Trusts Convention, Art
2.51 The fiduciary 'deviant propri=E9taire de biens formant un
patrimoine fiduciaire' (Art 5 of Title II of the 2003 Trust Law). The
fiduciary fund, as it might be translated, being rights both in and
over particular property, is distinct from the fiduciary's personal
patrimony. It is regarded by the 2003 legislation as 'owned' by the
fiduciary. His personal patrimony (or fund) the fiduciary owns in his
personal capacity. It is interesting for the purposes of this paper
that the 2003 Trust Law provides52 that the rules of the contract of
mandate apply as between the initiator (the trust creator/promisee)
and the fiduciary, other than those concerned with agency, and that
the initiator may renounce (peut renoncer) his right to give
instructions to the fiduciary.53 If during the subsistence of the
contract no renunciation is made, one assumes the promisee may give
amended or added instructions while the contractual promisor (the
fiduciary) is in the process of performing his contractual obligation.
One would have thought for the purposes of the Trusts Convention54
that mandate, agency and trust would have been fully distinguished,
because the distinction is conceptually crucial. However, it may be
that the initiator who does indeed retain his right to give continuing
instructions is here being paralleled with the common law settlor who
in Commonwealth mainland jurisdictions may reserve the permitted
modi*** of control powers. And it must be a modi***. For the settlor
to go too far invites the court to find the 'trust' is in law an
agency with the 'trustee' holding title on a bare trust only.55 The
obligation/contract model of the trust must lean over backwards to
avoid the categorisation of being an agency and contractual deposit of
title.
The Luxembourg 'fiduciary contract', however, is not an introduction
of a general contractual arrangement. Article 4 of the 2003 Trust Law
restricts its usage to products in the financial sector, and the Law
applies only where the trustee is a cor****ation. The cor****ation's
business must be lending, investment (pension scheme, mutual fund,
securitisation), insurance or the conduct of other provision of
financial services to the public. To the writer's mind the Luxembourg
contract is simply a contract between the initiator of a contract, A,
and the cor****ate manager, B, whereby with A's moneys B will provide
financial services to A. Funds are put in the hands of B as a term of
the services agreement. No reference is made in the 2003 Trust Law to
remedies that B has against third party wrongdoers who injure the
trust, and nothing is said of remedies that the customer, A, has
against wrongdoers if B, the manager, were involved in the wrongdoing.
The impression conveyed is that in the usual course of things the
continuing solvency of the Luxembourg financial organisation that
contracts in this way will be monitored by the Duchy; the
cor****ation's contractual obligations require it to protect the
fiduciary fund, and be liable if it fails to do so. Whether this
contract will be recognised as a trust by other ratifiers of the Trust
Convention remains to be seen. Certainly, because of the two
patrimonies, the cor****ation is said to own, and the legislation now
requires segregation of, the fund.56 It will be a considerable step
forward for the obligation trust model when this occurs, for the
'fiduciary contract' is the most stark form of that model.
France
The fiducie which the then French Government introduced into the
legislature in 1991 was on the other hand a vehicle of general
application. The Avant-Projet de Loi Relative a la Fiducie again
adopted the contractual model. 'La fiducie est un contrat', said the
proposed new Art 2062 of the Civil Code, and it was a fairly simple
structure -- it is a 'contract by which the settlor transfers all or
part of his assets to a fiduciary who is charged with achieving a
described end that benefits the beneficiaries'. Article 2063 then
provided that the contract must define the property of the fiducie,
the administrative and dispositive powers of the fiduciaries, the
description or mode of ascertainment of the beneficiaries, and the
benefits that the beneficiaries are to enjoy. Would-be variation or
revocation by the contractual parties, delegation by the trustee,
trustee dealings with third parties, and such usual matters that the
trust lawyer would expect are dealt with by successive articles. But
two proposed provisions should be particularly mentioned. What about a
segregated fund? The trust assets, held in title by the fiduciary, are
to form 'une masse s=E9par=E9e dans son patrimoine' (Art 2067). This was
the first mainland European formulation of the notion of a trustee
holding a segregated group of assets; here it is a fund within a
patrimony. Neither the fiduciary's creditors nor his heirs have claim
to the segregated fund.57 It is to be noted, however, that there is no
mention in these proposed articles of 'owner****p'; we are left to
infer that, unless the contract terms withhold one or more, the Art
2062 'transfer' confers upon the fiduciary all the rights and
obligations an owner would have. And it was expressly provided that
the fiduciary is to ensure that no benefit from those assets flows to
himself. The second provision to be mentioned speaks of the fiducie
creator (le constituant). Not surprisingly he may be a beneficiary
himself, but if the fiduciary commits a grave breach of trust either
the beneficiaries or the creator may ask the court to nominate in the
fiduciary's place a provisional administrator or a substitute
fiduciary. As to rights and remedies of creator or beneficiary against
wrongdoers, whether the wrongdoer be a stranger or the fiduciary
himself, and whether compensation or specific asset recovery is
sought, the Avant-Projet is silent.
The proposals contain nothing concerning the appointment of substitute
or new fiduciaries in the event of the fiduciary's death, retirement
or removal, and it is provided58 that, if the contract does not make
provision for the continuation of the fiducie, the court may terminate
the fiducie on the death of the fiduciary or the liquidation of the
cor****ate fiduciary. This is somewhat startling, but the article may
merely be saying that the contract has to provide for what is to
happen if a fiduciary (or the contract initiator) dies, becomes
incapacitated or is unable for other reasons to discharge the duties,
exercise the powers, or receive the property enjoyment created by the
contract. It is here confirming that the Law is not going to help. Or
it may be that the constituant and the fiduciaire are envisaged
throughout as parties to a contract and of course the only parties.
Once they are no longer in existence, the contract is at an end.
The 1991 Avant-Projet never saw enactment. Re****tedly because of the
fears of the French tax authorities that it would lead to significant
tax avoidance, not to mention evasion, it was withdrawn by the
government and discarded. A further draft fiducie was introduced in
1994, but that too came to nothing.
Though France has not ratified the Trusts Convention, and is
expressing no intentions in that direction, a third Proposition for a
fiducie was introduced by a member of the Senate into the legislature
in February 2005, and is currently being considered.59 The February
proposal is again for a contractual structure.60 Though the opening of
the proposed Civil Code, Art 2062, reads that 'the fiducie results
from a contract', it appears to be more exactly described in proposed
Art 2064 as le contrat de fiducie. As in previous drafts the whole
direction of the proposed legislation is an association between three
persons -- creator, fiduciary, and beneficiary -- the terms of which
association are as set out in the contract. It seems to be
contemplated, once beneficiaries are ascertained, that these persons
will live and be capacitated until the terms of the fiducie are at a
conclusion. The same provisions as existed in the 1992 draft fiducie
appear again in this 2005 draft; the court has power, if the contract
has not provided for the event, to terminate the trust on the death of
the fiduciary or on the liquidation of the cor****ate fiduciary. On
this occasion, however, the segregation is not of a fund within the
fiduciary's patrimony; the Proposition adopts the Luxembourg
arrangement of two patrimonies. The intended trust assets are
'transferred' to the fiduciary who then has title or more exactly
fiduciary owner****p61 in a patrimony distinct from the patrimony of
his personal assets. The distinct patrimony of the fiduciary assets is
described in Art 2062 as a patrimoine d'affectation called a
patrimoine fiduciaire.
In the course of its legislative examination of the proposed
legislation to date, it would appear that the Luxembourg 2003
'fiduciary contract' has been influential. It is understood that the
proposed articles for the Civil Code currently restrict the usage of
the fiducie to financial transactions. The Luxembourg 'fiduciary
contract' solely permits the use of its fiducie for cor****ate
financial trustee****p administering unit investment, securitisation,
insurance, the provision of security for loans, and cor****ate and
public pension schemes. A fiducie under the present French Bill may be
brought into being only by a cor****ate body subject to French
cor****ate tax, and, as in Luxembourg, only a cor****ate institution may
be appointed fiduciaire. In the present French Bill banks and
insurance companies are the permitted institutions.
This appears to be a not insignificant departure from the February
2005, proposal. The remarks made by Senator Marini on presenting his
Proposition to the Senate did indeed draw attention to the value of a
fiducie designed to provide property management for the constituant,
and the holding in fiducie for creditors the security with which they
have been provided by debtors. But he also foresaw the fiducie holding
property for the benefit of 'vulnerable' people such as those who are
injured in accidents or are suffering from illness. However, as things
are at present the use of the fiducie in France for testamentary and
inter vivos individual and family wealth management is not within the
scope of the proposed applicability of the French fiducie.62
Other European Jurisdictions
The Dutch recognise a bewind as agency. A bewind occurs when the owner
of assets transfers 'owner****p' to the beneficiary but irrevocably
assigns to the manager (the bewind-voerder in Netherlands law) for the
duration of the management all the rights and powers in the 'owned'
property.63 And we have seen64 the German Treuhand which is not
dissimilar in structure is classified as contractual; it is not
considered as fitting within any model of trust. The Liechtenstein
Treuh=E4nderschaft, adopted by legislation in 1926, is structured
expressly upon the common law trust.65 It is isolated from
Liechtenstein civil law as an independent, self-contained legislative
measure.
The 'mixed' jurisdictions
Several of today's offshore jurisdictions at one time or another,
often as small islands in larger seas or oceans, have been
periodically occupied by governing groups of both civil law and common
law backgrounds. Mauritius in the Indian Ocean comes at once to mind.
Then there is St Lucia in the Caribbean. However, within this category
of those with both indigenous civil law and common law influences, the
principal jurisdictions are Scotland, South Africa, Quebec and
Louisiana. Each one of these was a civil law jurisdiction when the
Anglo-Saxons with their common law arrived. Each has made its own
significant jurisprudential contribution in the development of a local
trust.
Scotland
Scotland is a civil law tradition that has its origin in Nordic laws.
At the same time it has been legally associated with England and Wales
since the union of these countries in 1705, with government and the
final court of appeal in London. Relations between the Scottish and
English legal professions have always been close. In practice there is
little difference between what can be achieved with the Scottish trust
and the English trust -- and that is something that should be
underlined -- but conceptually Scotland knows nothing of legal and
equitable owner****p. It is also influenced by the absolute owner****p
principle of received Roman law, and the numerus clausus doctrine.
Consequently, though the trustee has owner****p of the trust fund, the
obligation model is the conceptual basis of Scots law, with the trust
beneficiary having a personal remedy only in the event of any breach
of trust. What is seen as protecting the trust beneficiary from the
unlawful taking of trust property by the trustee or by the complicit
third party is not the right of in rem action against the taker, but
the fact that the trust assets constitute a segregated fund. Within
the last decade Scottish scholars have enthusiastically embraced the
rationalisation of two patrimonies, one patrimony the trust fund, held
fiduciarily, and the other the trustee's personal effects.66
South Africa
The law of South Africa when the English settlers arrived in the first
years of the nineteenth century was a mixture of Afrikaans law and the
laws, mostly concerning family and succession, of the indigenous
peoples. The Afrikaaners (originally the Boers) were originally mostly
from Holland and they had left that country in the seventeenth century
for settlement in South Africa. This was a time prior to the
codification of Dutch law. As a consequence, their law was the old
uncodified Roman-Dutch law, a mix of customary law and received Roman
law. A strong tradition of Roman law exists in South Africa to the
present day. Into this scene the English settlers brought the common
law trust following the 1806 English acquisition of the Cape Colony,
and thereafter it was part of the practice of law in South Africa when
ultimately the courts were called upon to explain its character in
this Roman-Dutch legal setting. The absolute owner****p of the civil
law again militated against the common law's dual owner****p, as it is
often called, and both testamentary and inter vivos trusts have come
to be expressed in Roman law terms as having an obligation base. Save
that an unjust enrichment action can be brought by the beneficiary
against a stranger in wrongful possession of trust property, the trust
beneficiary's remedies in the event of breach of trust are personal
only against the trustee. South Africans, also faced with the fact of
the absence of any in rem action in the beneficiary, have yet to be
persuaded, it seems, by the two patrimonies rationale that is so
evident in Scotland and France. The trust, with owner****p located in
the trustee, is the more widely used trust form in South Africa, but a
striking feature of South African law is its admission by statute of a
bewind trust. That is to say, the creator of the trust transfers
owner****p in the trust property to the beneficiaries, and the rights
of disposition and management are irrevocably assigned by the bewind
trust creator to the trustee for the duration of the trust. While The
Netherlands' Civil Code as amended in 1977 perpetuates the bewind as a
unique and distinct property administration arrangement, where the
'administrator' with the rights and powers of administration and
disposition is conceived of as an agent of the beneficiary owner, in
1977 The Netherlands was then very much of two views about the common
law trust.67 By contrast, though having in South Africa at that time
no clear conceptual base, the trust was very much a feature of South
African law and practice, and in 1988 the South African legislature
evidently considered that perpetuating the distinction between bewind
and the trust could no longer be justified.68
As with Scotland, South Africans scholars are of the view that any
usage of the trust that is possible with the common law trust -- from
family wealth management and distribution trusts to unit investment
trusts -- can be achieved and is achieved with the South African
trust. Increasingly familiar are businesses operated by trusts, in
lieu of cor****ate or partner****p organisation.69
Quebec
Quebec was a seventeenth century French-settled colony with the
coutume de Paris as its private law that in the eighteenth century
came under British sovereignty. In the following century English-
speaking businesses and trading companies became an established
feature of Montreal, the largest city of the province, and with these
residents came the usage of the common law trust. It may have been to
regularise this position that the Quebec legislature adopted in 1879 a
Bill introducing a trust (or fiducie) into the law and a little later
inserted these provisions into the 1865 Civil Code of the province.
Absolute or indivisible owner****p created real problems. The few
articles in the Civil Code that dealt with fiducie were a constant
source of debate as to what was the character of the trustee****p the
Code had created. It was considered conceptually impossible to
describe the trustee as 'owning' the trust property, and the Civil
Code seemed almost obtuse in saying only that the trustee was 'seized'
of the trust property. In 1994, however, a recodification took place,
and the op****tunity was taken to put the fiducie on a more firm
conceptual base. The continued need of a segregated fund was taken as
axiomatic, but fiduciary owner****p was rejected as a civil law
solution because a trustee with title to assets that he holds in a
fiduciary capacity only is not an absolute owner. The civil law
principle would continue to be violated. It was thought to be no
conceptual solution to describe the trustee as the owner of his
patrimony of personal effects, and owner also of an independent or
segregated patrimony that constitutes the trust fund.70 The ultimate
solution for Quebec was found in an adoption of Pierre Lepaulle's
theory of an 'unowned' fund.71 The trust fund is not 'owned' by anyone
(whether the creator of the fiducie, the trustee, or the beneficiary);
it is a fund 'dedicated' to an end or purpose, in which the trustee
has a personal right (whereby he can administer it) and the
beneficiary has also a personal right (whereby he compels the carrying
out by the trustee of the beneficiary's distribution entitlement). The
codifiers regarded the unowned patrimony as an entity, but not a legal
persona, which also appears to be the later conclusion of the drafters
of the Uruguay fideicomiso.72
The 1994 Quebec fiducie is created in just 38 Articles of the Code.73
The administration of property on behalf of others, which embraces
agency, tutor****p and curator****p and includes fiducie administration,
is covered in a further seventy-one articles.74 The fiducie articles
contain provisions on the nature of the trust (this is where the
patrimoine d'affectation is introduced), various kinds of trusts and
their duration, the administration of the trust (appointment and
office of the trustee, the beneficiary and his rights, and measures of
supervision and control), changes to the trust terms and to the
patrimony (as in common law jurisdictions, the court is given
extensive variation powers over both private and charitable trusts),
and termination of the trust.75
The new Civil Code carries articles that embrace almost all of the
trust issues that a reader of a common law trust text would expect to
find with regard to a common law trust. Civil Code articles are always
of a general statement character, it is true, but that being said
there can be little doubt as to the intention of the Quebec codifiers
to provide a compass of articles that adequately structure the
fiducie, when it is compared with the common law trust.
The 1879-1994 fiducie was limited to use in wills and donative inter
vivos transmissions, and was therefore exclusive of the many financial
and commercial trust usages familiar in common law Canada and the USA.
This meant that any such trust in Quebec had to be created in a common
law Canadian or a US jurisdiction, and be conducted administratively
out of that jurisdiction. The trustee would seek recognition in Quebec
for the trust, as a foreign trust under the province's conflict of law
rules. That necessity is now gone. The choice for the practitioner
between using for his client an Ontario trust, for instance, or a
Quebec fiducie is now based on business and taxation considerations,
as are other such inter-jurisdictional questions in other areas of the
law.76 The Code expressly provides that by way of fiducie a trust can
be created with any property for the benefit of an ascertained or non-
ascertained beneficiary, which includes a family discretionary or
successive beneficiaries trust, for a private or specific purpose, or
as an investment or pension trust for described persons or members of
a partner****p, association or cor****ation, including the shareholders
of the cor****ation. A fiducie may also be a charitable or non-profit
trust.77
As everywhere in the civil law world, the Quebec trust adopts the
obligation model. However, it is to be found in the Code under the
Book concerning 'Property'. This appears to be explained by the fact
that the fiducie fund is characterised as unowned dedicated property.
But the codifiers seem to have had something broader in mind. The
nineteenth century influence of the neighbouring common law trust on
the 1879 Quebec trust is accepted by most writers to be evident, but
the 1994 codification has a model in mind that, though as embracive in
applicability to different areas of law as any foreign trust law, is
nevertheless doctrinally of the civil law tradition. The 1994 trust,
more emphatically than its 1879 predecessor, is an independent or
stand-alone concept.
Louisiana
Claimed for France in the seventeenth century, the territory that is
now Louisiana became a French Crown colony in 1731 with its coutume de
Paris and was for a time under Spanish control thereafter, being again
a French possession when acquired by the USA in 1803. Its laws were
therefore of the Latin civil law tradition, and they were very much
influenced by the French Code Napol=E9on of 1804. For instance, because
the Code Napol=E9on abolished substitutions and the fideicommissum,
which was totally explicable, given French eighteenth-century history,
so did Louisiana. Trusts were recognised by the Louisiana legislature
on a more desultory basis prior to 1964, but in that year the present
Trust Code was adopted. Much modernity and comprehensiveness was
brought by the Trust Code, and greater flexibility in the use of the
trust idea was possible. Undoubtedly the costs associated with
Louisiana remaining the 'odd man out' in the American common law
immensity was the principal reason for the move, and the passage of
the new Trust Code was not easy for the academic community. Frequent
critical positions were taken in the journals by those civilian
scholars who felt the Trust Code was a 'sell out' to economic
considerations that were being allowed to prevail over the time-
honoured civil law traditions of the state. Some flavour of those days
is still to be sensed in civilian writing.
Though 'title' was conferred upon the trustee by the Trust Code, and
this has been widely interpreted by some civilian opinion within the
State as breaking with the civil law absolute owner****p doctrine, the
argument is made by these critics that in fact the Trust Code permits
the interpretation that the beneficiary 'owns' the trust fund, subject
to the real right of the trustee to dispose and administer.78 There
are obvious parallels here with the South African bewind trust, but
the rationalisation of the trustee having a right in rem is an
interesting addition to the South African analysis. Such a right would
certainly be resisted in 'mixed' jurisdictions other than Louisiana.
There are obvious differences in both expression and structure between
this Trust Code and the trust codes of the American common law states,
but it has to be said -- in the writer's opinion -- that the familiar
civil law rule that all beneficiaries must be ascertained and have
vested interests at the time of the creation of the trust is the only
conceptual rule that distinguishes the Trust Code from anything that
may be done, albeit in a somewhat different manner, with trust law in
any other state. The range of usages to which today the Louisiana
trust can be put -- from family property distribution to unit trusts
and trusts providing security to lenders -- is comparable with trust
case-law and statute in the USA at large. The Trust Code might indeed
be described as effectively an adoption of the property model of the
trust, but that description continues to be resisted by those proud of
the civil law inheritance of the state. Their argument would be that
it is instead firmly of the obligation model, and they would point to
the beneficiary being left by the Trust Code, as they contend, with
owner****p.79 However, they too would surely concede that the Louisiana
trust is a distinct, self-contained concept in Louisiana law.
Trusts in Civil Law Asia
Legislation as early as 1922 introduced the trust into Japan. It was a
response to a rapid development over the previous 20 years of Japanese
trust cor****ations acting as financial institutions; they were both
unregulated and operating in a conceptual area that was not known to
the Civil Code. This essentially cor****ate regulatory legislation of
1922 is still in force. So far as Korea, Taiwan and mainland China are
concerned, the story is different. Each of these three jurisdictions
was responding to an op****tunity for raising capital for large-scale
development projects with an investment concept that has obvious
attractions at a time of economic expansion. South Korea was already
into such expansion in the 1960s, and the date of its Trust Law is
1961. Taiwan introduced the trust into its law in 1996, and mainland
China did the same in 2001.
Japan
The Japanese trust has been described as 'a creature of contract and
statute',80 but, even if one were not aware that the legislative
drafters had before them in 1921 the then California Trust Code, it is
not difficult to surmise from the articles of the Trust Law of 1922,
that the legislators were attracted by the rights and remedies
associated with the common law trust model. In moving in that
direction, however, the challenge was considerable, because as their
Civil Code the Japanese had adopted in the last years of the
nineteenth century the first draft of the German Civil Code of Private
Law.81 In 1921 the trust concept meant for most lawyers across the
world the common law trust, and the Japanese would have known of
pre-1914 European writing that declared the trust unacceptable to a
civil law code. The absolute owner****p principle certainly led the
Japanese legislature into some ambiguity in explaining what is meant
by a 'trust' in this Trust Law. But, meaning of 'trust' aside, what
emerges from the remaining articles is an express trust that is
independent of the instrument of creation. It has provision for
effective trustee administration so long as the dispositive terms in
the instrument are still being carried out, unlimited -- that is -- by
the lives of contracting parties. Segregation of the trust property
from the trustee's own assets, rights and obligations is expressly put
in place.82 But in addition provision is made for the appointment,
resignation and court removal of trustees, the transfer of trust
property on the appointment of a new trustee, and interim
administration pending the new trustee being appointed (Arts 48-56).
Extensive court powers are introduced for the supervision of trust
administration and sup****t of trustee****p procedures.83 These are
features of the Trust Law that one immediately associates with a self-
contained and distinct working concept.
Earlier, however, reference was made to article 1, dealing with what a
'trust' means. This article has always presented a theoretical issue.
In defining a trust 'within the meaning of this Law', it creates
something of an enigma. The article states in the authorised English
translation that:
'trust ... shall signify to transfer or otherwise dispose of a property
right and cause another person to administer or dispose of the
property in accordance with a specific purpose.' (emphasis added)
There are two schools of opinion in Japan about this language. The
traditional view is based
on the proposition that every interpretation of a civil law code must
have its source within the code itself, and if there is ambiguity or
silence the ultimately prevailing interpretation is justifiable in
terms only of what the code impliedly provides. The Civil Code is
intended, the contention runs, to fill in where the Trust Law is not
clear or 'fails to provide'. It is a more recent minority view that
from internal evidence it can be seen the newly introduced Trust Law
was intended to be interpreted as a stand-alone legislative measure,
distinct in conception from Civil Code principles.
The traditionalists, seeking to harmonise the new Trust Law with the
Civil Code, interpret separately the two parts of the Art 1
definition. The first part is said to state that the transfer of which
it speaks creates in rem rights in the transferee,84 but the second
qualifies that by stating that the obligation and right of the trustee
to administer is in personam.85 It would follow from this, say the
minority, that the beneficiary too has only in personam rights to
enforce the trust. Yet that is not so. It is provided in Art 27 that
the beneficiary may demand of a trustee who has made away with trust
property that he make 'restitution of the trust property', a provision
that therefore confers -- so it is argued -- a right in rem upon the
beneficiary (as does the common law system).86 Also, Art 31 goes
further. It confers upon the beneficiary the right to have a transfer
of trust property to a third party set aside as a void act. This right
can be had by the beneficiary if the trust assets in question had been
registered as fiduciary property or the wrongdoing trustee transferred
to a third party who knows or ought to have known of the wrongdoing.
This again, the argument runs, should be understood as an in rem
beneficiary right; the transfer being void the third party must
retransfer the property in question to the trust fund. Article 1, the
minority say, should be read not as two parts but as one proposition.
The in rem rights of the beneficiary, and indeed the structure of the
whole Trust Law, point to this Law -- with its distinct provisions --
being intended as something separate and apart from the Civil Code
doctrine.
There are arguments favouring both sides. A traditionalist could point
to power being given to the trust creator/promisee to contest a trust
creditor's seizure of trust property,87 and also that same promisee's
power to demand of the trustee production of trust administration
do***entation.88 This shows, it can be said, that the act of transfer
to a trustee is not only limited in scope to the discharge of the
objects set out in the trust terms, but that there remains after the
limited transfer an ongoing agreement right of the trust creator to
ensure that the trustee is properly discharging his trust duties. The
minority contrary opinion could respond that on the other hand the Art
40.2 power expressly vests in the trust creator's heir as well as in
the creator. This would suggest the trust is an institution distinct
from the personalities who as contractual parties set it up.89
This Trust Law, as we have noted, is contract as well as statute, and
the promisee as promisee -- rather than any beneficiary -- might be
expected to be the party who enforces the promisor's duties. However,
in this Trust Law the promisee appears to have this authority only on
four occasions, and moreover in each of these cir***stances the
beneficiary too may bring action.90
As we have seen, the promisee, or his heirs, the beneficiary or other
trustees, may sue the wrongdoing trustee for personal or proprietary
relief,91 and the beneficiary may 'avoid [the] disposal' of trust
property to a culpable third party.92 Nothing is said, however, of the
duty of the trustee to pursue wrongdoers who cause loss to, or
misappropriate, trust assets. Other duties of the trustee are clearly
set out, but the Trust Law says nothing as to who has remedy to
enforce the trustee's performance of those duties. On the other hand,
when asked, the court is given extensive authority to assist in the
administration and the discharge of the trust and its objects, and
this is a characteristic of the common law system.
In fact the existing Trust Law appears to attempt within civil law
principles to achieve the results of a property model of the trust.
The more persuasive view is that the trustee owns as a result of the
transfer, but, when the trustee is responsible for wrongdoing vis-=E0-
vis the trust property, owner****p powers are given to the beneficiary.
This represents a subtraction of powers from the owner in order to
gain a protection of the trust fund that otherwise would not exist, at
least without a court intervention. Other articles suggest a fiduciary
character is intended to attach to trustee owner****p.93 There is no
mention of contract or agreement anywhere in the Trust Law, and were
it not for the retention of a role for the trust creator during the
trust's lifetime, as previously mentioned, the property model
character of this 'trust' would be apparent. It is only the linguistic
expression in Art 1 that can suggest otherwise, and it is hard to
accept another meaning when such an alternative explanation is not
maintained in the remainder of the Trust Law's provisions. However, a
new draft amending the 1922 Trust Law is to be brought before the Diet
in November 2006, and this will probably return to how the trust
should be defined or described for the purposes of the proposed new
Trust Law. Whether it leans towards the obligation model (eg, with a
substantially enhanced set of powers for the trust creator wherewith
to enforce the trust objects and protect the trust fund) or the
property model (eg, the express conferment of powers upon the
beneficiary to recover trust property for the trust fund) will be
seen.
Charitable trusts were introduced by the 1922 Trust Law, and
interestingly enough the same policy was later adopted by each of
South Korea, Taiwan, and mainland China. A much larger role for state
authorities is put in place by each of these Asian jurisdictions than
is familiar in common law jurisdictions.
A Trust Business Law of 1922 accompanied the Trust Law; this contained
rules concerning the rights of the trust creator and beneficiary, and
the trustee's duties and powers. In 2004 this Law was substantially
updated to accommodate contem****ary trust business, extending the
types of property which can be held in trust, enabling companies other
than trust banks to conduct fiduciary business, and modernising the
power of delegation. Save for specific limited cir***stances trustee
disclosure to the beneficiary is now mandatory. Common law
practitioners will particularly note the prospective trustee's
obligation to explain to the would-be settlor the trust agreement
about to be executed. The intended revision of the 1922 Trust Law is
to follow in the wake of the revision of the 1922 Trust Business Law.
It has been claimed by a knowledgeable source that 'the modernization
of the Trust Law will bring [the Law] in line with the commercial uses
of trusts'94, and it appears from this IFLR re****t that the
innovations now introduced follow closely the type of change that
common law trust jurisdictions have recently considered, or soon will
be considering.95 In line with the common law trust model, fiduciary
duties in Japan will become default instead of the present mandatory
rules, and provisions are proposed for the variation of trusts, the
division of a single trust's assets for investment purposes, the
investment of combined trust funds,96 and, though the trust agreement
be silent, the trustee power to delegate whenever it is appropriate to
do so.
One has always to remember that the Japanese are thinking of the trust
as a commercial and business concept. This explains the intention to
permit the beneficiaries by a majority vote to vary the trust terms,
provided the purpose of the trust is retained. Comparison with the
powers of cor****ate shareholders in general meeting is obvious, as is
the intention to combine distinct trust funds or to divide trust
assets for management purposes. The draft Trust Law also allows
declarations of trust (to permit securitisation by companies of their
receivables), trusts to manage businesses (in lieu of incor****ated
management), and asset security-against-loan (or mortgage) syndication
trusts, which hitherto have been known only to jurisdictions that have
the property model of the trust. The Japanese Diet is to be asked to
approve limited liability trusts, where the trustee's liability is
limited to the value of the trust assets, unless the trustee is
delictually liable. Trustee limited liability, where the trust is a
vehicle for investment by the public, constitutes a move beyond that
which any Commonwealth common law mainland jurisdiction has yet made.
The draft Trust Law allows trust units to be traded on the stock
markets as securities, and it introduces private purpose trusts.
There can be few with commercial or professional interests in Japan
who are not awaiting with considerable interest the fate of the draft
Trust Law. Not least of interest will be the attitude of the Diet
towards a Japanese trust that, while moving along the lines of the
Anglo-American property trust model, has still to find a modus vivendi
with the domestic Civil Code. On the other hand in the writer's
opinion the draft new Trust Law is unparalleled in the civil law world
in the sophistication of trust business to which it looks forward.
Enthusiasm for the placing of receivables and futures into
securitization trusts has already gripped Japan, and this proposed
legislation will make possible much of what at the moment is practice
built on something of an uncertain legal base.
The Republic of Korea (South Korea)
In 1961, when South Korea was in the early stages of its striking
economic expansion, government introduced into the legislature a Trust
Act. It closely reflects the 1922 Trust Law of Japan. Essentially it
seems to constitute a more 1960s organisation and expression of the
articles that are to be found in the earlier Japanese Trust Law.
On the controversial Art 1 meaning of 'trust', it describes the trust
as 'a legal relation' between the creator of the trust and the
trustee, involving -- as its equivalent of the first part to the
Japanese Art 1 -- the 'transfer' or 'disposition' of 'a specified
property' to the trustee. This more clearly suggests that property
owner****p p***** to the trustee. Otherwise this Trust Act gives the
same protection power over the trust fund as the Trust Law in Japan,
and has the same silences. It also gives the trust creator the same
four powers with regard to the enforcement of the trust.97
The Republic of China (Taiwan)
The Trust Law of Taiwan was promulgated only in January 1996, and
therefore would be expected to reflect experience both in Japan and
Korea. The long association over half a century until 1945 with
Japanese rule and administration, aside from past Dutch and Spanish
influences, have left a distinct civil law stamp upon Taiwanese
private law, and the adoption of the obligation model for any new
trust would have seemed to follow. What appears to have happened,
however, is that, though following that model, a pragmatic *****sment
was made, something that would not have been regarded as tolerable had
the doctrinalists held sway. It was the observed success of the
Japanese trust for no less than 50 years after 1946, and perhaps also
that of Korea, that appears to have led the Taiwanese government to
have before it the 1922 Trust Law of Japan as the draft for its own
legislation. The Taiwanese articles are frequently reworded to reflect
modern trust forms of expression, and with the insertion of new
articles there has been a rearrangement of the articles in the draft,
but in terms of the model of trust that has been adopted it is that of
the 1922 Trust Law.
The Taiwanese Art 1 defines the trust for the purposes of the Trust
Law, but differs in one respect from the Japanese and Korean Art 1.
The English translation provided on the official Republic of China98
website is worded as follows:
'the settlor transfers or disposes of a right of property and causes
the trustee to administer or dispose of the trust property according
to the stated purposes of the trust for the benefit of a beneficiary
or for a specified purpose.' (emphasis added)
This will be recognised as the two part statement found in the
Japanese Art 1.99 However, the above language in the Taiwanese Art 1
is preceded by the words, 'the term 'trust' refers to the legal
relation****p in which'. The 'legal relation****p' is not that in the
property model between trustee and beneficiary (or trustee and
abstract purpose), but between trust creator and trustee. This is
evident from Art 8 where it is provided that 'a trust relation' is not
extinguished by the death, bankruptcy or incapacity of 'the settlor or
trustee'. The continued significance of the trust creator once the
inter vivos trust has been created is shown by the provision100 that a
change in the 'method of administration'101 requires the consent of
the trust creator (or the trustee or beneficiary),102 and each of
those three parties is empowered to apply to the court for such a
change.103 Any one of the three may 'request' the trustee to
compensate or restore in the event of that trustee's breach of trust,
104 and the trust creator as well as the beneficiary has the right to
monitor and make copies of the accounts maintained by the trustee.105
All this is in line with the Japanese and Korean legislation, but the
Taiwanese Trust Law goes much further in conferring powers upon the
creator of the trust. However, Taiwan also places a much heavier
stress on the contractual 'legal relation****p' between creator and
trustee. The creator (beneficiary or other trustees) may sue the
trustee for breach,106 apply to court (as may the beneficiary) for the
removal of a trustee, and the creator may appoint a new trustee,107
without the reservation of this power to do so in the instrument. The
creator (the trustee in question or other trustees, or the
beneficiary) may apply to court for the increase or reduction of
trustee remuneration,108 but the creator can terminate the trust only
if the creator is the sole beneficiary.109
All these trust creator powers, save for the appointment of new
trustees,110 are also exercisable by the trust beneficiary,111 and by
the trustee or trustees where the nature of the power would call for
trustee action. As a result the continuity of trust administration and
disposition in favour of qualifying beneficiaries is maintained. In
other words, the law reserves powers to the trust creator, but
conceives of the beneficiary being a constant phenomenon, and the
trustee or trustees as occupying the office of asset management.
Moreover, as to the beneficiary and in rem rights, the beneficiary is
entitled to have the trustee 'restore' property that the trustee has
'damaged' or wrongly disposed of in the course of improper
administration. The trustee can also be required to 'disgorge'
improper gains.112
Though the Taiwanese Trust Law was enacted long after the coming into
force of the Trusts Convention and the commencement of the debate as
to the nature of the 'trust' to which in civil law Europe that
international convention gave rise, it is an obligation model of the
trust that, like the 1922 Trust Law of Japan, has a leaning towards
the property model. The trust is conceived of as something more than
the instrument, contract or last will and testament, by means of which
the trust is created. What more is a matter of opinion, but the
present writer would say that the emphasis is on the care with which
the legislation of these jurisdictions is concerned with the ongoing
nature of administration until the trust object is achieved,
regardless of the death of the trust creator or of the original
trustee.
That Japan and Korea, and now Taiwan, have moved towards the
attributes of the property model within a civil law code structure
offers to apprehensive civil law states an interesting interpretation
of what can be achieved, despite the difficulties.113
The People's Republic of China (PRC) (mainland China)
The PRC Trust Law which came into force on 1 October 2001, is not
dissimilar overall in its structure and provisions from the Trust Law
of Japan. It provides protection of the trust fund against wrongdoing
when a disposition in breach of trust is made by the trustee. The
court may 'nullify' the disposition, which in the hands of the trust
creator or the beneficiary, though it is merely a declaration of
voidness, could constitute an in rem remedy. An in personam remedy
against the trustee for the making good of loss to the trust fund is
also available. The trust creator and the beneficiary each has an in
rem remedy against the third party transferee who knows the trustee's
disposition was a breach of trust.114 These proprietary remedies, or
effects of remedy granted, again show a marked lean towards the
property model, but the PRC trust puts considerably more emphasis on
the contract than does Japan or Korea; it emulates the Taiwanese
example and extends the creator/trustee agreement character yet
further. Japan and Korea bring into existence four instances where the
creator has powers to intervene in the ongoing trust management; the
PRC bring into existence eleven.
There are two particular respects in which the PRC obligation model
goes its own way. First, though a trust may be created by a contract,
will or 'other written do***ents',115 so that agreement is not the
essential element of the creative act (or of the trust itself), the
Trust Law conceives of the creator, trustee, and beneficiary entering
into 'Trust relation****ps'.116 Each member of this 'trinity', as the
relation****p has been described, is actively engaged throughout the
duration of the trust, subject to the trust creator being alive and
capacitated.117 The creator has all the rights, the beneficiary always
having the same rights, that are granted by the Trust Law for the
monitoring of trust administration and the enforcement of the
trustee's duties. The trust creator is present as a party to the
agreement; the third party beneficiary must be empowered by the Trust
Law or he has little right to intervene in the defence of his
interests.
Secondly, though as we have seen the Trust Laws of Japan and Taiwan,
somewhat abstrusely, require the settlor to 'transfer' or 'dispose' of
'a property right' (Japan) or a 'right of property' (Taiwan) in order
to fund the trust, mainland China takes a different route. In the
English translation of the Trust Law, Art 2 provides that 'the term
"Trust" means the acts whereby the settlor, based on his trust in the
trustee, entrusts the rights in his property to the trustee ...'.
(emphasis added). Whether this means that for the purposes of the
trust the trustee acquires from the trust creator the owner's powers,
or that the trustee acquires possession and civil law 'owner****p', is
not said.118 Article 14 does describe the trust property as 'the
property that the trustee obtains as a result of his accepting the
Trust' (emphasis added), but obtaining property, like entrusting
rights in property, is a lay term.119 It has no legal meaning.120
In effect, whatever may be the practice among PRC settlors, the Trust
Law appears to be all embracive in how the trust may be created. The
Trust Law appears to permit trust creation by (1) a transfer by the
trust creator of 'fiduciary owner****p' to the trustee, (2) retention
by the creator of owner****p (definable as indivisible owner****p), but
an assignment to the trustee by the creator of the rights to dispose
and manage the property that will revert to the creator on the
termination of the trust, and (3) a transfer of owner****p to the
beneficiaries, with a reservation to the trust creator of the rights
to dispose and manage, and their assignment by the creator to the
trustee for the purposes of management and disposition, with
thereafter a similar reversion to the beneficiaries. In both the
second and third instances, it will be seen, the trustee while being
without owner****p would have 'the rights in property' wherewith to
manage and dispose.121
What evidence warrants a conclusion of such embraciveness? The
sup****ting interpretation is that the PRC Trust Law requires that the
trust creator make over to the trustee as a trustee 'rights of
property', and states that the trustee will exercise these rights
(management and disposition) 'in his own name' when discharging his
duties on behalf of the trust object. This is the character of the
South African bewind trust,122 and it is the technique of the German
testamentvolstrecher and the Dutch bewind. In the South African case
-- possible also, it seems, in mainland China -- there is a transfer
of 'owner****p' to the beneficiaries, and a passing by the trust
creator of the owner's rights and powers for the duration of the trust
to the trustee.123 On the other hand, if this technique of creating a
trust is indeed being authorised, it is strange that nowhere does the
Trust Law stipulate that the trust creator's grant of the rights of
administration and disposition be irrevocable until the trust ends or
is terminated. The counter argument to this interpretation of the
words, 'rights in his property', could be that, as the Japanese Trust
Law (Art 1) speaks of the trust creator transferring or disposing of
'a property right' and the Taiwanese Trust Law (Art 1) refers to the
trust creator doing the same with 'a right of property', so the
English translation of the PRC Law is simply referring to 'property'
itself when it says, 'rights in his [the trust creator's] property'.
These are fundamental structural questions, but such questions aside
the Trust Law creates a segregated fund and provides for the trust as
an independent property management vehicle until the object(s) are
achieved, or the trust is terminated. Like the trust legislation of
Japan, Korea and Taiwan, it expressly provides for the segregation of
trust assets from the trustee's own assets, his heirs and creditors.
124 It also contemplates a trust (an entrusting by S and management or
disposition by T) distinct from the instrument of creation. The trust
therefore endures beyond contract or any other mode of creation. The
Trust Law expressly provides as a default rule that the trust shall
not terminate because of the inability of the trust creator or trustee
to continue acting.125 It provides for trustee resignation126 and
removal, plus interim care for trust affairs,127 and for the
appointment of substitute trustees.128
At the same time, with the creator able to retain owner****p of the
trust property, determine the content of the dispositive and
administrative terms of the trust, and monitor and enforce proper
trustee administration, it is capable of being similar in character to
agency.129 It must be said, however, that the trust creator has no
power to amend (or vary) the terms of an on-going trust; he can only
under certain cir***stances terminate the trust.130
Trusts in Islamic law
The waqf or wakf of the shari'ah has for centuries been an institution
in the Arabic and Muslim countries, and among Muslim minorities
elsewhere. It involves the passage of the management control over
property by donatory transfer of control to one person and the right
of benefit in that property to others. Traditionally the waqf is
comparable to the role of a common law charitable or public trust, and
to this day it is closely associated with religion and pious
almsgiving. However, some countries or jurisdictions do permit the
individual to create a waqf for the benefit of the individual's
family, and family waqfs are an accepted phenomenon in these
jurisdictions. It is normally provided that on the death of the last
family member the waqf assets shall pass to charitable purposes for
perpetual dedication to such purposes thereafter. It is said that the
residuary assets then pass to God. The application of the waqf to
commercial or business purposes has not to date taken place.
The property in question is normally land, the continuous use of which
will further the described purpose, or benefit the designated persons.
Shari'ah law confers forced share rights upon the family members of
the deceased over his assets, and the waqf must respect those
distribution rights. The assets are likely to be immovable property
within the jurisdiction of a shari'ah court. Investment for return on
the part of the waqf manager -- and therefore 'real subrogation' or
the trust concept of a 'fund' -- appears not to be a characteristic of
a waqf. There are what might be called shari'ah schools of doctrinal
opinion within Islam, and as a consequence in some jurisdictions the
donor transfers full owner****p to the mutawalli (or manager), while in
others the donor retains the owner****p, and the manager operates with
such rights as are transferred to him. The manager will also be liable
for his wrongful conduct vis-=E0-vis the waqf.
The analogies with a trust are obvious, but it seems hardly possible
to contend that the waqf is itself a functional model of the trust
concept.131
In 2005, Dubai enacted a Trust Law,132 the first such act to have
taken place in the Middle East. It applies to the international
financial centre of Dubai. It validates express trusts, whether
private or commercial, and whether for persons or purposes (charitable
or non-charitable). The legislation incor****ates elements that are
familiar within all offshore trust jurisdictions, such as no
recognition of the forced share laws of other jurisdictions. It is
designed as a property model of the trust, and indeed in default of
provision in the trust instrument or the laws of Dubai resort may be
had by Dubai courts to the common law of trust and equity. However,
considerable care is required of the trust creator in honouring the
requirements of shari'ah law that may be applicable.
1 Professor Emeritus, University of Victoria, BC, Canada.
2 This paper was published in John Glasson (ed), The International
Trust (Jordans, 2002), second edition to be published shortly. It was
reproduced in John Glasson (ed), International Trust Laws, looseleaf,
vol I (Jordans), and in Donovan Waters, 'The Future of the Trust from
a Worldwide Perspective' [2004] JTCP 199.
3 Resulting and constructive trusts raise issues that are distinct
from intended trusts, and are largely singular to the common law
trust. Statutory trusts are excluded also. This later paper,
subsequently revised for publication, was given at the STEP Asia
Conference, Hong Kong, 13 October 2006.
4 The word, 'models', was introduced by Professor Maurizio Lupoi
in his work, Trusts: A Comparative Study (CUP, 2000), originally
published in Italian as Trusts (Giuffr=E9 Editore, 1997), where he
speaks (see pp 5-8) of an English-law (or traditional) trust model and
an international trust model. He is there referring on the one hand to
the trust law in the USA, England, Canada, Australia and New Zealand,
and on the other hand to the trust law of the offshore 'financial
centre' jurisdictions, the mixed common law and civil law
jurisdictions, and the pure civil law jurisdictions. He suggests a
third civil law model may be emerging from the international trust.
5 The text for each of these 53 'trusts' is supplied in M Lupoi
(ed), Trust Laws of the World, Vols I and II (ETI Editore, 2000). As
of the present date the number is in excess of 53.
6 JH Langbein, 'The Secret Life of the Trust: The Trust as an
Instrument of Commerce' (1997) 107 Yale LR 165, at pp 166 and 178.
7 The use of the Latin, persona, emphasises the features
associated with legal personification, whether the object in question
is human or an abstract concept. An entity is an unincor****ated or non-
personified body or concept, to which the law extends certain
qualities or attributes. A 'trust' has been described in this manner,
though the trust is but a management or distribution obligation
assumed by a trustee vis-=E0-vis a beneficiary (or trust creator). A
trust may also be described as a 'legal institution', again as
distinguished from a legal persona. See, eg, E Cameron and others
(eds), Honor=E9's South African Law of Trusts (Juta, 5th edn, 2002), at
para 1, p 1.
8 Though we recognise that within the civil law system some
jurisdictions regard the trust as a legal entity, and in the common
law US statutory business trusts are described as 'juridical
entities'. These trusts will be discussed later.
9 In this paper I will hereafter use the term, 'trust creator', to
refer to the settlor. That term is ponderous, but 'settlor',
'trustor', 'founder', etc, are terms used in particular jurisdictions.
Regrettably none of those alternatives is comprehensive of all trust
creators, its meaning unambiguous and recognised in all jurisdictions.
10 Several offshore jurisdictions, both of common law and civil
law background, have challenged this element of trust theory by
granting statutorily to the settlor or others powers to direct or veto
trustee decisions. By force of statute, therefore, a trust remains
such though control may thereby be lost to the trustee. This will be
discussed later in the paper.
11 In the present writer's opinion it is an indispensable element
distingui****ng trust from agency. Trident Holdings Ltd v Danand
Investments Ltd (1988) 49 DLR (4th) 1 (Canada; Ont CA). The only
debateable issue is how minimal the degree of control may be.
12 So-called 'types' of trust would then refer to express and
implied trusts, resulting and constructive trusts, a formulation that
is not under discussion in this paper.
13 It is often forgotten that the popularly used term,
'owner****p', has no meaning in the common law. Most common law people
think of owner****p as title.
14 Since under feudalism all land was owned by the Crown, the
Crown's subject could have only a derivative interest in the land. The
subject held an estate, quantified as a fee simple, fee tail, life
estate, or estate pur autre vie. Land in the tenth and subsequent
centuries was by far the most im****tant property. In the fourteenth
century for the purposes of the use, extended in the seventeenth
century to the trust, the Court of Chancery adopted the same estates
doctrine for all property (realty and personalty) held by the feoffee
to uses, and later by the trustee.
15 Though the proprietary right is now established, there were
scholars well into the twentieth century who insisted that the
beneficiary has only a personal right against the trustee.
16 Today's so-called offshore trust jurisdictions across the world
discussed in Part II of this paper, are all of the 'pure' common law
tradition or, being 'mixed' jurisdictions or of neither civil nor
common law tradition, have adopted the property model of the trust.
See, eg, Dubai, at n 132, below, which is of the Islamic tradition.
17 Trait=E9 th=E9orique et pratique des trusts, Paris, 1932.
18 A set of lectures given in the UK in 1975, prior to this
period, under the title, Trusts and Trust-Like Devices, edited by WA
Wilson, an eminent Scots professor, was published by the UK National
Committee of Comparative Law in 1981. Though the Convention solely
concerned conflict of law rules, it was the Hague Convention on the
Law applicable to Trusts and their Recognition, in force January 1986,
that was the stimulation for much of the subsequent writing and
interest in this subject.
19 Some have seen the common law trust as a jurisprudential
hodgepodge of property and obligation law and various remedies, the
whole being impervious in civil law terms to analytical organisation.
20 M Lupoi, op cit n 5, above; M Lupoi, op cit n 4, above. See
also AM Honor=E9, 'Trusts: The Inessentials', in J Getzler (ed),
Rationalizing Property, Equity and Trusts: Essays in Honour of Edward
Burn (OUP, 2003).
21 Personification produces segregation. The persona says, 'That's
mine.'
22 'Obligation' refers to a model drawn from the law of
obligations (contract and delict).
23 The 'mixed' jurisdictions represent both traditions, codified
(eg, Quebec) and uncodified (eg, Scotland).
24 The Quebec Civil Code of 1994 classifies it as 'Property' (Book
IV), as opposed to 'Obligations' (Book V). It seems this is because
the fund is itself an unowned proprietary entity.
25 The distinguished Japanese scholar, Professor ****nomiya, who
was an advocate of the Lepaulle 'entification' theory, described the
Japanese trust as 'unique and isolated in the Japanese legal system,
like a drop of oil floating on a pool of water.' See the International
Financial Law Review, Guide to Japan 2004.
26 This is discussed later with regard to the trusts of several
Asian jurisdictions.
27 There is no reason why, as in a unit trust, contributors to an
existing trust fund should not hold units of value in a pooled trust
fund. The fund will constitute the assets that are segregated.
28 The common law creates the same problem in permitting the
settlor to declare himself the trustee of his own assets.
29 For such an instance, see the South African bewind trust.
30 It is often said that the common law leans too heavily towards
the interests of the trust beneficiary, and would be better advised to
'balance' things more evenly by stopping short of in rem remedy
against the interests of the wrongdoer's innocent creditors. The
common law prefers the argument that the unsecured creditor makes a
considered decision to take the business risk. The civil law finds
that argument unacceptable unless perhaps all valid trusts, their
terms and assets must be fully disclosed in a public register.
31 It may be asked whether the fiduciary obligations of (1) no
conflict of interest and duty, or conflict of duties, and (2) no
delegation, are not essential elements. The common law has always
allowed both such obligations to be set aside at the choice of the
trust creator, and the fact that they have long been mere default
rules must mean that, though these obligations are a significant
feature of the property trust concept, they cannot be seen as
essential elements.
32 See op cit n 25, above.
33 See, eg, the real subrogation of the Ethiopian Civil Code
fid=E9icommis, which enables the fid=E9icommissaire to manage new
investments acquired in the process of investment and reinvestment
(Book I, Title III, Chapter III, Section III, Art 530.)
34 A Dyer and H van Loon, Re****t on Trusts and Analogous
Institutions, Preliminary Do***ents No 1, May 1982. Proceedings of the
Fifteenth Session, 984, Book II, Trusts.
35 H K=F6tz, 'National Re****t for Germany', in DJ Hayton, SCJJ
Kortmann, HLE Verhagen (eds), Principles of European Trust Law
(Kluwer, 1999), at pp 85-103.
36 DJ Hayton (ed), Extending the Boundaries of Trusts and Similar
Ring-Fenced Funds (Kluwer, 2002).
37 There is also the civil law rule that all the assets 'owned' by
a person must be available to his creditors, and segregating the trust
fund clearly permits a person to claim that certain of his former
'owned' assets are held for others and are therefore not available to
his creditors.
38 See n 17, above.
39 Panama (1971), Columbia (1971), Ecuador (1993), Peru (1993),
Argentina (1995) and Uruguay (2003).
40 Declaration of trust, ie the settlor declares himself to be a
trustee of assets that are already his, is not a feature of this Latin
American code provision. Most civil law lawyers would steer clear of
this type of common law trust.
41 Ley 17.703.
42 The beneficiary's personal creditors may only lay claim to the
proceeds of realisation of fideicomiso assets: Art 7.
43 It is understood that the trust interno of Italy operates in a
similar manner. See n 50, below.
44 With unregistered fideicomisos the third party need not be
concerned with whether the fiduciary has power to enter the intended
contract, unless the fiduciary's action is quite obviously a breach of
his trust or the third party knew the act was breach.
45 The Law confers no specific powers upon the fiduciary, or upon
anyone else.
4 See further L Turano, 'Uruguay's Fideicomiso' (2005) 19 Trust
Law International 149, including an unofficial translation of the Law
which has been drawn upon for this paper.
47 The jurisdictions of Jersey, Guernsey, Malta, Gibraltar and
Cyprus are of civil law background, and each possesses a domestic law
of trusts. But, |
