Gilly Kennedy-Smith

Gilly Kennedy-Smith

Mourant LP Partner

The Detrimental Face of Power

[Dynamic date]

07 October 2021

Senator Bernie Sanders of Vermont once said, “A nation will not survive morally or economically when so few have so much, while so many have so little”,1 and the same can be said of a trust. If a settlor wants their trust to be well fortified against any form of matrimonial, state or future creditor attack then they just need to step back from the start. It is imperative that advisers counsel their clients that power is best held in the hands of independent fiduciaries who will not pander to the every wish of the settlor. This article, originally published in Private Client Business, Issue 5 2021, examines where planning can go awry and the steps that advisers should encourage their clients to take, in order to avoid issues arising when the face of Janus2 turns around and fortune no longer favours them.

Trustees and drafters are commonly asked, when assisting a settlor of a trust, what level of involvement the settlor can have in the trust and its ongoing administration. The answer is not as straightforward as statute might suggest. Being legally entitled to do something does not mean this is a sensible choice and in reality it can put the settlor at greater risk from third-party claimants, for example, creditors of the settlor3 and spouses or former spouses in the event of a divorce.4 In addition the matrix of facts surrounding the settlor and their onshore and offshore connections may suggest that retaining any control is not recommended.

In the case of Pugachev his political position as a Russian oligarch and the high probability of a change of favour and the disastrous change of circumstances this can bring should have led to him taking a back seat to the future running of his trusts. We will look at what he put in place and also how he could have changed his planning, but first it is useful for us to take a step back.


What creates a valid trust?

Any trust must meet the English common law test (which also applies in Jersey, Guernsey, the Cayman Islands and BVI) of the three certainties of intention, subject matter and object.5 With regard to the certainty of intention, on which this article will focus, it is a question of whether or not the settlor gave up their rights in the property held in trust and genuinely intended to create immediate rights in another person or class of persons. If they did, then the first of the three certainties is satisfied; if not, only a bare trust or nomineeship will have been created (assuming legal title has been vested in someone other than the settlor) and resulting tax issues can arise.

That is not always a clear-cut question though. In Re Shirley6 “the settlor was entirely free to deal with the fund as his absolute property” and clearly this is not sufficient to meet the requirements of a valid trust.

The irreducible core of obligations placed on a trustee and examined in Armitage v Nurse7 is also relevant and care should always be exercised when reserving powers to bear this obligation in mind.8 However, it is also worth noting that Armitage related to the validity of exoneration clauses and the minimum standard of care, so in relation to reservations of power it is relevant but not directly.

What does Guernsey law permit?

In 2007 the law in Guernsey was updated and s.15 of the Trusts (Guernsey) Law 2007 was introduced (the Guernsey Trust Law). That section provides that the reservation or grant by the settlor of powers (to themselves or to any other person) does not, in and of itself, invalidate a trust. The list of powers that can be legally reserved is extensive and includes:

"(a) a power to revoke, vary or amend the terms of the trust or any trusts or functions arising thereunder, in whole or in part,9

(b) a power to advance, appoint, pay or apply the income or capital of the trust property or to give directions for the making of any such advancement, appointment, payment or application,

(c) a power to act as, or give directions as to the appointment or removal of, a director or other officer of any corporation wholly or partly owned as trust property,

(d) a power to give directions to the trustee in connection with the purchase, retention, sale, management, lending or charging of the trust property or the exercise of any function arising in respect of such property,10

(e) a power to appoint or remove any trustee, enforcer, trust official or beneficiary,

(f) a power to appoint or remove any investment manager or investment adviser or any other professional person acting in relation to the affairs of the trust or holding any trust property,

(g) a power to change the proper law of the trust or the forum for the administration of the trust,

(h) a power to restrict the exercise of any function of a trustee by requiring that it may only be exercised with the consent of the settlor or any other person identified in the terms of the trust,

(i) a beneficial interest in the trust property.”

As you can see the list is extensive but it is rare for clients to reserve all of these powers as this can lead to complications in the event of a dispute, as is examined in this article. Where a client retains too much control and influence it calls into question their intention to create the trust for the benefit of others and also to what extent they could recall those assets by various means set out in (a)–(i), above. It is the job of the legal adviser to explain why and what issues can arise from retaining too much control or influence over the trust.

It is important to note that s.15(2) of the Guernsey Trust Law confirms that anyone who holds those powers is not a trustee of the trust and holding that power does not impose a fiduciary duty upon the holder.11 It is also relevant that any trustee who acts in compliance of the terms of a trust where powers are reserved or granted will not be acting in breach of trust, as a matter of Guernsey law.

It is clear therefore that the settlor could retain a benefit or control as a beneficiary, trustee or third party (such as a protector). They can also retain the power to revoke the trust12 or reserve a life interest in the assets. All of these elements could put the planning at risk, depending on the jurisdictions involved of course. An exception to the rule can be found in Cayman where it is common (and perfectly legal) for trusts to be established purely for probate avoidance. In such cases this is not an issue but, when reserving powers in the context of asset protection, in the broader sense, then this could well put that planning at risk.

How does this compare to other jurisdictions?

Guernsey is not the only jurisdiction which permits reserved power trusts. The BVI was the first jurisdiction13 to introduce statutory provisions expressly permitting powers to be reserved, however, they are also expressly permitted to varying degrees in Jersey, the Cayman Islands, Hong Kong, the Bahamas, Bermuda and Singapore. In Hong Kong and Singapore the powers the settlor is able to reserve are more limited but they all result in the same debate as to the settlor’s exposure and risk.

Is there a difference between reserving powers to the settlor and granting powers to others? How far is too far?

Powers can be reserved in varying degrees. It is possible to have powers to act, give direction, veto or consent.

The most robust position, where one of the purposes of the trust is asset protection is for an independent professional corporate trustee to have full discretion to exercise all trust powers and for them to have documented how and when they consider exercising their powers as well as where they have historically said no to requests from the settlor or other beneficiaries. Any reserved powers should be expressly stated as such to avoid any accusations of obfuscation.

Where there is to be a power conferred on someone other than the trustee, this would ideally be granted in favour of a third-party independent fiduciary, who again documents historic instances of turning down requests that do not suit the motivation and objectives underlying the trust.14

Where powers are reserved to a protector who simply complies with a settlor’s request, is a close friend of the settlor, or in the worst-case scenario to the settlor themselves, then this is the weakest position to be in with regard to any asset protection objectives.

If the protector simply follows orders and never acts contrary to the settlor, or does not fully consider its duties and how to exercise its power of either: acting, giving a direction or providing a veto or consent, then this could suggest that the protector, if a fiduciary, is acting in breach of trust and further weakens the trust. A settlor may well retain wholly beneficial powers, in which case any suggestion of a sham arrangement would not arise, although the asset protection benefits would be reduced. Further, depending on the nature of those powers, it may mean that there is no substantive trust at least on the basis of the judgments in Pugachev and Webb, although of course those judgments have been heavily criticised.15

What issues can arise?

A trust can be at risk from attack from multiple angles. There is always the risk that either the trust instrument will be held, on its true construction, not to have created a substantive trust (such that, in reality, the settlor remains the sole beneficial owner of the assets) or it does create a substantive trust on its face, but is in fact a sham as the settlor and trustee never intended for it to take effect according to its terms. In either case, the arrangement will not satisfy the first of the three certainties, being the intention to create a trust for the benefit of others. If, upon review, a third party would question if the settlor has given up sufficient control over the assets (despite there being a trust) and that the reality is the beneficial interest remains with the settlor then it is likely that in the event of a creditor claim that trust could be voidable.

It is difficult to prove an intention to create a sham arrangement, however, if the trustees have historically shown themselves to be dependent upon the settlor’s views and in reality act merely as a banking portal for the settlor then the case would be easier to prove. A trust can only be a sham from the start though, so the standard is high and must be met from the outset.

As we commonly see in the global divorce courts, trusts and other structures are under increasing levels of scrutiny. Assets that are given to a trustee where no powers are reserved to the settlor and where that settlor has no way of benefiting may be considered outside the scope of the family court, but this has not stopped the English family courts from purporting to vary a foreign law nuptial settlement or proceeding on the basis that the trust is [not] a resource of the settlor which should be taken into account in making financial awards. In these cases historic patterns of benefit and influence are always relevant.

The more channels of influence and control that a settlor retains (either directly to themselves or granted to something they have a strong influence over) the greater the risk that the court will look through the planning and treat at least a portion of those trust assets as being part of the matrimonial asset pool; at least to the extent that the court feels the settlor could request and would be granted a distribution after the decree absolute was granted.

It is important to look at two key cases in this area in more detail.


The judge in Pugachev16 explained that he did not find the reintroduced concept of an “illusory” trust17 as being a helpful one and stated that it is one we should move away from.

Between 2011 and 2013 Mr Pugachev, in consultation with his family office, settled in the region of $95 million of assets on five discretionary trusts which were subject to New Zealand law. The trust assets included a London property worth £12 million and a St Barths property worth $40 million.

The New Zealand solicitor and his wife were shareholders in the trust companies and were directors, along with a couple of Mr Pugachev’s colleagues. The beneficial class of each of the trusts was well defined and comprised of Mr Pugachev and his family members. On face value the trustees appeared to be independent.

A significant number of powers were reserved to the protector (Mr Pugachev), whose consent was required before the trustee could exercise several powers of the trust such as making distributions of income and capital or making investments, varying the terms of the trusts, making changes to the beneficial class or terminating the trust. The protector was also given the power to change the trustees of the trusts.

In 2015 claimants were successful in obtaining a Russian judgment against Mr Pugachev for $1 billion and they brought enforcement proceedings in England. The claimants brought three claims against the trusts:

  1. “The true effect of the trust claim”: Mr Pugachev had not properly divested himself of his beneficial ownership in the assets as he was the settlor, protector (with extensive control powers)18 and beneficiary;
  2. “The sham trust claim”: Mr Pugachev intended to retain control and the trust deeds were a “sham” and therefore had no effect; and if those claims were unsuccessful19;
  3. “The section 423 claim”: If the trusts were a successful divesting of his control in the assets then this was done to prejudice his creditors who included the claimants.20


The “true effect of the trust claim” was successful. In considering the “sham trust claim” the court took into account Re Esteem Settlement21 and highlighted the following points of interest:

  1. A finding of sham, where there was no intention by the parties to create a trust and it was intended from the outset for the settlor to retain total beneficial interest in the trust assets, involves careful consideration of the facts, including external evidence. An artificial arrangement is not necessarily a sham and nor does it automatically result in a sham if the parties depart from the original agreement.
  2. The unilateral intentions of the settlor alone are not sufficient to declare a trust a sham; there must be common intention by both the settlor and the trustee and in this case, reckless indifference is enough to indicate a common intention.
  3. A trust must be a sham from the start, it cannot become one over the course of time.

The court held that when considering the “true effect of the trust” claim it was misleading to refer to the arrangement as illusory; Mr Pugachev had retained control through his position as protector and he had no intention of ceding control. Birss J concluded that

“… whatever label is to be applied in this case, in my judgment the combination of circumstances here mean that the court should not give an effect to these instruments that would result in the assets being regarded as outside of Mr Pugachev’s ultimate control.”22

As the court supported the first and second claims they did not consider the third, alternative, “section 423 claim” in any great detail.

A key factor in the court’s decision was the level of control retained by Mr Pugachev, which allowed him to control the exercise of powers in his favour. As a result the assets were held by the trustees upon resulting bare trusts for Mr Pugachev, as settlor, and therefore were available to settle the creditor claims made against him. It did not help that Mr Pugachev had been accused of numerous allegations of contempt of court, had been shown to have lied to the court and could not be shown to be trusted, having previously breached the court’s orders.

Had Mr Pugachev been advised to reduce the level of his control, appointed an independent protector to act and been able to demonstrate that the trust was independently and well administered for the whole beneficial class he would have arguably been more protected from these claims.

Webb v Webb

It is interesting that the Webb judgment does not rely upon Pugachev or refer to it in any way despite only being handed down a few years later. In the case of Webb the settlor sought to persuade the court that two trusts known as “The Arorangi Trust” and “The Webb Family Trust” should not form part of the marital pot following the dissolution of his marriage. This case was heard in the Cook Islands and the court there found that

“… the two deeds of trust fail to record an effective alienation of the beneficial interest in the assets in question. The powers retained by [Mr Webb] meant that at any time he could have recovered, and still could recover, the property which he had purported to settle on the trusts.”23

This decision was upheld by the Privy Council due to the fact Mr Webb was trustee, a beneficiary, the “consultant” (whose consent was required to change the trustee) and also contained a provision allowing him to nominate himself as the sole beneficiary. As a result the Privy Council held that the trust deeds failed to record an effective alienation by Mr Webb of the property held in trust.

The impact of this finding in Webb was further complicated by a debt which potentially impacted upon the pool of matrimonial assets. The court applied the principles set out in Livingstone v Livingstone24 to say that a debt can only be factored into the matrimonial pot if it is likely to be paid or recovered (due to resistance by the debtor and/or some jurisdictions not enforcing foreign revenue laws, for example).25

They also distinguished between personal and matrimonial debts.

Again, had Mr Webb been advised to pass the role of consultant to an independent third party and further not to retain ultimate control to nominate himself as sole beneficiary he would have been offered greater protection.

Final thoughts

It is important to remember that a power in itself is not property although certain powers might be treated in the same way as property.26 However, it is not always as clear-cut as that if the reserved power allows the settlor to retain beneficial ownership of the trust assets. If they do retain power then it is important to demonstrate clear and independent administration of the trust and that no beneficiary rights are excluded, such as the right to information.

While statute permits the legal reservation of significant powers when settling a trust it is always recommended that detailed and considered thought needs to be given to whether those powers should be reserved or granted to a third party who is ideally independent. You need to ask yourself if those assets are in reality held to the direction of the settlor and if the answer to that question is yes, then if the trust is established in that format the settlor and the trust they wish to settle will remain exposed.

The settlor, protector and trustee, as well as any reserved powers should always be considered carefully in light of the matrix of facts surrounding the settlor. This might include consideration of which jurisdictions are involved, what the settlor’s source of wealth is and any areas of risk to the structure arising as a result of those facts and the family’s circumstances.

If the settlor can reserve a power controlling the direction in which the assets flow or controlling how they are held and by whom, or who can enjoy them, then this this offers a potential avenue to challenge the trust, regardless of whether or not the trust was a sham from the start. Likewise, if too many powers are retained and there is no change in the nature of the ownership then the trust is at risk of being called “illusory” and at worst, a sham. As with all things, close attentiveness to avoiding points of danger is ultimately in the settlor’s best interests.


1 @SenSanders on Twitter, 24 January 2014.
2 The two-faced god of dualities.
3 JSC Mezhdunarodniy Promyshlenniy Bank v Pugachev (“Pugachev”) [2017] EWHC 2426 (Ch); (2017–18) 20 I.T.E.L.R. 905.
4 Webb v Webb [2020] UKPC 22; [2021] 1 F.L.R. 448; [2020] W.T.L.R. 1461.
5 Knight v Knight (1840) 3 Beav. 148; 49 E.R. 58; affirmed sub nom. Knight v Boughton (1844) 11 Cl. & Fin. 513; 8 E.R. 1195.
6 Re Shirley (1965) 49 DLR (2d) 474.
7 Armitage v Nurse [1998] Ch. 241; [1997] 3 W.L.R. 1046.
8 The irreducible core is an idea expressed by Millett LJ in the case of Armitage v Nurse [1998] Ch. 241, 254 and suggested that certain duties must be imposed upon trustees and should not be possible of exclusion, so as to ensure the valid existence of the trust.
9 A revocation would not invalidate any actions taken validly by the trustee prior to the revocation.
10 Reserving a power to direct investments may be a fiduciary or non-fiduciary power, but a non-fiduciary power requires “the most direct and express words”: Vestey’s Executors (Lord) v IRC [1949] 1 All E.R. 1108, 1115 per Lord Simonds.
11 Although there is arguably some tension between s.15(2) and s.32(3)(b) of the Guernsey Trust Law, the latter of which provides that powers of consultation or consent are not fiduciary “if the terms of the trust so provide”: see Re K (Judgment 31/2015). This is touched on in Tony Pursall and Matthew Guthrie, Guernsey Trust Law, 1st edn (Oxford: Hart Publishing, 2020), pp.185–186.
12 Tompson v Browne (1835) 3 My. & K. 32; 40 E.R. 13 which is confirmed by Guernsey Trust Law ss.15(1)(a), (k).
13 Trustee Act 1961 s.86, introduced by the Trustee (Amendment) Act 1993.
14 It is important to note that turning down requests is not a requirement for a valid trust, however, it is sensible to demonstrate that the trustees have true and independent control.
15 See, for example, Pursall and Guthrie, Guernsey Trust Law, 1st edn (2020), pp.80–82.
16 Pugachev [2017] EWHC 2426 (Ch). For an earlier exploration of the case see C. Moss, “A Sham Trust, and an Illusory Trust but Not in Those Terms; Pugachev explained” [2018] P.C.B. 126.
17 I.e. a trust which does not quite meet the bar set for a sham arrangement, but it also does not satisfy the first certainty fully.
18 Mr Pugachev was the protector until he was under a disability which included a claim by creditors. His son was nominated as successor protector in those circumstances. The court rejected this attempt to make the trust “judgment proof”. It was further clear from the list of reserved powers that the trustees could take little or no action without his consent and while such reservation was permitted it could not stop another jurisdiction challenging them. This was further compounded by the fact his powers as protector were personal, not fiduciary, in nature.
19 The judge stated that if he was incorrect in regards to the first claim and the protector’s powers were fiduciary powers then the trust was a sham because the trust showed a different intention from those held by the parties when Mr Pugachev settled the assets upon trust.
20 Insolvency Act 1986 s.423.
21 Re Esteem Settlement [2003] J.R.C. 092.
22 Pugachev [2017] EWHC 2426 (Ch) at [442].
23 Webb v Webb [2020] UKPC 22 at [65].
24 Livingstone v Livingstone [1980] 4 M.P.C. 129 (NZHC).
25 The English common law position is that the English court will not entertain the enforcement of foreign revenue laws, which is a principle set down in the case of Government of India v Taylor [1955] A.C. 491, HL; [1955] 2 W.L.R. 303.
26 Tassarruf Mevduati Sigorta Fonu v Merrill Lynch Bank and Trust Co (Cayman) Ltd [2011] UKPC 17; [2012] 1 W.L.R. 1721 the Cayman Islands Grand Court has since confirmed that it has no power to appoint receivers over future distributions that a beneficiary (who is a judgment debtor) may receive from a discretionary trust, as a beneficiary of a discretionary trust has no legal or beneficial interest in the trust assets: Y v R 2018 (1) CILR 1.
Scroll To Top