Some thoughts on...flee clauses

Not as easy as one, two, flee.

I don’t propose to dwell too much on further discussion about the future of Hong Kong. My thoughts remain positive as per my recent article[1]. However, there are many that remain uncertain. One consequence of this has been an increased number of queries about the pros and cons of adding a “flee clause” to Hong Kong trusts to mitigate against perceived risks in Hong Kong. The enquiries mostly relate to trusts with a Hong Kong trustee and governed by Hong Kong law but also the increasing number of trusts with a Hong Kong trustee and a foreign governing law[2].

My view has been, and remains, that flee clauses can be a dangerous addition to a trust and should mostly be avoided. In addition, it is difficult to give any guarantee that they will work as intended. Some more specific thoughts are set out below.

Before so doing, it is useful to set out some thoughts about risk management. Flee clauses are seen as part of a risk management strategy. In my mind, for the sophisticated client, more suitable risk management strategies should revolve around some or all of the following:

·       Use of multiple jurisdictions both for structures and asset holding,

·       Use of multiple types of jurisdictions such as avoiding only using British Overseas Territories or only Crown Dependencies,

·       Use of different service providers,

·       Use of different financial institutions,

·       Different family members with different residences and citizenships,

·       Use of insurance against political risk / asset appropriation[3],

·       Use of bilateral investment treaties[4],

·       Use of different types of structures,

·       Diversification of asset type and location,

·       Use of a regulated family vehicle to centralise compliance information.

If there is a real concern about a specific jurisdiction (over and above others) consideration should be given to using alternatives.

Furthermore, most trusts have a power to change trustee and/ or governing law in the course of the trust’s administration without recourse to an automatic flee clause. Many Hong Kong trusts went through such a change in the period prior to 1 July 1997 and, once the perceived handover risk had dissipated, many “returned”.

So what about a flee clause? A flee clause is a provision in a trust designed to enable a change of trustee and, usually governing law, due to the occurrence of some major event (“Trigger Event”) happening in the jurisdiction of the trustee.

Usual Trigger Events are related to matters such as:

(i)              invasion of the territory where the trustee is located,

(ii)             severe disruption relating to a major weather event, natural disaster or pandemic,

(iii)           significant changes to the internal governance of a territory (new tax laws, asset forfeiture, coups d’état, insurrection etc.) including declaration of a state of emergency[5]; and

(iv)           (less common in my experience) claims by creditors against a trust(ee).

In most cases the Trigger Event automatically effects the change of trustee and /or governing law; in others there is a discretion of the trustee or protector. In most cases, the change is permanent.

There are some immediate issues which spring to mind. The last one relating to the trust assets is the most important.

1.     Clarity of drafting.

It is crucial to know whether or not the Trigger Event has occurred. For example -what is meant by a “catastrophic weather event” or “insurrection”? There is for example often debate in insurance circles as to the strength of a storm.

2.     Is a flee clause actually needed for the event in question?

Due to regulatory oversight requiring disaster recovery protocols, system backups in the “cloud”, remote access, improved building standards and multiple offices around the globe, the impact of severe weather (whilst a major concern on the ground), does not have the same potential disruptive impact on trusteeships as it had, say, 20 years ago and does not warrant as much consideration via a flee clause[6].

The Cold War is over, so the thought of, say, Cuba invading the Cayman Islands has long passed!

A real risk of a Corbyn-government disrupting the British Overseas Territories has gone.

For those dwelling on Hong Kong, Hong Kong is perhaps the least likely trust jurisdiction to be invaded by a foreign power!

Flee clauses should really only concentrate on events that can be clearly defined and which would impact on the operation of the trust – such as currency controls, asset appropriation, taxation etc.

3.     The Trigger Event occurring when the change is not needed.

Usual examples are the assassination of the Governor of Bermuda in 1973 and the Bahamas case in 2020[7].

4.     The Trigger Event should be automatic.

If a powerholder has to step in to activate the flee clause, there is always risk of delay, of injunction or of the exercise of the power being set aside. Since a change of trustee could result in expense and tax liabilities, this is why, in my view, the Trigger Event should be a true emergency. 

5.     The role of the “Emergency Trustee”.

Since the switch is meant to be immediate, it is imperative that the Emergency Trustee is onboard from day one. In addition to the trust documents, the Emergency Trustee will need real time access (unaffected by any Trigger Events) to KYC/AML documents, source of funds information, annual reviews, asset holdings etc. The Trust documents should expressly permit the sharing of such information with an external party.

The Emergency Trustee’s compliance/risk/legal team must have pre-approved the change. There can be no delay for onboarding. In that regard the Emergency Trustee will invariably be part of the same group as the Trustee. A family-owned PTC is an option but far more expensive and far less likely to be ready at short notice.

The location of the Emergency Trustee is also an issue. Such a location should ideally be insulated from as many of the same Trigger Events as possible, else frying pans and fires spring to mind.

6.     Public policy risks

Trigger Events linked to tax changes, transparency requirements and the like, will be subject to more scrutiny than say, Trigger Events designed to mitigate against hurricanes or typhoons. The regulatory environment concerning trusts, the jurisdictions from which many trustees operate and the risk appetite of the trust companies themselves has changed over recent years.

The FATF stated in 2005[8], that “flee clauses can constitute an obstacle to an effective anti-money laundering framework, in particular in terms of international legal assistance.”

The OECD[9] issued the following in 2002: “any flee clauses that require a trust and information about a trust to be moved to a different jurisdiction upon receipt of service of process or inquiry by the authorities only serve to encourage and protect illicit behaviour. Jurisdictions should consider reviewing the use of flee clauses to evade service of process or inquiry by the authorities with a view to either abolish them or severely limit their availability.”

Take the following scenario. Jurisdiction A introduces a tax on trusts in Jurisdiction A at 00:00 on 1 January. A scintilla of time later, due a flee clause, a trust in Jurisdiction A has “moved” to Jurisdiction B via a change of trustee and governing law. Could there be an argument that there has been some form of tax evasion[10]? Are there any issues for the Trustee in Jurisdiction B – even if reputational? Could the Trustee in Jurisdiction A remain liable (but perhaps without access to the trust fund)?

Could Jurisdiction A take the OECD’s lead and first declare that any flee clauses were void on public policy grounds if they were to be triggered by fiscal or compliance related events? Would an attempt to abolish flee clauses also trigger some flee clauses?!

Standard fiduciary powers to change trustee and governing law would not be caught but trustees would need to act quickly.

7.     The assets

This is the main issue with flee clauses – the need to “automatically” move the assets from the trustee to the Emergency Trustee in a manner which is recognized and accepted by all relevant parties.

Clearly, the assets cannot be in the same jurisdiction as that of the trustee. This does reduce some of the asset protection benefits of, say, settling Cayman Islands shares onto a Cayman Islands law governed trust with a Cayman Islands trustee.

The transfer of assets needs to have been pre-approved. No-one wants to face analysis of powers of attorney, third parties requesting additional documents and/or wet signatures from the original trustee etc. before the trust assets can be dealt with. 

Since most trustees do not hold assets directly but via underlying limited liability structures, it also makes sense if the asset holding is done by vehicles controlled by members of the same group as the trustee and Emergency Trustee. This again can deal with compliance and onboarding issues – powers of attorney can be agreed in advance and there is certainty that company registers will be updated.

We therefore need to deal with transfer of the ownership of the holding entity in a manner which will allow operation of such holding entity’s bank and financial accounts and satisfy the compliance obligations of the financial institution. The common use of nominee shareholders also needs to be taken into account.

In terms of effecting a switch automatically and with certainty, other than powers of attorney and share transfer forms, a degree of bespoke drafting may be needed. If we use two classes of shares with different rights depending on whether or not a Trigger Event has occurred, it is still necessary to issue one class to the Emergency Trustee in advance and have a declaration of trust.

It is also possible, for example, to have a Cayman Islands LLC where the occurrence of a Trigger Event (mirroring the trust) “switches” the ownership from the trustee to the Emergency Trustee[11].

Another possibility could be to hold assets via a PPLI policy (so the asset owner is the insurance company) and the terms of the policy reflect the Trigger Event perhaps via an arbiter provision.

Bahamas and COVID

To conclude it is worth highlighting the Bahamas[12]. The article by Rhyan Elliott of Higgs & Johnson is worth a read. In summary, the Bahamas declared a “state of emergency” due to COVID on 17 March 2020[13] using the Emergency Powers Act. On 19 March 2020, the Government amended the regulations by adding an extra regulation 24:

“Wherever there is a provision (a “flee clause”) in a trust instrument governed by Bahamian law to the effect that the trust in question shall, in the event of a proclamation of a state of emergency in The Bahamas, cease to have —

(a) Bahamian law as its governing law;

(b) The Bahamas as its forum and situs of administration; or

(c) as its trustee a trustee in The Bahamas,

such trust instrument shall be treated (despite any provision in the trust instrument that might be considered a contrary provision) as having from its date contained immediately thereafter the following provision —

“Provided always that the flee clause does not extend to the proclamation of a state of emergency as a result of a pandemic; and provided further that any actions carried out without taking account of such proviso shall be null and void from the outset.”

This seeks to disapply the activation of the flee clause but after the fact. It is an interesting debate as to the effect of this provision on a trust with an automatic flee clause ( so no exercise of a fiduciary power to set aside). On 17 March, such a trust would have left the Bahamas, a new trustee taken over and a new governing law in effect. Assets may have moved across and trustee discretions been exercised. Two days later, the Bahamas sought to disapply all of this. It would be interesting to know how many such trusts were caught up in this.

[1] https://www.ifcreview.com/articles/2021/february/the-future-of-hong-kong-as-an-international-financial-centre/

[2] The reasons for this are outside the scope of this article but happy to elaborate further.

[3] https://www.dhirubhai.net/pulse/trustees-how-safe-your-trust-fund-richard-grasby-tep

[4] https://www.dhirubhai.net/pulse/trustees-how-safe-your-trust-fund-richard-grasby-tep

[5] See Bahamas example below.

[6] See also John Gibbons – “Flee clauses: a fresh look in a modern context”. Trusts & Trustees Vol 23 No 10 December 2017 pp1028-29.

[7] See for example: https://www.macfarlanes.com/what-we-think/in-depth/2020/fight-or-flight-flee-clauses-and-offshore-trusts-in-the-time-of-covid-19/

[8] https://www.fatf-gafi.org/media/fatf/documents/reports/Misuse%20of%20Corporate%20Vehicles%20including%20Trusts%20and%20Company%20Services%20Providers.pdf

[9] https://www.oecd.org/corporate/ca/1961539.pdf

[10] On which see K Foo. “Does tax evasion generate criminal proceeds?” Singapore Academy of Law Journal. (2019) 31 SAcLJ pp845-874.

[11] See for example s10(3) and s13(1) of the Limited Liability Companies Act (2021 Revision)

[12] https://higgsjohnson.com/the-bahamas-trustees-pursuing-protection-in-uncertain-times/

[13] https://www.bfsb-bahamas.com/wp-content/uploads/2020/05/EmergencyPowers-Covid19-Regulations2020-20May2020.pdf

? 2021 RDG Fiduciary Services Limited