Tax treaty tie-breaker: The Upper Tribunal's decision in Haworth v HMRC

The Upper Tribunal in Haworth v HMRC [2024] UKUT 00058 (TCC) has found that the First Tier Tribunal applied the correct test for the “place of effective management” of a corporate trustee and confirmed that the test for place of effective management is not the same as the “central management and control” test applied to determine the residence of a company.

See our article on the First Tier Tribunal decision for more detail.

The scheme

The Jersey trustees of a family trust held shares in a UK-incorporated company. They participated in a “round the world” scheme, which was devised in the UK and was designed to avoid the trustees incurring UK capital gains tax on a share disposal.

The existing Jersey trustees resigned in favour of Mauritian resident trustees, who disposed of the shares. The Mauritian resident trustees then resigned in favour of UK resident trustees. The Mauritian resident trustees would not incur capital gains tax on the disposal of the shares.

The scheme took advantage of the UK-Mauritius Double Tax Treaty. Article 4(3) of the Treaty provided a tie-breaker test for determining the treaty residence of a person who was liable to tax in both contracting states. It stated that a person "shall be deemed to be a resident of the contracting state in which its place of effective management is situated." If the “place of effective management” was in the UK, then the trust had to pay capital gains tax on the disposal of the shares; if it was in Mauritius, they did not.

The place of effective management test

The place of effective management test was derived from HMRC v Smallwood [2010] EWCA Civ 778, which concerned the same “round the world scheme”. In Smallwood, the judges disagreed on how to apply test.

• Giving the leading judgment, Hughes LJ held that the place of effective management should be interpreted as the centre of top-level management, i.e. where the key management and commercial decisions of the trustees “as a continuing body” are actually made, and this could be distinct from the day-to-day management of the trustees in office at the time. This meant consideration could be paid to any overarching scheme in place for the management of the trust.
• Dissenting, Patten LJ said that the test should be applied to a corporate trustee in accordance with the central management and control test provided for in Wood v Holden [2006] EWCA Civ 26. The court should consider where the location of central management and control was at the time of the disposal of the shares, that is, where binding decisions are made by its constitutional organs (e.g. the board of directors), unless the decision-making is usurped.

Adding to the confusion were the comments of Chadwick LJ in Wood v Holden that, on the facts of that case, the test for identifying the location of central management and control of a company is in substance the same test as the place of effective management test.

Haworth: The First Tier Tribunal

In the First Tier Tribunal, Judge Morgan questioned Hughes LJ’s reasoning in Smallwood. She seemed to agree with the view of Patten LJ, stating that she found it difficult to understand why it is not appropriate to seek to assess the place of effective management of a trust by adopting a similar approach to that taken in Wood v Holden.

However, she conceded that she was bound by the majority decision of the Court of Appeal in Smallwood. She found that the trust was effectively managed in the UK at the relevant times by reference to the place of the “top-level management” – the UK settlors of the trust and their UK advisors had “devised, decided upon, facilitated, orchestrated and superintended” the scheme “on an on-going basis throughout the relevant period”.

Haworth: The Upper Tribunal

The appeal in the Upper Tribunal turned solely on the interpretation of the place of effective management. The appellants argued that the First Tier Tribunal erred in holding that the Court of Appeal’s decision in Smallwood was applying a different test from that in Wood v Holden, and, as such, it should have applied that test to identify the location of the central management and control of the trustees. The Mauritian trustees had made the relevant decision to dispose of the shares, so no capital gains tax would arise. The fact that certain decisions had been made in the UK was not sufficient to conclude that the decisions of the Mauritian trustees had been usurped.

The Upper Tribunal considered in detail the reasoning in Smallwood and the meaning of the place of effective management. It concluded that there was no consideration of whether the decision-making of the trustees had been usurped in Smallwood, and that “the majority of the Court of Appeal had endorsed a test for the place of effective management which involved looking at the circumstances in which the scheme was devised and implemented” (emphasis added). It was not necessary to apply Wood v Holden and consider the test for central management and control, and the comments of Chadwick LJ in that case had not formed part of the decision.

The Upper Tribunal found that the First Tier Tribunal, therefore, had applied the correct test for the place of effective management, identifying “in which state the real top-level management (or the realistic, positive management) of the trustee qua trustee is found”. The binding authority from the Court of Appeal in Smallwood required it to do so.

Why is this important?

This decision re-affirms that the interpretation of the “place of effective management” as a tie-breaker in double tax treaties is not the same as the test for “central management and control” to determine corporate residence. In fact, it is potentially much wider. It will not be sufficient to show that the current offshore trustees actually took the relevant decisions (even on advice), rather, it will be necessary to consider all of the circumstances leading up to those decisions being taken as well, considering the trustees as a continuous body rather than as individuals from time to time.

It remains to be seen whether the case will be appealed to the Court of Appeal, who will be able to depart from the decision in Smallwood if they see fit.

Trainee solicitor, Sophie Greenstreet, is a co-author on this article.