Garron: Determining Residence of Trusts
Offshore trusts are major tax-planning and estate-planning tools which can be used to protect assets from taxation and to defer accrued capital gains for a long term. For all such planning tools to work, however, the “offshore” aspect of the trust—that is to say its non-Canadian residence—has to be a central feature of the structure. A trust resident in Canada would be subject to Canadian income tax on its worldwide income.
The established common law principle for determining the residence of trusts is that a trust is resident in the jurisdiction where its trustees reside and operate (Trustees of the Thibodeau Family Trust v. The Queen,  78 DTC 6376 [Thibodeau]). In Thibodeau, two of the trustees were resident in Bermuda and the third trustee was resident in Canada. The Federal Court held that the trust was resident in Bermuda because the majority of the trustees resided in Bermuda and the trust document allowed for majority decision-making.
The Tax Court of Canada’s decision in Garron Family Trust v. Her Majesty The Queen, 2009 TCC 450 [Garron], released September 10, 2009 abandons this established approach in favour of the “central management and control” test used to determine the residence of corporations (De Beers Consolidated Mines, Limited v. Howe, (1906) A.C. 455). Garron holds that the residence of a trust is determined by the jurisdiction where the central management and control of the trust resides regardless of the residence of the trustee.
Garron concerned the residence of two trusts formed in Barbados. The settlor for the trusts was a resident of St. Vincent Islands, the trustee was a holding company resident in Barbados, and the trust beneficiaries were Canadian residents.
The beneficiaries initially owned a Canadian Controlled Private Corporation called PMPL Holdings Inc., which in turn had full ownership of the subsidiaries Progressive Moulded Products Inc. and Progressive Tools Limited. These two subsidiaries were in the business of manufacturing and assembling automobile parts.
In 1998, a corporate reorganization similar to an estate freeze was carried out on the capital of PMPL Holdings Inc. The common shares held by the original owners were converted into fixed-value preferred shares and new common shares were issued to two newly-established holding companies, 1287325 Ontario Ltd. and 1287333 Ontario Ltd. The shares of these new holding companies were then issued to the Barbados trusts. The result was that any accrual in paid-up capital flowed out to the offshore trusts.
In 2000, shares for the two holding companies, 1287325 and 1287333 Ontario Ltd. (owned by the two Barbados trusts), were sold in an arm’s-length sale to Oak Hill Capital Partners LP. As a result of the sale, the two trusts realized capital gains of $450 million on which Canadian tax was paid due to a withholding mechanism. The trusts subsequently claimed tax refunds on these capital gains based on Article XIV(4) of the Agreement Between Canada and Barbados for the Avoidance of Double Taxation and the Prevention of Fiscal Evasion with Respect to Taxes on Income and Capital. Under Article XIV(4), capital gains may only be taxed in the jurisdiction where the taxpayer is resident. The issue before the Tax Court of Canada was the residence of the two Barbados trusts.
The Central Management and Control Test
Neither the appellant taxpayers nor the Minister of National Revenue disputed the fact that the trustee for both trusts was a resident of Barbados and not resident in Canada. Relying on Thibodeau, the appellants argued that the residence of the trust is determined by the residence of the trustee. However, the Minister argued for the use of the central management and control test in determining residence of the trusts. The Minister submitted that throughout the period in question, the beneficiaries were in control of the trust; the offshore trustee had played merely a subordinate role.
Madam Justice Woods sided with the Minister, holding that the central management and control test was the appropriate test for determining trust residence. Justice Woods found that Thibodeau‘s reliance on the residence of trustee alone could be restricted to the unique facts of that case. She further held that Thibodeau‘s rejection of the central management and control test was erroneous: “in my view the Thibodeau decision does not form a solid foundation for rejecting the Minister’s position that residence should be determined by a central management and control test” (at para. 151).
According to Justice Woods, Thibodeau relied on the assumption that trustees will always comply with their fiduciary obligations. Since the trustee has a fiduciary obligation to manage the trust for the benefit of the beneficiaries, the residence of the trustee is sufficient to determine the residence of the trust. In her view, this assumption was erroneous and key to the rejection of the central management and control test in Thibodeau.
The Garron judgment falls short in providing convincing rationale for moving away from the Thibodeau approach to trust residency. Madam Justice Woods approaches the decision as developing a test for trust residence rather than following or moving away from established jurisprudence. The rationale for provided by Justice Woods is that “adopting a similar test of residence for trusts and corporations promotes the important principles of consistency, predictability and fairness in the application of tax law” (at para. 160). The judgment lacks a discussion of the legal and structural differences between corporations and trusts and why a central management and control test may be necessary for determining corporate residence but not necessary for determining trust residence.
A better approach may have been to carve out an exception from Thibodeau to deal with the facts of this case: situations where there is evidence that that the trustee is not fulfilling its fiduciary duties and legal obligations towards the trust and the beneficiaries of the trust. In such cases, residence could be found in jurisdictions other than the jurisdiction where the trustee is resident because of the limited or subordinated role of the trustee.
Applying the new approach to the facts of the case, Justice Woods held that there was a lack of evidence to show that the trustee actually played a role in the management of the trust assets and investment portfolio. The trustee was selected by the beneficiaries to deal with the administrative aspects of the trusts and had no effective decision-making authority beyond this.
Another factor that weighed against the appellants in Garron was that the trust indentures provided for the appointment of a protector who could replace the trustee. On its own, this would not be a problem (like the trustee, the protector in this case was also not a Canadian resident). However, the trust indentures further provided that a majority of the beneficiaries could replace the trustee and retain control of the trusts. For Justice Wood, the subordination of the trustee was apparent and “effectively enforceable” through this mechanism.
Implications for Offshore Trusts
Despite the drastic shift in law resulting from Garron, the management and control test may not be as arduous to get around as it would seem on first glance. The UK case Wood v. Holden,  EWCA Civ 26 [Wood], discussed in the Garron judgment is indicative of this.
Wood also concerned determining an off-share entity’s residence for purposes of calculating capital gains tax. A London financial services firm set up a holding company, Eulalia, in the Netherlands. The arrangement produced favourable tax consequences for Mr. and Mrs. Woods, who were residents of the UK and who were disposing their family business. However, the sole managing director of the holding company was another corporation in the Netherlands.
The Court of Appeal held that Eulalia was not a resident in the UK despite the involvement of UK residents in its set-up and its affairs. The court found that:
- i) the London firm “did not dictate any decisions” that the director was to make, and the directors of Eulalia “were not by-passed nor did they stand aside”; and
- ii) the managing director exercised its duty as managing director by signing and executing relevant documents.
The court also noted that “[t]he documents [adduced as evidence] showed guidance and influence coming from [the London firm], but no more than that” (at para.31).
In Garron, Justice Woods relied on the lack of evidence adduced by the appellants to distinguish the application of the management and control test from Wood: “[i]n contrast [to Wood], the appellants led very little evidence as to the formation and operation of the Trusts. In these circumstances, there is no basis for concluding that St. Michael [the trustee] did not agree to assume a limited role in the management of the Trusts” (at para. 262).
The management and control test in the offshore trust context could be satisfied when the trustee exercises its independent authority over the trust, such as affecting changes in the trust’s investment portfolio and management of the trust’s property. The trustee could do this even with guidance and influence from a Canadian resident beneficiary or an investment advisor, as long as the trustee’s authority is not bypassed or undermined. Avoiding a set-up where the trustee can be ousted at any time by a Canadian resident protector or beneficiary would further bolster the argument for vesting central management and control in the offshore trustee.
Nonetheless, if the hurdle is not too difficult to overcome, it is a higher hurdle than the Thibodeau approach of determining residence based on the residence of the trustee. As of yet, Garron has not been appealed to the Federal Court of Appeal; the taxpayers have 30 days from the release of the decision to file an appeal. Those of us on the sidelines watch with interest for further developments on this issue.