Sham Loans

Sham Loans

The Commissioner of Taxation in two recent tax cases (Millar[1] and Normandy Finance[2]) and Liquidators in BCI Finances[3] (supported by the Commissioner of Taxation), have had success in alleging certain loans were shams. This had the effect that in the tax cases, that the taxpayer failed to discharge its onus of proof that the assessments were excessive. In the case of the Liquidators asserting breach of fiduciary duties by directors of companies whose only creditor was the Tax Office, that certain loans were in fact disguised repatriation of undeclared income, and that interest paid on the sham loans was not deductable. The fact that some of the directors and a “shadow” director had put the companies into a position not in the best interest of the companies, meant that some of the directors and the “shadow” director failed in their fiduciary duties.

Millar

In Millar, the taxpayer was a borrower from Hua Wang Bank in Western Samoa, and their Australian superannuation fund had made a deposit with that bank.  The Commissioner applied the sham doctrine so as to assess the taxpayers on effective early access to their superannuation balance, which access was assessable income.

In relation to Millar, the High Court refused the taxpayer special leave to appeal on 16 November 2016, on the basis that the result in the Full Federal Court was not sufficiently uncertain.  In the Full Federal Court there was a division between Logan J, who concluded that there was no sham in that case, and majority of Pagone and Davies JJ who concluded that the original finding of sham was not an error of law.  However, only Logan J referred to the High Court decision in Equuscorp Pty Ltd v Glengallan Investments Pty Ltd[4] and quoted therefrom, sham “is an expression which has a well understood legal meaning.  It refers to steps which take the form of a legally effective transaction but which the parties’ intent should not have the apparent, or any, legal consequences”.  Sham requires a common intent of the parties: Snook[5].

Equuscorp

In Equuscorp, the assignees of a debt originally owed by Investors in a “tax effective” investment scheme, were held to have enforceable rights against the borrowers as the loans to them were not shams notwithstanding that the lender had “no money to lend”.  The High Court in that case relied on the fact that the investors borrowing the money accepted that they had been loaned the money because it discharged their obligation to the manager of the scheme, to provide the funds required.  This was notwithstanding that, in the end, the investors did not obtain their tax deduction for their investment as promoted by the scheme, presumably due to the application of the general anti-avoidance provision, Part IVA.

Lender’s Rights

It is to be remembered that a finding of sham would render the lender’s rights in the loan document unenforceable as such (at least in this jurisdiction), not just undermine the characterisation of the transaction for the taxpayers/borrower. In Millar and BCI Finances, there was no direct evidence of the lenders’ intention, but only an inference drawn as to the lenders’ complicit purpose.

Raftland

All three judges in the Full Federal Court in Millar referred to the later decision of the High Court in Raftland[6].  Raftland was tax case where as always, the onus is on the taxpayer to prove that an assessment is excessive.  What this means is that whereas in ordinary civil litigation, whilst a party making a serious accusation such as sham is not required by the Briginshaw principle[7], to prove their allegation at a level beyond that normally required i.e. beyond the level of on the balance of probabilities[8], but it does require a court to show great care in finding that they are satisfied the matter so proved.  In tax cases on the other hand, if the Commissioner makes a serious allegation such sham, the burden of disproving the Commissioner’s allegation still rests on the taxpayer.  Some might think this would have put some special obligation on the Commissioner as a “model litigant” not to make allegations of sham in circumstances where his anti-avoidance armoury such as the anti-avoidance provision, Part IVA could be applied[9].  However in none of Raftland, Millar nor Normandy Finance did the Commissioner apply Part IVA, but simply relied on making the assertion of sham.  Of course, he was successful in those cases with his sham argument, however, had he not been successful, and failed to run a potentially successful Part IVA argument, he would not have done the Revenue any favours.

Raftland was controversial as the High Court found that the trust resolution in that case was a sham notwithstanding that the tax scheme would only work if the resolution was effective to make the selected beneficiary “presently entitled” to the income of the trust. However, as the scheme also involved changing control of the beneficiary so that the former controllers of the beneficiary would only ever get $250,000, rather than the whole of the trust income, the High Court concluded in effect that the disparity between the legal effect of the resolution and the “fiscal” effect, taking into account all the evidence of various actors’ intentions, was such that the resolution was a sham[10]. In Millar, the full Federal Court was divided as to the significance of the intention of the adviser to the scheme (who was not a party to the documents), compared to the intention of the parties.

Whilst Logan J said in Millar that the intention of the legal adviser in Raftland was part of the reason for the decision in Raftland[11], but he noted in Millar that the finding of the Tribunal as the intention of the taxpayers was the end of the matter, as the accountant/promotor (Mr Vanda Gould) was not that agent for the taxpayers[12]. However, Pagone[13] & Davies JJ held that Mr Gould’s intention in Millar was important, as they said the intention of the legal adviser was in Raftland[14].

Normandy Finance

In Normandy Finance the corporate taxpayers borrowed money from Hua Wang Bank initially without a written agreement.  The Commissioner asserted in the Full Federal Court, that a subsequently drawn written agreement (providing for interest) was a sham and as the taxpayer said, it was the written agreement which governed their relations, the question was whether the written agreement was a sham not whether whatever oral agreement was originally in existence when the funds were advanced, was a sham.

At first instance, Edmonds J held that the original oral loan was not a sham, but that the only evidence as to whether interest was payable on the loan was a subsequently drawn loan agreement, which he held to be a sham, and so the interest was not deductible even though the loan itself was not[15]. The Commissioner asserted that the loan was in fact assessable income, but from the decision of the Full Federal Court is not entirely clear whether the funds lent had their origin in undisclosed income of an Australian resident.

Again, the Full Court split with Logan J finding for the taxpayers[16], and Jagot and Davies JJ finding for the Commissioner. However, the intention of Mr Gould, who was again the accountant/promotor[17] was not relevant, as the principal behind the borrower and lender entities was Mr Gould’s client, Mr Townsing, who understood the detail of the transactions his companies entered into[18].

BCI Finances

As noted above, in BCI Finances the Commissioner was not a party, but supported the Liquidators of companies that were creditors of the Commissioner.  In BCI Finances, certain funds were placed on deposit with Israeli banks from which BCI Finances and related companies made borrowings.  There was evidence that the funds placed on deposit with the Israeli bank were undisclosed income of some of the principals.  The taxpayers went to great length to prevent the Commissioner from becoming aware of the deposits with the Israeli banks.  Gleeson J found that the documents provided by the taxpayers to the ATO “were designed to conceal the totality of the arrangements between those applicants and the Israeli banks (whatever was the totality of those arrangements) and hence the documents were shams”.

However, had the loan from the Israeli banks been on standard terms, for instance simply specifying that the bank could offset any security that it held against the borrower’s liability in the event of the borrower’s default, and there just happened to be such security, is not clear that a mutual shaming intent would be present even if the taxpayer went to great lengths not to tell the Commissioner about the back-to-back nature of the borrowings, and details concerning the security.  It is not clear whether the Commissioner asserted Part IVA against BCI Finances or only sham.

Interest Offset Accounts

It should be readily apparent that not all back-to-back loan arrangements are shams. For instance for some thirty years, the Commissioner has accepted home loan interest offset accounts.  That is, where a borrower of a home loan maintains a deposit account with the bank which is provided as security for default by the borrower, and upon which the interest rate for the borrower is reduced due to the non-payment of interest to the lender under the interest offset arrangement.

Intention to repay a loan

In relation to whether an oral loan or a loan evidenced by a document is a sham, there is some confusion as to whether an mutual intention of the parties that the loan not be repaid is a sufficient finding of fact to not have to make a finding of sham, as on one level, this is said not to be a loan (but a disguise for some other transaction). However, where there is a document which represents to outsiders that the amount is to be repaid, there may need to be a finding of sham[19].

It is also interesting to note the justification for the sham loan findings in cases such as Millar, by the subsequent conduct of the parties, in the context of arguments about sham trust.

SHAM TRUSTS

It is to be remembered that the trust in Rafland was not found to be a sham, only the trustee resolution[20].

It has now become clearer that subsequent conduct of a party in relation to a trust should only be categorised as being in breach of trust if the original intention of the settlor and the trustee was that a trust be formed. So perhaps subsequent inconsistent behaviour in the context of a loan, may be in breach of the loan agreement, rather than evidence of a sham loan? Take for example, a documented loan by a parent to an adult child. Whilst both parties may have intended that the loan be repaid when it was made, it would not be uncommon for the child to not comply with its terms, nor for the parent not to enforce its terms, without having reached any further agreement.

A v A

It was stated by the English High Court in A v A[21] that a trust “which is not initially a sham cannot subsequently become a sham”.[22] The view of Justice Munby was that the only way a properly created trust which is not a sham could become a sham subsequently is if all the beneficiaries were, with the requisite intention, to join together with the trustees for the purpose of that sham.[23]

Reynolds

In comparison, the New Zealand Court of Appeal in Reynolds [24] in finding that the trust in that case was not a sham, noted that a finding that a properly created and executed trust was sham would be an extreme finding because of the effect it has on the interests of beneficiaries.[25]Therefore, caution would need to be exercised when determining if acts or omissions provide compelling evidence at a high level to justify the conclusion that the trust is a sham.[26]

In Reynolds there was an absence of resolutions or minutes, no annual accounts, intermingling of financial arrangements and no record of decisions. This evidence of poor administration was held not to establish an intention that the trust operated as a sham but rather may be evidence of a breach of trust.[27]Further, it now seems clear that in the trust context, the mutual intention must be between the trustee and the settlor[28].  Otherwise the consequences for the “innocent party” trustee would be visited upon them, when they were not a party to the arrangement.

Clearly this is highly important in the case of an alleged sham loan, as if it was not so, and if the shaming intention was only known to the borrower, then the lender might not be able to enforce its loan (at least in this jurisdiction).

Clayton v Clayton

Subsequent to Reynolds, the recent New Zealand Supreme Court decision in Clayton v Clayton [2016] NZSC 29 again declined to find the trust in question was a sham, as the husband settlor who was also the trustee, intended the trust to be valid. However, like in the well-known Australian family law case of Kennon v Spry[29], the husband’s power of appointment was treated as though it was an asset valued the same as the value of the trust itself[30].

Re Esteem Settlement

In the Jersey case of Re Esteem Settlement[31], it was said a sham trust, not only covers a case where the trustee merely went along with the original intention of the settlor, but also if the trustee is reckless by going along with the settlor by not knowing or caring what the trustee is signing, he is taken to have the necessary intention to create a sham.[32]

The court in A v A agreed with the approach above from Re Esteem Settlement that “reckless indifference will be taken to constitute the necessary intention”.[33]

Conclusion

Whilst it is to be hoped that the Commissioner’s recent successes in arguing sham do not become commonplace, lawyers and accountants advising clients may be well advised to ensure their clients are sufficiently aware of what the documentation prepared for them requires them to do in their capacity e.g. as a trustee or director of a trustee company. After all, at a practical level, if the client disregards the legal form, the Commissioner may be inclined to do the same?

If you wish to discuss this article, please contact the writer Robert Gordon, or Anthony Pointon on (03) 961 7707.

 __________________________________________________________________________________________

[1] [2016] FCAFC 94

[2] [2016] FCAFC 180

[3] [2016] FCA 1351

[4] (2004) 218 CLR 471

[5] [1967] 2 QB 786

[6] (2008) 238 CLR 516

[7] (1938) 60 CLR 336

[8] “a court will not find that a transaction is a sham if another inference is at least equally open: Sharrment (1988) 18 FCR 516; there is a “strong and natural presumption against holding a provision or a document a sham”: National Westminster Bank Plc v Jones [2001] BCLC 98 at [59]. A finding of sham would have serious consequences for lawyers involved in the drafting of the documents if they were knowingly involved in the improper purpose of the documents. For a case where a lawyer was found to be liable in advising to phoenix companies, see Somerville (2009) 77 NSWLR 110.

[9] For instance, where there is a loan from a non-resident company to taxpayers in Australia that may be associates of the shareholders in the non-resident company, the Commissioner should pursue s47A and Div. 7A if the loan agreements are not on Div. 7A complying terms. Also, a purported loan may be outstanding for so long, notwithstanding that it is recorded yearly in the debtors accounts as owing, as perhaps not being a loan on the basis that there was never an intention to repay?

[10] Pagone J writing extra-judicially “Sham Trusts Revisited” STEP Hong Kong 9 Oct 2014. It is interesting that his paper didn’t refer to what many would consider to be the starting point in modern times, to the issue of sham trust in Abdel Rahman v. Chase Bank (CI) Trust Company Limited and others, a decision of the Jersey Royal Court of Jersey [1991] JLR 103. The result in Raftland sits uncomfortably with the UK authorities which say that a document will not be a sham if it is intended to have its legal effect, regardless of the tax motives of the parties: see Pagone ibid at pp11-12. That paper also appears to raise for discussion e.g. whether resolutions to make beneficiaries presently entitled but to not pay out the entitled for year after year, may raise the question after Raftland, of whether the resolution is a sham resolution?

[11] At [3]

[12] At [8]

[13] Whose judgment interestingly, did not cite his own paper on the topic. In contrast, in his very recent decision in Full Federal Court in Chevron [2017] FCAFC 62 he dismissed Gordon J’s extra-judicial suggestion that deeply retrospective legislation may be unconstitutional. Whilst there may be different views as to whether a judgment should cite the judge’s own extra-judicial work (as to which an analysis of Kirby J in the High Court might be directed), one view is that it assists understanding whether the judge had a pre-conceived view and therefore a potential bias in determining the issue judicially, although this is probably an issue for the judge to declare his extra-judicial writings to the parties before he hears the case. In any event, even if this has been done, the public may know nothing of it as the parties have not objected to his hearing the case, and it may be that it would be helpful for the reader of a judgment to know such background? After all, Pagone J’s 9 Oct 2014 paper was at least the fourth reincarnation of the same paper, pointing to some deep interest in Raftland, perhaps one could go so far as to say that he was implicitly critical of the High Court’s decision? Of course, extra-judicial writing can be extremely influential on other members of the same court, if not higher up the judicial hierarchy. One is reminded of Gzell J’s multiple articles on pre-CGT treaties and CGT, which found favour in Undershaft [2009] FCA 41 and Virgin Holdings [2008] FCA 1503.

[14] Pagone J at [45]; Davies J at [82]

[15] [2015] FCA 1420 at [112]

[16] Logan J has since written extra-judicially about Edmonds J decision at first instance, “What is the point of having Part IVA? (in the Income Tax Assessment Act 1936), Tax Institute, Nat. Conv. 15 March 2017.

[17] It will be recalled that Hill J’s decision on Part IVA in Consolidated Press (2001) was that the taxpayer’s intention was attributed from the advisers to the scheme, because the parties would not have understood the detail of what they had done. Interestingly, the ATO did not purse that line thereafter, perhaps because Part IVA is an objective test, whereas what the taxpayers thought they were doing, or what the advisers said, is the subjective purpose. Having said that, the subjective purpose of parties in anyway involved in a tax scheme, must for part of the factual matrix against which the objective purpose can be determined.

[18] It is noteworthy, that Logan J did not dispute Edmond J’s analysis at first instance, that Raftland was not dependent on the intention of the legal adviser (at [32]), whereas Jagot & Davies JJ did not comment on Raftland at all.

[19] It is worthy of note that in the Div. 7A context, there is a statutory provision which is somewhat of a “sleeper” at present, but in the writer view, will become a potentially huge problem i.e. s109R allows the Commissioner to disregard a loan repayment in working out whether the minimum loan repayment conditions have been meet, if he forms the view that a re-borrowing was to be used to make a repayment. This is a statutory recognition of the practical possibility that notwithstanding the formal loan agreement, it may be the case the Div. 7A loan may not be able to be repaid in the statutory period (7 years for an unsecured loan).

[20] In Fitzroy Services [2013] FCA 471 at [34] Edmonds J found that the loan itself was not a sham, but the interest on those loans was not deductible as the nexus with the production of income was not proven. Again, in that case, the Commissioner did not assert Part IVA in the alternative to sham.

[21] [2007] EWHC 99.

[22]Ibid at [42].

[23] Ibid at [44].

[24]Official Assignee in Bankruptcy in the Property of Reynolds v Wilson and Harvey and another [2008] NZCA 122.

[25] Ibid at [77]

[26] Ibid

[27] Ibid at [92]

[28] For a survey of the cases see A v A at [32]-[58]

[29] [2008] HCA 56

[30] The Court discussed the lower courts’ findings of “illusory trust” but was inclined to view their finding as being one that the settlor/trustee/husband intended to create a valid trust, but may have failed to do so if the trust deed provided such wide powers to him, that it failed to retain the “irreducible core” of trust duties to constitute a valid trust (referring to Armitage v Nurse) at [130]. Note there was no allegation of sham in Kennon v Spry. Also note the Court in Clayton did not refer to the decision in Raftland.

[31] 2003 JLR 188 at [58]

[32] Re Esteem Settlement at [58]

[33] Op cit at [52]. The Court also made it clear that there is not one concept of trust for Chancellery purposes and one for family law purposes.