Tax avoidance: Inland Revenue sheds further light on their approach in practice

By Campbell Rose and Brad Bowman

Inland Revenue has released a draft Question We’ve Been Asked, Income tax: scenarios on tax avoidance – 2015.  The draft QWBA has resulted from a panel discussion on the scenarios in question during the 2014 CAANZ Tax Conference.

The draft QWBA concludes that:

• Structuring an investment into an existing business using a limited partnership is not tax avoidance;
• Borrowing funds from, and investing in a portfolio investment entity (PIE) sponsored by the same bank constitutes tax avoidance; and
• Taking into account the tax profile of a beneficiary when distributing income from a discretionary trust on its own is not tax avoidance (but factual variations may alter this view).

Although the draft QWBA leaves some important questions unanswered, Inland Revenue are to be commended both for participating at the Conference and for committing resources to produce (and consult on) the draft QWBA.  This latest guidance adds to the other recent QWBAs that also addressed tax avoidance in practice: QB 14/11 and QB 15/01.

Structuring using a limited partnership

The first scenario involves a company with available tax losses (Loss Co), which wholly owns a profitable operating subsidiary (Op Co).  Loss Co wishes to introduce a new 50% investor (Investor Co).  The investment is structured by Op Co selling its business to a limited partnership with Loss Co and Investor Co as 50:50 investors.

The effect is to enable Loss Co to continue to offset its losses against its 50% share of Op Co’s income.  If Investor Co had simply purchased 50% of Loss Co’s shares in Op Co (an economically equivalent transaction), that tax outcome would not have arisen.  It is clear that tax considerations significantly influenced the manner in which the investment was structured.

The draft QWBA importantly reconfirms that, in undertaking the avoidance analysis, the economic effects of the structuring must not be confused with comparing the transaction actually undertaken with one that is economically equivalent (but which produces different, and perhaps less optimal, tax outcomes).  This is consistent with the Supreme Court’s observation in Ben Nevis that in commerce there are different means of producing the same economic outcome which have different tax consequences.

Of equal significance is the draft QWBA’s confirmation that taxpayers are not obliged to elect a structure that requires them to pay the highest amount of tax: “…there is no general requirement for the parties in this scenario to adopt an alternative, less tax-favourable, arrangement”.  As the Supreme Court stated in Ben Nevis and in Penny & Hooper, “… taxpayers have the freedom to structure transactions to their best advantage” (assuming that provisions are used in a way intended by Parliament).

Determining the commercial and economic reality of the arrangement therefore should be based on what actually happened, not what could have happened in the alternative.  We are seeing Inland Revenue investigators postulate hypothetical (and less tax efficient) alternatives as part of their tax avoidance analysis, and so it is hoped that the principles reaffirmed in the draft QWBA will be respected in practice.

This principle has parallels with Sir Ivor Richardson’s observations in CIR v BNZ Investments Ltd [2002] 1 NZLR 450 (at paragraph [40]), that “commerce is legitimately carried out through a range of entities and in a variety of ways; that tax is an important and proper factor in business decision making and family property planning; that something more than the existence of a tax benefit in one hypothetical situation compared with another is required to justify attributing a greater tax liability” [emphasis added].

The draft QWBA concludes that all of the facts, features and attributes Parliament would expect to see present in the arrangement to give effect to Parliament’s purposes for the specific provisions are present – the tax outcomes are therefore within Parliament’s contemplation, and accordingly do not fall foul of the general anti-avoidance provision.

We agree with this conclusion.  Our final observation is that the tax effects of the arrangements listed in the draft QWBA should also include the tax consequences of the business sale itself (e.g. income may arise from the sale of trading stock or fixed assets) – supporting the commercial and economic reality of what has been implemented.

Borrowing funds to invest in a PIE

The second scenario involves an individual on the top marginal tax rate of 33% who borrows funds from Bank A (incurring interest at 5.0% p.a.) and invests the funds in a multi-rate portfolio investment entity (PIE) sponsored by Bank A.  The PIE’s asset is deposits with Bank A which earn a fixed pre-tax return of 4.9% p.a.  The individual notifies the PIE to apply a prescribed investor rate of 28% and deducts their interest expenditure.

The arrangement is pre-tax negative (a loss of 0.1% p.a.), but post-tax positive (net return of 0.178% p.a.).  According to Inland Revenue’s general anti-avoidance interpretation statement, this is an indicator that suggests tax avoidance, however it is not determinative in itself (refer paragraph [349] of IS 13/01).  The draft QWBA reiterates that a strong tax influence in structuring does not automatically give rise to tax avoidance (refer paragraphs [56] and [87]).

In addition, the arrangement involves a circular movement of funds from Bank A to the investor, and then back to Bank A via the PIE.  Circularity of funds is another indicator of tax avoidance, but similarly it is not conclusive (refer paragraph [24] of IS 13/01).

The QWBA finds that the arrangement has no investment or savings element as a matter of commercial and economic reality, which is contrary to how Parliament contemplated the PIE rules should be used (as compared with, say, a taxpayer who withdraws funds off deposit and invests them in a similar cash PIE – which IS 13/01 concludes is not tax avoidance).  Inland Revenue has also reiterated its view that a general purpose of the arrangement (the generation of investment income) is not sufficient to displace a tax avoidance purpose or effect as being merely incidental; the purpose must explain the particular structure adopted.

The purpose or effect of this arrangement could be described as the generation of investment income.  For this arrangement, the general purpose of generating investment income could have been achieved in multiple different ways.  It does not explain the specific structure of the arrangement.  The draft QWBA therefore concludes that the tax avoidance purpose or effect was not merely incidental to the general purpose of generating investment income.

Ultimately, the draft QWBA concludes that borrowing funds from a bank and investing them in a PIE sponsored by the same bank is tax avoidance.  Although a number of factors are relied upon in reaching this view, we believe the most influential factors are likely to be the lack of real risk assumed by the taxpayer in making the investment, and the circularity: would the same conclusion be reached if the taxpayer borrowed from Bank A and invested in a PIE sponsored by Bank B?  This was briefly canvassed at the Conference panel session, but has not found its way into the draft QWBA.

Tax-influenced distributions from a discretionary trust

The third scenario involves the trustee of a discretionary trust distributing income to a taxpayer who is either:

• An individual adult beneficiary on a marginal tax rate lower than the trustee tax rate; or
• A corporate beneficiary with tax losses available that are equal to, or greater than, the income distributed; or
• A corporate beneficiary, where the income is a dividend from a foreign company and exempt under section CW 9.

As a starting point, the draft QWBA notes Parliament would contemplate that the facts, features and attributes present in an arrangement involving beneficiary income would include the existence of a valid trust (including satisfying the necessary prerequisites in the formation of the trust), and the trustees of the trust acting in accordance with the trust deed and general trust law.

The draft QWBA effectively goes on to observe that if the recipient beneficiary is validly entitled to a distribution, the distribution is properly effected and the recipient does actually benefit from the distribution, then the commercial and economic reality of the arrangement is the same as its legal form (i.e. a distribution of income to a beneficiary as beneficiary income). Again, this is what Parliament contemplates for the trust rules.  On this basis, the arrangement’s tax outcomes are not tax avoidance.

We agree with the draft QWBA’s conclusion.  This is welcome guidance given that the distribution of income by trustees of discretionary trusts to a variety of beneficiaries, including by reference to their tax profile, is widespread and common-place in New Zealand – which is notorious for its proliferation of trusts.

However, we do not agree with all aspects of the draft QWBA’s analysis, and it does leave open some significant questions:

• The draft QWBA contemplates a “solvent” corporate beneficiary with tax losses available to it.  The scenario discussed at the CAANZ Tax Conference did not involve a solvent beneficiary and, in practice, a beneficiary with tax losses is quite likely to be insolvent (or barely solvent).  If the corporate beneficiary is owned by a beneficiary or beneficiaries of the trust, it is quite plausible that the trustee would consider it entirely appropriate to distribute funds to the corporate in order to bolster its balance sheet – to put it in a better position to continue to trade, or assist it to satisfy bank covenants, or to enable it to repay debt (etc).  It is difficult to see how this additional factual feature should make any difference to the anti-avoidance analysis.  Indeed, it seems axiomatic that the beneficiary in such circumstances clearly is “benefiting” from the distribution.  This should be revisited in the finalised QWBA.
• The QWBA refers to distributions within six months of the income year-end constituting beneficiary income.  A relatively recent law change extended the six month time limit to the earlier of (a) the date on which the trustee actually files the return of income or (b) the date by which the trustee must file the return: in practice it is often greater than six months.
• The draft QWBA correctly states that a valid declaration or resolution by a trustee allocating income to a beneficiary will be sufficient for an amount to be “paid” (see [77] – at [78] it is confirmed that physical payment is not required).  Despite this, in the context of its avoidance analysis the draft QWBA appears to place significance on whether distributions are actually received by beneficiaries (see [82], [91] and [92]).  In our view actual cash payment/receipt is not a fact, feature and attribute that Parliament contemplated to be present in an arrangement involving beneficiary income.  Given that a valid and binding trustee resolution creates a receivable/sub-trust in the beneficiary’s favour, which can be dealt with by them for their benefit, we consider that this aspect of the draft QWBA needs to be modified when it is finalised.
• Linked to the “paid” point, the draft QWBA suggests that journal entries may function to “pay” or otherwise effect a distribution (refer [91] and [92]).  This is not correct, and appears to implicitly (and, we consider, wrongly) suggest that a journal entry ‘in combination’ with other features may lead to a tax avoidance conclusion.  A journal entry is simply an accounting entry recording a transaction that has already occurred.  The journal entry is not the “payment” of the distribution; it simply records that the beneficiary has a beneficial interest in an amount equal to the distribution that has already occurred by virtue of the trustee’s declaration or resolution.  This should be corrected in the finalised QWBA.
• Disappointingly, the draft QWBA is silent on what factual variations in this scenario would give rise to tax avoidance (or, at least, Inland Revenue scrutiny).  It would not be difficult to non-exhaustively list some of these in terms of what Inland Revenue have encountered in practice.  If Inland Revenue published anonymised Disputes Review Unit reports (whether taxpayer favourable or unfavourable), then this would go a long way to inform taxpayers and their advisers of broadly what factual features of arrangements are likely to be considered problematic (or not) in the avoidance analysis.

Inland Revenue’s deadline for comment on the draft QWBA is 23 July 2015.  If you have any comments or if you wish to discuss the draft QWBA, please contact your usual Deloitte advisor.

July 2015 Tax Alert contents

Tax Alert - July 2015

CIR v Trustpower Ltd: Unfeasibility expenditure

Bright-line test for sale of residential property - issues paper released

Currency conversions for branches – draft released

Tax avoidance: Inland Revenue sheds further light on their approach in practice

Take care when transferring funds from a UK pension plan

New tax bill – the start of the business transformation process

Increase in pooling value threshold for depreciation purposes