10 - Policy report: BEPS – transfer pricing and permanent establishment avoidance submissions and policy decisions

Document number	10
Date	22 June 2017
Type	Policy report
Title	BEPS – transfer pricing and permanent establishment avoidance submissions and policy decisions
Description	Report on submissions and policy changes resulting from the Government's discussion document BEPS – Transfer pricing and permanent establishment avoidance (March 2017) – for the submissions see document #24. Report number: IR2017/330, T2017/1577
Downloads	PDF (2.1 MB; 29 pages) DOCX (222 KB; 29 pages)
Contents	Cover page Executive summary Recommended action Background General views on the proposals PE avoidance Summary of proposed rule General reaction Threshold for the application of the new measures Interaction with MLI Overriding DTAs Source rules PE avoidance source rule Anti-avoidance source rule Life insurance rules / FIF rules Transfer Pricing General reaction Extending the time bar to 7 years Shifting the burden of proof from Inland Revenue to the taxpayer The test for reconstructing a transfer pricing arrangement Administrative measures Summary of measures General reaction Earlier payment of disputed tax Collection of tax from a local subsidiary Collection of information Application date and grandparenting APAs Application date Grandparenting APAs Appendix One: List of submitters

---

Tax policy report: BEPS – transfer pricing and permanent establishment avoidance submissions and policy decisions

---

Date:	22 June 2017	Priority:	Medium
Security level:	In Confidence	Report no:	T2017/1577 IR2017/330

Action sought

	Action sought	Deadline
Minister of Finance	Agree to the recommendations	29 June 2017
Minister of Revenue	Agree to the recommendations	29 June 2017

Contact for telephone discussion (if required)

Name	Position	Telephone
Gordon Witte	Senior Policy Advisor, Inland Revenue	[Withheld under section 9(2)(a) of the Official Information Act 1982]
Sam Rowe	Senior Policy Advisor, Inland Revenue
Steve Mack	Principal Advisor, The Treasury
Carmel Peters	Policy Manager, Inland Revenue

 

22 June 2017

Minister of Finance
Minister of Revenue

BEPS – transfer pricing and permanent establishment avoidance submissions and policy decisions

---

Executive summary

1. In March this year the Government released a tax discussion document called BEPS – transfer pricing and permanent establishment avoidance which proposed the introduction of a package of transfer pricing and permanent establishment (PE) avoidance rules targeted at countering large multinationals engaged in aggressive tax practices. This report provides advice on the 16 submissions we have received on this discussion document. It also seeks policy decisions on the reform package following this consultation.

2. In summary, we recommend proceeding with all but one of the proposals in the discussion document (we do not recommend proceeding with the proposal to require large multinationals to pay disputed tax earlier). We recommend making a number of refinements to some of the original proposals in response to submissions. These refinements will make the proposals more certain for taxpayers and better targeted at the base erosion and profit shifting (BEPS) arrangements we are concerned about.

3. Agreeing to our recommended changes will not change the previously estimated forecast tax revenue from the transfer pricing and PE proposals (which is $25m in 2018/19 and $50m per annum thereafter).

Summary of submissions

4. Two submitters (Oxfam and NZCTU)[1] expressed support for all the proposals on the grounds that they would help ensure multinationals pay their fair share of tax.

5. Most submitters accepted in principle the need for measures to address the transfer pricing and PE avoidance issues identified in the discussion document. However, they did raise issues with certain features of the proposed measures and made suggestions to make them more workable and better targeted. Many submissions focused on when the PE avoidance rule would apply and when Inland Revenue would reconstruct a transfer pricing arrangement. We are confident we can refine the proposals to address many of the submitters’ concerns while ensuring the measures are just as effective at combatting BEPS.

6. Some submitters were opposed to proceeding with any of the proposed measures as they considered the new rules were unnecessary and would make New Zealand’s tax environment more uncertain and unattractive for multinational investment.

7. Most of the submitters are tax advisors or represent businesses that could be negatively affected by the proposals. Therefore, the submissions are understandably critical of some of the measures. As expected, submitters strongly opposed the proposals that increased Inland Revenue’s powers to investigate large multinationals. These administrative proposals included:

• extending the transfer pricing time bar from 4 years to 7 years. This is the period within which Inland Revenue can adjust a transfer pricing tax position taken by a taxpayer;
• allowing Inland Revenue to request information that is held by an offshore group member; and
• requiring large multinationals to pay disputed tax upfront (rather than at the end of a dispute).

Our response

8. We agree with submitters that the proposal to make large multinationals pay disputed tax upfront is unnecessary, and recommend not proceeding with the proposal. Inland Revenue already charges “use of money interest” on tax owing, which provides a strong incentive for paying tax which is in dispute.

9. We recommend proceeding with all of the other proposals in the discussion document, subject to a number of refinements to make the proposals more certain for taxpayers and better targeted. These refinements should not reduce the overall effectiveness of the proposed reforms.

10. Otherwise, we consider the measures are well-targeted at the specific problems that Inland Revenue has actually observed in its investigations of multinationals. Currently only a small number of multinationals use the aggressive PE avoidance or transfer pricing arrangements which are targeted by the proposals. This suggests the new rules will only increase uncertainty or tax costs for a small number of multinationals.

11. The following table summarises the main issues raised by submitters and our recommended responses:

Submission	Recommended response
The anti-avoidance threshold for the application of the PE avoidance rule should be narrowed so it does not apply to ordinary commercial arrangements.	Accept the submission. We consider the rule should be more narrowly targeted at avoidance arrangements. We could do this either by requiring a more than merely incidental purpose of tax avoidance, or by adopting into domestic legislation the OECD’s widened PE definition. We would like to give further consideration as to which approach we should adopt.
The PE avoidance rule is not necessary in light of the OECD’s new widened PE definition (which New Zealand is implementing with some countries through the Multilateral Instrument and through future double tax agreement (DTA) negotiations).	Accept the submission in part. In cases where the applicable DTA includes the OECD’s new widened PE definition, the proposed PE avoidance rule seems unnecessary. However the OECD’s widened PE definition will not be included in most of our DTAs under the MLI (although it may be included under subsequent bilateral DTA negotiations). To reflect this, we recommend that the proposed PE rule apply only where an applicable DTA does not include the OECD’s widened PE definition.
The PE avoidance rule should not override New Zealand’s DTAs.	Decline the submission. For the rule to be effective it needs to override those DTAs which do not include the OECD’s new widened PE definition. This is consistent with the Australian and UK approaches.
The proposed anti-avoidance source rule is too broad and should be more targeted at the perceived problem.	Accept the submission. We consider the rule should be more narrowly targeted at the existing issues Inland Revenue has identified with the source rules.
The life insurance proposals represent an unfair and unilateral reconstruction of the tax treatment of life insurance premiums and should not proceed.	Accept the submission in part. We consider that the proposed reinsurance amendments are necessary to ensure that the rules apply as intended and to protect the tax base. However, there is little revenue at risk in relation to the foreign investment fund amendments and a significant likelihood of accidental non-compliance under the proposed change. Accordingly, we recommend that the foreign investment fund related life insurance changes do not proceed (meaning that any life insurance policies issued in New Zealand by life insurers from Singapore, Russia, and Canada would remain exempt from the foreign investment fund rules).
The time bar which limits Inland Revenue’s ability to adjust a taxpayer’s transfer pricing position should remain at 4 years (not be extended to 7 years as proposed).	Decline the submission. We consider there is still a good justification for extending the time bar to 7 years for transfer pricing issues (consistent with Australia and Canada).
The burden of proof for transfer pricing matters should remain with Inland Revenue (rather than being shifted onto the taxpayer as proposed).	Decline the submission. The burden of proof is on the taxpayer for transfer pricing matters in most OECD and G20 countries, including Australia. This means most multinationals already have to prepare transfer pricing documentation that satisfies the burden of proof for other countries. Also, the burden of proof is on the taxpayer for other tax matters.
The test for reconstructing a transfer pricing arrangement should align with the OECD’s transfer pricing guidelines.	Accept the submission. New Zealand’s legislative test for reconstructing an arrangement should be based on the corresponding test in the OECD’s transfer pricing guidelines.
The proposal to make large multinationals pay disputed tax upfront is unnecessary and should not proceed.	Accept the submission. We recommend not proceeding with this proposal as Inland Revenue already charges “use of money interest” on tax owing, which provides a strong incentive for multinationals to pay tax that is in dispute.
The proposal to require a New Zealand member of a multinational group to pay tax owed by a related non-resident group member should not proceed.	Decline the submission. However, we agree that the rule should only apply if the non-resident fails to pay the tax itself.
The proposed extension of Inland Revenue’s information collection powers to allow Inland Revenue to request information that is held offshore by a related group member should not proceed. Submitters also raised concerns about the new civil penalty of up to $100,000 for failing to provide requested information (which replaces the current $12,000 maximum criminal penalty).	Decline the submissions. We consider that these information proposals are necessary to ensure that the multinational group is required to provide Inland Revenue with the requested information and has appropriate incentives to comply with these requests. However, we recommend allowing the multinational to appeal the penalty.

12. We also propose widening the scope of the original proposal to deem an amount of income to have a New Zealand source under our domestic legislation if we have a right to tax the income under a DTA. The rule proposed in the discussion document was limited to income covered by the PE and royalty articles of our DTAs. We should extend the rule to all types of income that we can tax under a DTA – as Australia does. This ensures we can exercise a taxing right that we have negotiated under a DTA. We will consult further on this wider proposal in the next round of consultation.

13. Officials are available to discuss this report at your joint Ministers’ meeting on 29 June. Further information about the next steps is set out in the cover report included in this package (BEPS – submissions on March 2017 discussion documents – covering report T2017/1578 / IR2017/329).

Recommended action

We recommend that you:

(a) Agree that the proposal in the discussion document to require large multinationals to pay disputed tax upfront should not proceed as it is not necessary given that Inland Revenue already charges “use of money interest” on tax owing.

Agreed / Not Agreed   Agreed / Not Agreed

(b) Note that the recommendations below agree to implement all the other proposals in the discussion document, subject to some refinements in response to submissions (as identified in this report).

Noted   Noted

(c) Note that the recommended refinements will provide more certainty for taxpayers without reducing the overall effectiveness of the proposed reforms. Therefore agreeing to officials’ recommendations will not affect the estimated forecast tax revenue from implementing the transfer pricing and PE avoidance measures, which is $25m in 2018/19 and $50m per annum thereafter.

Noted   Noted

(d) Agree to introduce a new PE avoidance rule that will apply to large multinationals that structure to avoid having a permanent establishment (taxable presence) in New Zealand. The rule will only apply to multinationals with over EUR €750m of consolidated global turnover.[2] The rule will not apply if the relevant DTA already includes the OECD’s new widened PE definition.

Agreed / Not Agreed   Agreed / Not Agreed

(e) Note that, in designing the detail of the new PE avoidance rules, officials will consider options for narrowing the original scope of the PE avoidance rules without reducing their effectiveness. We will report back with our recommendations on this matter.

Noted   Noted

(f) Note that we will consult further on a new source rule which will deem an amount of income to have a source in New Zealand if New Zealand has a right to tax that income under any applicable DTA.

Noted   Noted

(g) Agree to introduce an anti-avoidance source rule which will broadly provide that, where another group member carries on a non-resident’s business in New Zealand, the non-resident will be deemed to carry on that business itself for the purpose of determining whether its income from New Zealand customers has a New Zealand source. (This is more narrowly targeted at the existing issues Inland Revenue has identified with the source rules than the original proposal.)

Agreed / Not Agreed   Agreed / Not Agreed

(h) Agree to address a potential weakness of the life insurance source rules by ensuring that no deductions are available for the reinsurance of life insurance policies if the premium income on that policy is not taxable in New Zealand including where the income is not subject to New Zealand tax under a DTA.

Agreed / Not Agreed   Agreed / Not Agreed

(i) Agree that the proposal to amend the FIF life insurance rules should not proceed as there is little revenue at risk and a significant likelihood of accidental non-compliance under the proposal.

Agreed / Not Agreed   Agreed / Not Agreed

(j) Agree that the time bar that limits Inland Revenue’s ability to adjust a taxpayer’s transfer pricing position should be increased to 7 years (in line with Australia).

Agreed / Not Agreed   Agreed / Not Agreed

(k) Agree that the burden of proof for demonstrating that a taxpayer’s transfer pricing position aligns with arm’s length conditions should be shifted from Inland Revenue to the taxpayer (consistent with the burden of proof being on the taxpayer for other tax matters).

Agreed / Not Agreed   Agreed / Not Agreed

(l) Agree to strengthen the transfer pricing rules so they align with the OECD’s transfer pricing guidelines and Australia’s transfer pricing rules. This involves amending New Zealand’s transfer pricing rules so that:

•  they disregard legal form if it does not align with the actual economic substance of the transaction;
• they provide Inland Revenue with a power to reconstruct transfer pricing arrangements which are not commercially rational because they include unrealistic terms that third parties would not be willing to agree to;
• the legislation specifically refers to arm’s length conditions and using the latest OECD’s transfer pricing guidelines as guidance material for how the rules are applied; and
• the new legislation codifies the requirement that large multinationals will provide Inland Revenue with the information required to comply with the OECD’s country-by-country reporting initiative.

Agreed / Not Agreed   Agreed / Not Agreed

(m) Agree that New Zealand’s legislative test for reconstructing a transfer pricing arrangement should be based on the corresponding test in the OECD’s transfer pricing guidelines.

Agreed / Not Agreed   Agreed / Not Agreed

(n) Agree that in addition to applying to transactions between related parties, the transfer pricing rules should also apply when non-resident investors “act in concert” to effectively control a New Zealand entity, such as through a private equity manager.

Agreed / Not Agreed   Agreed / Not Agreed

(o) Agree that if a large multinational group (over EUR €750m worldwide revenues) does not cooperate with Inland Revenue, then Inland Revenue may more readily dispute the multinational’s tax position based on the information available to Inland Revenue at the time.

Agreed / Not Agreed   Agreed / Not Agreed

(p) Agree that tax owed by any member of a large multinational group can be collected from any wholly-owned group member provided the non-resident fails to pay the tax itself (this is slightly narrower than the original proposal in the discussion document).

Agreed / Not Agreed   Agreed / Not Agreed

(q) Agree to extend Inland Revenue’s information collection powers so that in respect of large multinational groups, Inland Revenue can request information that is held offshore by a related group member.

Agreed / Not Agreed   Agreed / Not Agreed

(r) Agree to extend Inland Revenue’s information collection powers so that Inland Revenue can deem an amount of income to be allocated to a New Zealand group member or New Zealand PE of a large multinational group in cases where they have failed to adequately respond to an information request in relation to New Zealand-sourced income. (Currently the existing power only applies in respect of deductible payments.)

Agreed / Not Agreed   Agreed / Not Agreed

(s) Agree to create a new civil penalty of up to $100,000 for large multinational groups which fail to provide requested information (which replaces the current $12,000 maximum criminal penalty), but clarify that the taxpayer would be able to appeal this penalty.

Agreed / Not Agreed   Agreed / Not Agreed

(t) Agree that advance pricing agreements (APAs) existing prior to the application date of these proposals should be grandparented.

Agreed / Not Agreed   Agreed / Not Agreed

 

Steve Mack
Principal Advisor
The Treasury

Carmel Peters
Policy Manager
Inland Revenue

Steven Joyce
Minister of Finance

Hon Judith Collins
Minister of Revenue

Background

14. In March this year the Government released a tax discussion document called BEPS – transfer pricing and permanent establishment avoidance. This report provides advice on the 16 submissions (from 15 submitters) we have received on this discussion document. It also seeks policy decisions on the proposals, including a number of suggested refinements to address issues raised by submitters.

15. We have met with six of the main submitters (CA ANZ, CTG, PwC, KPMG, EY, DEG) to discuss their submissions and explain the proposals. We will continue to work with these and other submitters to develop the detailed design of the legislation.

16. This report advises on the important issues relevant to the main policy decisions to be taken by Cabinet in July. Following these decisions, we will design the detail of the proposals, on which there will be further targeted consultation in August to October of this year. A number of the submissions related to the detail of the proposals or to Inland Revenue’s operational approach. For example many taxpayers asked that Inland Revenue develop practical guidance on how the proposed new rules would apply. We will advise you on these detailed design and operational submissions following the next round of consultation.

General views on the proposals

17. The proposed transfer pricing and PE avoidance rules are targeted at countering large multinationals engaged in aggressive tax practices.

18. Some submitters welcomed these proposals as a positive step by the Government to ensure that all large multinationals pay their fair share of tax (Oxfam and CTU).

19. Most submitters accepted in principle the need for measures to address the transfer pricing and permanent establishment (PE) avoidance issues identified in the discussion document. However, they did raise issues with certain features of the proposed measures and made suggestions to make them more workable and better targeted. Many submissions focused on when the PE avoidance rule would apply and when Inland Revenue would reconstruct a transfer pricing arrangement.

20. Other submitters argued that the proposals will have a detrimental effect on New Zealand being an attractive investment destination and should not be implemented (CTG, CA ANZ, and NFTC). These submitters argued that the proposals introduce complex and onerous rules which may incentivise foreign companies to remove their existing personnel from New Zealand, which would reduce GDP and lower employment levels.

21. We disagree with these submissions. First, the majority of multinationals operating in New Zealand are compliant and the Government is committed to making sure New Zealand remains an attractive place for them to do business. However, there are some multinationals that set out to circumvent New Zealand’s tax rules. These multinationals should not be allowed to exploit weaknesses in the transfer pricing and PE rules to achieve a competitive advantage over more compliant multinationals or domestic firms. Second, it is highly unlikely that foreign companies will remove their existing personnel from New Zealand as a result of these proposals. Most of the affected foreign companies are dependent on having personnel in New Zealand to arrange their sales. Without personnel on the ground, they would not be able to service their New Zealand market. It is also unlikely that they would cease to operate in New Zealand altogether.

22. The transfer pricing and PE proposals introduce a set of rules to reinforce the integrity and efficiency of the tax system so that there is a level playing field for multinationals and domestic firms.

PE avoidance

Summary of proposed rule

23. Where a DTA applies, New Zealand is only able to tax a non-resident on its income from sales to New Zealand customers if the non-resident has a PE in New Zealand. The discussion document proposed a rule to prevent non-residents from structuring their affairs to avoid having such a permanent establishment in New Zealand where one exists in substance.

24. The rule proposed in the discussion document would deem a non-resident to have a PE in New Zealand if:

• the non-resident supplies goods or services to a person in New Zealand;
• the non-resident is part of a multinational group with more than EUR €750m of consolidated global turnover;
• a related entity (either associated or commercially dependant) carries out an activity in New Zealand in connection with that particular sale for the purpose of bringing it about;
• some or all of the sales income is not attributed to a New Zealand PE of the non-resident; and
• the arrangement defeats the purpose of the relevant DTA’s PE provisions.

General reaction

25. Submitters were not strongly opposed to a new PE rule in principle, with 2 submitters supporting the proposal (Oxfam and NZCTU) and the remainder mostly accepting the need (or inevitability) for some form of PE avoidance rule. However 7 submitters considered that we should not adopt any PE avoidance rule at this stage. These submitters argued that:

• The OECD’s Multilateral Instrument (MLI)[3] includes a widened definition of a PE so any PE avoidance issues should be addressed under this. Alternatively, we should defer consideration of a PE avoidance rule until the impact of the OECD’s BEPS measures has been determined (EY, AmCham, DEG, CA ANZ).
• The rule is unnecessary, as any current issues with PE avoidance can be addressed through our transfer pricing rules (NZLS, DEG, CA ANZ).
• The rule will apply to non-abusive transactions, is outside the OECD’s BEPS initiatives, and will erode taxpayer certainty (CTG, NFTC, Deloitte).

26. In response to these submissions, we consider that:

• Where the widened definition of a PE in the MLI applies, a domestic PE avoidance rule would not be necessary. However the widened definition applies only under the MLI where both countries choose to adopt it. We are aware that most countries do not intend to adopt the widened definition under the MLI (including the US, the UK, and Australia). We note that the widened definition is being added to the OECD’s model treaty, and we expect it to eventually be incorporated into most of our DTAs (including DTAs with countries that did not elect to adopt the widened definition under the MLI) as each DTA is bilaterally renegotiated. However it will be many years before all our DTAs are bilaterally renegotiated. Therefore a domestic rule is necessary now to address PE avoidance by taxpayers resident in these countries.
• The principles underlying transfer pricing and PE profit attribution, while similar, are not the same. The transfer pricing rules seek to determine an arm’s length price for transactions between related entities. The PE profit attribution rules seek to determine what part of an enterprise’s overall profit should be attributed to a PE in a particular country. The OECD guidance is clear that profit may still be attributable to a PE even after the correct application of the transfer pricing rules (depending on the circumstances). In addition, deeming a PE to exist will allow us to charge non-resident withholding tax on any royalties paid by the non-resident that relate to its New Zealand sales. This will not be possible under the transfer pricing rules. Accordingly application of the transfer pricing rules alone would not produce the correct amount of tax for New Zealand in many cases where a PE is being avoided.
• Our proposed PE avoidance rule is broadly consistent with the OECD’s BEPS initiatives, as it should have a similar effect to the widened PE definition in the MLI. We recommend below some changes to our PE rule which should ensure it is better targeted at abusive transactions. Finally we acknowledge that the rule will reduce taxpayer certainty, which is undesirable. However we consider that this disadvantage is outweighed by the benefits of the proposed rule in terms of protecting the integrity of the tax system, fairness, revenue, and economic efficiency.

27. We therefore recommend proceeding with the introduction of a PE avoidance rule.

Threshold for the application of the new measures

28. The discussion document proposed that the PE avoidance rule would only apply to arrangements which defeated the purpose of the PE provisions in the applicable DTA. The explanation of this test in the discussion document focussed on the economic substance of the non-resident’s activity in New Zealand, and particularly whether a PE would arise if the non-resident and its local New Zealand subsidiary were treated as a single entity.

29. A majority of submitters (EY, NFTC, DEG, Deloitte, CTG, CA ANZ, PwC, KPMG, Russell McVeagh) considered that this proposed PE avoidance test was too broad. They considered that it would widen the PE definition in substance rather than just prevent its abuse. They were also concerned that it could capture ordinary commercial arrangements and discourage foreign investment. Some submitters considered that the PE avoidance rule should either be targeted at abusive or artificial arrangements or should adopt the wording of the OECD’s widened PE avoidance definition in the MLI (CTG, PwC, Deloitte, CA ANZ, DEG).

30. We agree with these submissions. We recommend that the rule be more narrowly targeted so that it does not apply to ordinary commercial arrangements, and so does not unduly discourage non-residents from doing business in New Zealand.

31. Submitters suggested two options for narrowing the scope of the rule:

• Option 1: Replace the current requirement that the rule defeats the purpose of the PE article in a DTA with a purpose of avoidance test. Under this new test, the rule would only apply if the relevant arrangement had a more than merely incidental purpose of tax avoidance. This would target the rule at avoidance transactions and align the rule with similar rules in Australia and the UK, each of which requires the taxpayer to have a purpose of avoiding tax. Because it is an express anti-avoidance rule, it would also be consistent with our DTAs. The rule would apply more broadly in the context of PE avoidance than our current general anti-avoidance rule.[4]
• Option 2: Replace the PE avoidance rule proposed in the discussion document with the OECD’s widened definition of a PE, which we would add to our domestic legislation as a standalone rule. The OECD’s widened definition provides that a PE arises if a representative of the non-resident plays the principal role leading to the conclusion of contracts that are routinely concluded without material modification by the non-resident. The widened definition also includes some supplementary provisions to counter particular avoidance strategies observed overseas. This rule is an objective test and it is the result of the OECD’s BEPS Action Plan, which included a report on “Preventing the Artificial Avoidance of Permanent Establishment Status” (Action 7). As indicated by the title of the report, the purpose of the rule is to prevent the artificial avoidance of PE status under various BEPS strategies. The widened PE definition would apply in respect of a DTA regardless of whether the other country elected to adopt it under the MLI. This test would be more certain for foreign investors than option 1 and would replicate the OECD’s recommended measure.

32. We consider that either option would be effective in addressing the PE avoidance arrangements we have seen in New Zealand. Accordingly adoption of either option would not affect the previously forecast revenue estimates for the measures. We would like to further evaluate the merits of the two options before recommending one.

Interaction with MLI

33. Several submitters questioned how our proposed PE avoidance rule would interact with the widened definition of a PE in the MLI. As noted above, some submitters considered that PE avoidance should be dealt with under the MLI (EY, AmCham, DEG, CA ANZ). Other submitters questioned how our domestic PE rule would work in relation to a DTA that included the MLI’s widened PE definition, or considered that any domestic rule should be consistent with the OECD’s BEPS actions (KPMG, CTG, NZLS, PWC).

34. There is merit in these submissions. The widened PE definition from the MLI should address PE avoidance, provided it is included in the relevant DTA. Accordingly it is not necessary for our domestic PE avoidance rule to apply where there is an applicable DTA that includes the MLI’s widened PE definition. We also generally prefer to follow the OECD’s approach where that is practicable.

35. Accordingly, we recommend that our domestic PE avoidance rule apply only in respect of DTAs that do not include the widened PE definition from the MLI (or an equivalent definition that is negotiated bilaterally).

Overriding DTAs

36. A majority of submitters considered that our PE rule should not override our DTAs (CTG, KPMG, CA ANZ, NFTC, NZLS, EY, Russell McVeagh, DEG). This is because DTAs are important to international trade, and New Zealand exporters also need to rely on them. Submitters also considered that we should not depart from the OECD’s agreed BEPS measures, particularly where the country of the non-resident has declined to adopt the widened PE definition in the MLI.

37. The OECD’s Commentary to the Model Tax Convention (the Commentary) states that, as a general rule, there will be no conflict between domestic anti-avoidance provisions and the provisions of a DTA. It also confirms that States are not obliged to grant the benefits of a DTA if the DTA has been abused (noting that this should not be lightly assumed). In the present case, our first option for a PE rule is an anti-avoidance measure that only applies if there is a purpose of tax avoidance. Accordingly it should not conflict with New Zealand’s DTAs in the vast majority of cases. The second option is also an anti-abuse measure however it functions as a black letter amendment to the terms of our PE articles. Accordingly we would like to further consider its consistency with our DTAs before deciding which option to recommend.

38. In either case we consider that the PE rule should expressly override our DTAs. This is to simplify the application of the rule. Otherwise it would be necessary to show that the application of the rule was consistent with a DTA in each particular case. This would be a time-consuming and resource intensive exercise. It would significantly undermine the practical effectiveness of the rule. We also note that both the UK and Australian PE avoidance rules override their DTAs.

39. We also consider that the PE rule should apply in respect of DTAs where the other country has elected not to include the widened PE definition from the MLI. The existing position is that anti-avoidance rules are generally consistent with DTAs. We do not consider that a country’s decision not to adopt the widened PE definition in the MLI changes this principle. In particular, we do not consider that such a decision evinces a common intent that a DTA can now be abused by the taxpayer of either jurisdiction.

40. We note that in relation to the second option (which incorporates the OECD’s widened PE definition into our domestic legislation), the widened PE definition will be added to the OECD’s model tax treaty, and so represents what the OECD considers to be the current best practice. Countries may also not want to adopt such a provision multilaterally under the MLI, but may be happy to agree to such a provision in bilateral negotiations with New Zealand (such as Australia). Accordingly, the failure to adopt the widened PE definition under the MLI does not mean that they object to such a provision in their DTA with New Zealand.

Source rules

PE avoidance source rule

41. The rule proposed in the discussion document stated that an amount of income would have a New Zealand source if we had a right to tax the income under the PE or royalty article of a DTA. Only 2 submitters opposed this rule, on the basis that it was circular (CTG) and could result in a breach of our DTAs (EY). We do not consider there is any circularity to the proposal – the proposal will ensure that we are not prevented from taxing income under our domestic legislation where we have an agreed right to tax that income under our DTAs. Also, for this reason, the rule would never apply in contravention of our DTA rights.

42. We note the rule proposed in the discussion draft applied only in respect of income covered by the PE article and the royalty article of our DTAs. This was because we were only aware of issues with our domestic legislation in relation to these kinds of income. On reflection, we consider the rule should apply in respect of all types of income we can tax under an applicable DTA (e.g. interest, dividends, income from alienation of property, etc.). This is sensible because it ensures that we do not negotiate taxing rights under a DTA that we cannot exercise because of differences in the formulation of the source rules in our DTAs and our domestic law source rules. This is the same position which Australia takes under its DTAs, and the proposed rule already applies to all income covered by an article of our DTA with Australia. Since we are broadening our original proposal we should consult further with stakeholders as part of the generic tax policy process (GTPP).

Anti-avoidance source rule

43. The rule proposed in the discussion document provides that a non-resident’s income would have a source in New Zealand (and therefore give us a domestic law taxing right) if it would have a New Zealand source, treating the non-resident’s multinational group as a single entity. This would stop non-residents from dividing their activities between wholly-owned group members in order to prevent their income from having a New Zealand source.

44. Some submitters considered that this rule should not proceed in its current form (CTG, EY, CA ANZ, Russell McVeagh). They considered that it was unnecessary, as our existing source rules were already adequate. Submitters also considered that the rule was too broad and struggled to understand how it would work in practice. Two submitters noted that a more targeted rule could be more appropriate (CA ANZ, EY).

45. The rule was partly intended to address an existing technical issue with the source rules, and partly intended to prevent possible future attempts to circumvent the source rules. In light of submitters’ concerns, we consider that the rule should be more narrowly targeted at the existing issues with the source rules. In particular, the rule should broadly provide that, where a group member carries on a non-resident’s business in New Zealand, the non-resident is also deemed to carry on business in New Zealand to that extent. This will prevent non-residents from being able to avoid a New Zealand source for their income from sales to New Zealand customers by arranging for a wholly owned subsidiary to carry out their local business activities. If the rule applied, only the portion of the sales income that is attributable to the group member’s activities in New Zealand would generally be taxable here.

Life insurance rules / FIF rules

46. Life insurance premiums can be used to shift income out of New Zealand. As such, the Income Tax Act denies a deduction for reinsurance premiums when the corresponding premium income is not taxable in New Zealand.

47. Life insurance can also be used as a type of investment savings. For this reason, the foreign investment fund (FIF) rules apply to life insurance policies owned by New Zealand residents.

48. New Zealand’s DTAs typically preserve New Zealand’s entitlement to tax insurance premiums whether or not a permanent establishment exists. However, under New Zealand’s DTAs with Canada, Russia, and Singapore, New Zealand is unable to tax life insurance premiums if a resident of those countries does not have a permanent establishment in New Zealand. New Zealand’s inability to tax life insurance premium income under these DTAs means that the rules denying reinsurance deductions and the application of the FIF rules may not work as intended when the premium is paid to a non-resident life insurer or reinsurer from these countries. Furthermore, non-resident life insurers who are residents of Canada, Russia, or Singapore, are able to receive an unintended tax advantage by being able to deduct life reinsurance premiums.

49. The discussion document proposed an amendment to the Income Tax Act to specifically provide that no deduction is available for the reinsurance of policies if the premium income on that policy is not taxable in New Zealand (including under a DTA). An amendment to the definition of a FIF was also proposed to specifically provide that New Zealand residents are subject to the FIF rules in respect of any policies that are not subject to New Zealand tax under the life insurance rules or any applicable DTA.

50. Submitters argued that the life insurance proposals should not proceed as they represent an unfair and unilateral reconstruction of the tax treatment of life insurance premiums (KPMG, EY, Deloitte, CTG, CA ANZ). Submitters argued that during treaty negotiations with Canada, Russia, and Singapore, New Zealand must have either accepted to change its standard practice of taxing insurance premiums, or inadvertently made the change – neither reason providing sufficient justification for the proposals. The submitters considered that the correct approach would be for New Zealand to renegotiate the relevant provision with Canada, Russia, and Singapore.

51. Submitters also argued that the proposals unfairly penalise the reinsured party by placing a significant burden on them to have completeness of information regarding their insurer’s place of tax residence and PE status in NZ (CTG and Deloitte). Should the proposals advance, Deloitte considers that appropriate grandparenting should be provided.

52. We agree with the submissions in part. We consider that the proposed reinsurance amendments are necessary to ensure that they apply as intended. These proposals will also ensure that life insurance businesses operating out of Canada, Russia, and Singapore will no longer benefit from more favourable tax treatment compared with those operating in New Zealand or other countries. However, we recommend that the FIF life insurance changes do not proceed as there is little revenue risk involved and a significant likelihood of accidental non-compliance under the proposed changes. This means that any life insurance policies issued in New Zealand by life insurers from Singapore, Russia, and Canada would remain exempt from the foreign investment fund rules.

Transfer Pricing

53. Transfer pricing rules guard against multinationals using related party payments to shift profits offshore by requiring these payments to be consistent with an arm’s length/market price that unrelated parties would agree to. Chapter 5 of the discussion document outlined a package of proposals to strengthen the transfer pricing rules.

General reaction

54. Three submitters (CTG, EY, KPMG) considered the transfer pricing proposals were unnecessary as the existing transfer pricing rules were sufficient. We disagree as New Zealand’s existing transfer pricing legislation would not allow us to fully implement the OECD’s new transfer pricing guidelines (that were developed to combat BEPS) as it does not explicitly require transfer pricing practices to align with the actual economic activity (if this differs from the legal contracts) and does not include a reconstruction provision.

55. Other submitters generally accepted that there was a need to update New Zealand’s transfer pricing legislation so it aligns with the OECD’s transfer pricing guidelines and Australia’s transfer pricing rules. However, as expected, there was strong opposition to the administrative proposals to extend the time bar for transfer pricing adjustments to 7 years and to shift the burden of proof onto the taxpayer for providing evidence that they comply with the transfer pricing rules.

Extending the time bar to 7 years

56. Inland Revenue currently has 4 years from the end of the tax year in which a taxpayer files an income tax return to investigate and amend the tax position taken by the taxpayer in their return. This 4 year limit is known as the time bar. The discussion document proposed that transfer pricing issues should have a longer time bar of 7 years (consistent with fact that Australia and Canada have 7 year time bars for transfer pricing).

57. Eight of the 15 submitters (CTG, PwC, KPMG, CA ANZ, EY, AMP (NZ), Russell McVeagh, NFTC, DEG, NZLS) opposed this proposal. The main reasons for opposing a longer time bar were:

• A longer time bar increases uncertainty for taxpayers and does not promote efficiency in transfer pricing disputes (will delay timely resolution).
• The discussion document argued that a longer time bar is needed because transfer pricing issues are complex and fact-specific, but this is also true of other areas of tax such as tax avoidance, the capital/revenue boundary and complex financial arrangements.
• Most countries have the same time bar for transfer pricing and other tax issues, and in most cases this was less than 7 years.
• If a transfer pricing dispute is resolved in favour of Inland Revenue, the taxpayer will be at risk of double tax in jurisdictions where the time bar has already passed.
• Imposing a longer time bar is inconsistent with Inland Revenue’s Business Transformation goals of real-time review and helping taxpayers get it right from the start.
• Inland Revenue should invest more resource into its transfer pricing team if the investigations are taking longer than 4 years.

58. Officials are not convinced by these arguments and consider there is still a good justification for extending the time bar to 7 years for transfer pricing issues. There are a number of reasons why transfer pricing investigations can take more time than other types of tax investigation:

• The factual review for transfer pricing cases is typically much more detailed than other tax issues and may involve discussions with numerous staff and the taxpayer, in addition to the usual review of legal documents etc. It may also involve wider industry interviews, e.g. with regulators, competitors, customers etc. to provide the necessary market context. The relevant documentation or information may be held outside New Zealand which can delay when this information is provided to Inland Revenue.
• Assessing compliance with the arm’s length principle requires very detailed and specific information and analysis of how a comparable transaction between unrelated parties would have been conducted. This means there are effectively two parallel investigations – determining the facts of the actual related party transaction and identifying a comparable arm’s length arrangement.
• Certain complex transactions require input from market experts typically based overseas. Vetting, engaging, and briefing an overseas expert takes time. Depending on the nature of the issues, the expert’s opinion may also take some time to prepare.
• There is usually a range of possible answers in transfer pricing cases and this leads to more frequent and extensive discussions and negotiations throughout the process. Taxpayers generally wish to engage in discussions and negotiations (and exchange issues papers) prior to entering the disputes process. There are also often settlement discussions during the disputes process that can go on for many months at a time.
• There may also be numerous and lengthy discussions with treaty partners in the course of a transfer pricing investigation to not only obtain additional information but also endeavour to resolve differences without double taxation arising.

59. Currently, most transfer pricing investigations take less than 4 years and we expect this will continue under the proposed new rules. The longer time bar is therefore only expected to be relevant in a handful of complex cases. However, it is important to have more time available to identify, investigate and resolve these cases as they can involve very large sums of tax.

60. One concern with a longer time bar is that it could lead to more years of income being part of a dispute, which could reduce incentives for taxpayers and Inland Revenue to agree on a settlement to the dispute. However, Inland Revenue is increasingly picking up the vast majority of the arrangements it wants to challenge on a relatively real-time basis (often year two, taking into account filing timeframes which generally mean a return is not filed until the start of year two) which should lead to fewer years being under dispute.

61. New Zealand is adopting Article 17 of the MLI which will update our DTAs so that they require our DTA partners to make appropriate corresponding adjustments in transfer pricing cases. This will ensure that double taxation does not arise due to New Zealand making a transfer pricing adjustment, even if this is beyond the other country’s time bar.

62. This also means that if New Zealand has a shorter time bar than other countries, we could be disadvantaged as we would be required to provide tax relief under our treaties, but would not be able to make tax positive adjustments in respect of those same years. In particular, Australia has a 7 year time bar for transfer pricing so New Zealand must provide up to 7 years of tax relief to Australian businesses, whereas we can only currently go back 4 years when adjusting the transfer prices of taxpayers that owe tax to New Zealand. Our DTA with Australia provides that both countries are allowed to propose transfer pricing adjustments up to 7 years after tax returns have been filed.

63. Having a longer time bar for transfer pricing does not preclude having shorter time bars in other areas where there is less risk or complexity. The discussion document noted that Australia and Canada both have 7 year time bars for transfer pricing even though their standard time bars are 4 years. Australia also has a shorter 2 year time bar for individuals and small businesses. Therefore we consider that having a longer time bar for transfer pricing is not inconsistent with Inland Revenue’s Business Transformation objectives.

64. Many submitters have suggested that as an alternative to extending the time bar, Inland Revenue should look to better resource its transfer pricing team. Inland Revenue may need to recruit a larger team of transfer pricing specialists to investigate transfer pricing issues. However, we do not agree that additional transfer pricing specialists would eliminate the need for a longer time bar. The longer time bar will only be necessary in a small number of complex cases. These cases require commissioning of overseas experts and multiple rounds of site visits, interviews and negotiations with taxpayers. These tasks are best performed by a small project team working in a logical sequence. Trying to use a larger team to simultaneously perform each task would be unlikely to shorten the overall time needed to resolve the dispute. Finally, it can be difficult for Inland Revenue to recruit or retain the relevant expertise as there is high global demand for transfer pricing experts.

Shifting the burden of proof from Inland Revenue to the taxpayer

65. The discussion document proposed shifting the burden of proof onto the taxpayer for providing evidence that their related party dealings are consistent with those that would be agreed by third parties operating at arm’s length.

66. This proposal is consistent with the fact that burden of proof is already on the taxpayer for other tax matters. Self-assessment is at the heart of how New Zealand’s tax system works and helps encourage taxpayers to comply with the law and get it right from the start rather than having to subsequently amend their tax position as a result of an Inland Revenue investigation.

67. Four submitters (CTG, KPMG, CA ANZ, Russell McVeagh) argued that the burden of proof for transfer pricing should remain with Inland Revenue. The main arguments raised by submitters were:

• Shifting the burden of proof will increase compliance costs, especially in conjunction with the other transfer pricing proposals.
• Inland Revenue should provide more guidance on what transfer pricing documentation they expect to be prepared (or explicitly mandate for transfer pricing documentation to be prepared), rather than shift the burden of proof.
• The current ability for Inland Revenue to shift the burden of proof to the taxpayer when a transfer pricing position is undocumented is an effective way to encourage documentation.
• Inland Revenue may have better information on comparables than the taxpayer and should not be able to use secret information (that it cannot share with the taxpayer) to adjust a taxpayer’s transfer pricing position.

68. The burden of proof is on the taxpayer for transfer pricing matters in most OECD and G20 countries, including Australia. This means most multinationals already prepare transfer pricing documentation that satisfies the burden of proof for other countries. For this reason, the additional compliance costs that would be imposed under New Zealand’s transfer pricing rules from shifting the burden of proof onto taxpayers is not expected to be substantial.

69. When New Zealand’s transfer pricing rules were introduced in 1995, most multinational transactions in New Zealand closely resembled easily observable market transactions. Two decades later, related party transactions and transfer pricing practices have become a lot more complex, specialised and sophisticated.

70. Multinationals typically have better information than Inland Revenue on market prices in their industry and on their supply chains. For this reason they are better placed to identify a relevant uncontrolled comparable and apply the arm’s length principle.

71. One submitter (KPMG) suggested the legislation should explicitly mandate the type of transfer pricing documentation that taxpayers have to prepare as an alternative to shifting the burden of proof. Others (EY, PwC, AmCham) suggested that Inland Revenue should prepare additional guidance on what documentation would be required to satisfy the burden of proof.

72. We consider that taxpayers are best placed to consider the amount of documentation or evidence that is required to demonstrate compliance (as this will vary based on the tax effect or materiality of the transaction). Imposing a minimum standard for documentation could impose additional compliance costs in respect of lower-risk transactions (which may require no or little documentation) and may not lead to adequate documentation for higher-risk transactions (which should require a higher standard to discharge the burden of proof). The OECD has recently issued extensive international guidance on transfer pricing documentation, which New Zealand endorses, and Inland Revenue has issued some short supplementary guidance as well.

73. Three submitters (CTG, KPMG, Russell McVeagh) raised concerns that Inland Revenue could potentially use information that it held on comparable transactions to adjust a taxpayer’s transfer pricing position and then not share this information with the taxpayer on the ground that it was tax secret. They considered this was a reason why the burden of proof should remain with Inland Revenue (either more generally, or just in this particular scenario).

74. However, in the 22 years since the transfer pricing rules were introduced, Inland Revenue has never used a secret comparable to adjust a taxpayer’s transfer pricing position. In practice, because New Zealand is a small market, Inland Revenue mainly sources comparable information from commercial databases that can also be purchased/accessed by taxpayers (as opposed to its own tax information). In any case, if Inland Revenue did in fact ever make an adjustment based on information that was not accessible to the taxpayer, it would be able to anonymise the relevant information in order to share the basis for the adjustment with the affected taxpayer.

The test for reconstructing a transfer pricing arrangement

75. Consistent with Australia’s rules and the OECD’s transfer pricing guidelines, the discussion document proposed providing Inland Revenue with a power to reconstruct transfer pricing arrangements which are not commercially rational because they include unrealistic terms that third parties would not be willing to agree to.

76. Two submitters (TPEQ and Russell McVeagh) argued that New Zealand should not include a specific reconstruction rule in our transfer rules as New Zealand’s existing general anti-avoidance rule already allows the Commissioner to challenge and reconstruct tax avoidance arrangements. We note that the general anti-avoidance rule would be more difficult to apply as it requires an explicit purpose of tax avoidance, whereas the proposed rule (and the OECD’s transfer pricing guidelines) would not. Therefore, we consider that a specific transfer pricing reconstruction rule is still necessary.

77. Eight of the 15 submitters (CTG, KPMG, TPEQ, CA ANZ, EY, PwC, AMP (NZ), Deloitte) raised concerns about the potentially broad scope of the proposed reconstruction rule and submitted that the proposed reconstruction rule should only apply in the “exceptional circumstances” described in the OECD’s transfer pricing guidelines.

78. The discussion document proposed that we adopt Australia’s provision which allows transfer pricing arrangements to be reconstructed when: “independent entities dealing wholly independently with one another in comparable circumstances would not have entered into the actual commercial or financial relations.”

79. Australia’s rule was developed in 2012 before the OECD’s new transfer pricing guidelines were published in 2015. The OECD guidelines suggest that tax authorities should only reconstruct those arrangements that: “differ from those which would have been adopted by independent enterprises behaving in a commercially rational manner in comparable circumstances, thereby preventing determination of a price that would be acceptable to both of the parties...”

80. Although Australia’s test is intended to be applied on a consistent basis to the test in the OECD’s transfer pricing guidelines, Australia’s wording is potentially broader, which creates uncertainty for taxpayers. Unlike the OECD’s transfer pricing guidelines, it doesn’t explicitly specify that the original arrangement should be commercially irrational to the extent that third parties wouldn’t be able to reach such an agreement.

81. New Zealand endorses the OECD’s transfer pricing guidelines. Therefore in order to provide more certainty for taxpayers, we recommend that New Zealand’s legislative test for reconstructing an arrangement should be based on the corresponding test in the OECD’s transfer pricing guidelines.

Administrative measures

Summary of measures

82. The discussion document proposed measures to help Inland Revenue investigate and assess un-cooperative multinationals. These included the following proposals:

• If a large multinational (over EUR €750m worldwide revenues) does not cooperate with Inland Revenue, then Inland Revenue may more readily issue a notice of proposed adjustment, and any subsequent documents under the disputes process, based on the information available to Inland Revenue at the time.
• Any disputed tax must be paid by a large multinational during the disputes process, rather than at the end of the final Court case. This only applies in respect of disputes over transfer pricing, the amount of New Zealand-sourced income, and the application of a DTA.
• Tax payable by any member of a large multinational can be collected from any wholly-owned group member (or the related New Zealand entity in case of the new PE avoidance rule).
• Inland Revenue will be empowered to collect more information from large multinationals, including information about its various non-resident group members. If the multinational does not provide the information, then penalties may be payable and Inland Revenue will be expressly authorised to assess the taxpayer based on the information available to it.

General reaction

83. Submitters were generally accepting of some form of administrative measures in relation to uncooperative multinationals.

84. Submitters did note that any legislation needed to make it clear when the measures applied, and there needed to be sufficient safeguards (both in terms of legislative requirements and Inland Revenue’s internal processes) to ensure the measures were not misapplied (KPMG, PwC, TPEQ, AMP (NZ), Russell McVeagh, NFTC, DEG). Some submitters also stated that the rules should be narrowly targeted (CTG, NZLS, DEG, NFTC, PwC), while others called for an increase in Inland Revenue’s resources to help taxpayers comply with the new BEPS measures (EY, CA ANZ, PwC). We note these concerns and will consider them when we design the detail of the measures.

Earlier payment of disputed tax

85. Submitters strongly opposed the proposal to advance the time at which multinationals must pay any tax in dispute (KPMG, EY, Russell McVeagh, NFTC, DEG, NZLS, AMP (NZ), CTG). Some submitters argued that the proposal to make large multinationals pay disputed tax upfront was unnecessary. This was because Inland Revenue already charges “use of money interest” on tax owing, which provides a strong incentive for multinationals to pay any tax that is in dispute. In addition, the “use of money interest” paid by Inland Revenue to taxpayers on tax that is paid, but not legally owed, is significantly below market rates. This could unduly pressure taxpayers to settle.

86. We agree with the views of submitters. The proposal was based on similar rules in Australia’s and the UK’s diverted profits taxes, and was intended to disincentivise taxpayers from deliberately prolonging disputes. However in light of the current “use of money interest” rates, we consider the rule is not necessary, and may instead unduly pressure taxpayers to settle. In addition such a rule appears better suited to a diverted profits tax regime, which is intended to incentivise taxpayers to pay the correct amount under the ordinary income tax rules. It seems less appropriate to include it in the ordinary income tax rules themselves.

87. Accordingly we recommend that you do not proceed with the proposal.

Collection of tax from a local subsidiary

88. Some submitters opposed the proposal to allow tax owing by a non-resident to be collected from a wholly owned subsidiary in New Zealand (or the related entity where the proposed PE avoidance rule applies) (CTG, PwC, Russell McVeagh). They questioned the practical need for such a rule, noted that it undermined the separate legal identity of corporate subsidiaries, and were concerned that it could cause risk assessment and banking covenant issues for lenders. One submitter (PwC) noted that if the rule was to proceed, it should apply only where the non-resident did not pay.

89. We consider that such a rule is useful to allow the collection of tax from non-residents with no direct presence in New Zealand. We also think it is reasonable to apply the rule where the non-resident and the New Zealand subsidiary are part of the same wholly-owned group, as they are part of a single economic entity.

90. Accordingly we recommend that the rule be retained. However we agree that the rule should only apply if the non-resident fails to pay the tax itself, and if the non-resident and the New Zealand entity are part of the same wholly-owned group. This should mitigate some of the submitters’ other concerns about risk assessment and banking covenant issues.

Collection of information

91. A majority of submitters considered that the information collection powers should not proceed (CTG, Russell McVeagh, PwC, NZLS, NFTC, AMP (NZ), AmCham, DEG). Submitters variously considered that the rules were unnecessary in light of enhanced international information sharing protocols (such as country-by-country reporting), would be unworkable in practice, and unfairly penalised the New Zealand resident, who may not be able to get the information from their multinational group members. Some submitters also considered the issue should be addressed by Inland Revenue improving its relationship with other tax authorities (AMP (NZ), Russell McVeagh, AmCham, DEG, NZLS).

92. Submitters raised further concerns about the new civil penalty of up to $100,000 for failing to provide requested information (which replaces the current $12,000 maximum criminal penalty) (CTG, CA ANZ, Russell McVeagh, PwC, NFTC, NZLS). Submitters considered the penalty should not be increased, given that the New Zealand subsidiary may not control the relevant information. If the penalty was to apply, they considered that only a court should be able to apply it. Finally they considered that directors and company employees should not be liable for the penalty personally.

93. We recommend that the information gathering proposals proceed (with some changes), notwithstanding submitters’ views. In our view it is unacceptable for Inland Revenue investigations to be frustrated because a multinational group fails to provide information that is under its control.

94. We do not think the New Zealand subsidiary’s difficulty in obtaining the information is a valid objection to the proposals. The New Zealand subsidiary is simply part of the multinational’s economic group. Therefore any consequences suffered by the New Zealand subsidiary are economically borne by the wider group and its shareholders. Accordingly our proposed measures effectively make the entity which controls the information liable for the economic consequences of its failure to provide that information.

95. Further, the inability of the New Zealand subsidiary to legally require the information to be provided is the reason the proposed measures are necessary in the first place. There must be incentives for the multinational group to provide Inland Revenue with the required information in the absence of any legal ability to compel its provision. This means that any failure to provide the necessary information must be to the multinational’s detriment, not Inland Revenue’s. Otherwise multinationals will be incentivised not to provide such information.

96. In relation to submitters’ other points:

• The information shared under new international protocols, such as country by country reporting, is at a more general level and will not be sufficient for Inland Revenue to assess particular taxpayers. In fact, tax authorities are explicitly prohibited from using country-by-country reports as a basis for assessing taxpayers.
• We are committed to improving our relationship with other tax authorities, but this will not practically address the current issue. The required information may not be held by the other tax authority, or it may be slow to obtain it.
• The impracticality of a New Zealand subsidiary obtaining the required information from another group member seems to be caused by the internal processes and priorities of the multinational group. This impracticality may ameliorate once the inability to obtain the information starts having negative consequences for the group. In the event that it does not, it should be the multinational that bears the negative consequences arising from its own processes and priorities, rather than Inland Revenue.
• Inland Revenue should be able to charge the penalty for not providing information, without requiring court approval. This is the normal position for civil penalties under the Tax Administration Act 1994 and we do not see why an exception should be made here. Further, such a requirement would also significantly undermine the practicality of imposing the penalty, and it is difficult to see what additional benefit it would provide. We also note that Australia has recently introduced a similar penalty, with a maximum amount payable of $450,000. However we agree that taxpayers should be able to challenge the penalty once it is applied (in common with other similar penalties). We will ensure this is provided for when we design the detail of the measures.
• We agree that the penalty should not apply to directors or employees of a company. We will clarify this when the detail of the rule is designed.

97. We also want to ensure that the proposed information gathering powers and penalties are used by Inland Revenue in a reasonable manner. Accordingly we will consider ways to ensure this is the case when we design the rules in more detail.

Application date and grandparenting APAs

Application date

98. The planned application date for these measures is the income year starting on or after 1 July 2018.

99. At the time the discussion document was released for public consultation, the planned application date was not publicly known.[5] For this reason, Inland Revenue has not received any submissions on the 1 July 2018 application date. However, some submitters expected the Government to seek an early application date and argued that it would be better to allow taxpayers time to consider the proposals and rearrange their affairs if necessary (PwC and CTG). PwC argued that the application date for these proposals should be no earlier than the first income year after 31 March 2019.

100. Cabinet has already noted that the reforms are expected to apply from income years beginning on or after 1 July 2018 (CAB-17-MIN-0164 refers). This is based on the expectation that the legislation will be progressed to enactment before this date.

101. We expect to receive more submissions on, and opposition to, the application date once the public becomes aware it is proposed to be 1 July 2018.

Grandparenting APAs

102. A taxpayer is able to apply for an advance pricing agreement (APA), which is essentially a binding ruling confirming that the taxpayer’s planned transfer pricing positions are compliant with the transfer pricing rules for up to five years. Some submitters argued that existing APAs should be grandparented and allowed to run their course (PwC and CTG). This would reduce any uncertainty taxpayers may face in light of the changing environment. Without grandparenting, taxpayers may be disincentivised to engage with Inland Revenue in the interim as the high cost of obtaining an APA proportionally increases if the length of the APA is shortened.

103. We agree with this submission. There is a high cost and a rigorous process involved in obtaining an APA and it would be unfair if the new proposals rescinded APAs issued before the 1 July 2018 application date – especially considering APAs only run for five years.

104. We therefore recommend grandparenting all APAs existing prior to the 1 July 2018 application date.

Appendix One: List of submitters

Abbreviation	Full name	Description	IL[6]
AmCham	The American Chamber of Commerce in New Zealand	AmCham is a New Zealand business organisation which promotes two-way trade and investment relationships primarily between New Zealand and the United States, but also within the Asia-Pacific region.	✔
AMP (NZ)	AMP Capital New Zealand	AMP is a specialist investment manager that manages a number of Portfolio Investment Entity funds, as well as private equity investments.
CA ANZ	Chartered Accountants Australia and New Zealand	Chartered Accountants Australia and New Zealand is the incorporated body representing the Institutes of Chartered Accountants in Australia and New Zealand. CA ANZ represents over 100,000 members in Australia, New Zealand, and overseas.	✔
CTG	Corporate Taxpayers Group	CTG represents 40 large New Zealand corporates and also include tax advisors from Deloitte, Russell McVeagh, and OliverShaw.	✔
DEG	Digital Economy Group	DEG is an informal coalition of leading US and non-US software, information/content, social networking, and e-commerce companies that provide goods or services through digital and non-digital means.
Deloitte	Deloitte	Deloitte New Zealand is an accounting firm providing audit, tax, consulting, enterprise risk, and financial advisory services.	✔
EY	Ernst & Young	EY New Zealand is a professional services firm which specialises in assurance, tax, transaction and advisory services.	✔
KPMG	KPMG	KPMG refers to the New Zealand arm of KPMG International – the global network of professional firms providing audit, tax, and advisory services.	✔
NFTC	National Foreign Trade Council	NFTC is an association of approximately 250 United States business enterprises engaged in all aspects of international trade and investment.
NZCTU	New Zealand Council of Trade Unions Te Kauae Kaimahi	NZCTU is one of the largest democratic organisations in New Zealand. NZCT is made up of 30 unions and has 320,000 members.	✔
NZLS	New Zealand Law Society	NZLS controls and regulates the practice of the law profession in New Zealand. The NZLS also assists and promotes law reform for the purpose of upholding the rule of law and the administration of justice.	✔
Oxfam	Oxfam New Zealand	Oxfam is a world-wide development organisation that mobilises the power of people against poverty. Oxfam NZ is the New Zealand arm of the global organisation.	✔
PwC	PwC	PwC refers to the New Zealand arm of PwC International – a multinational professional services network which advises on tax.	✔
Russell McVeagh	Russell McVeagh	Russell McVeagh is a New Zealand commercial law firm with offices in Auckland and Wellington.	✔
TPEQ	TP Equilibrium | AustralAsia	TPEQ is a boutique transfer pricing advisory firm which covers numerous industries for both the Australian and New Zealand markets.	✔

1 A full list of all the submitters, together with a brief description, is included in the appendix to this report.

2 The EUR €750m threshold has been chosen to align application of the proposed rule with the OECD’s threshold for requiring large multinationals to file country-by-country reports

3 The Multilateral Convention to Implement Tax Treaty Related Measures To Prevent Base Erosion and Profit Shifting. The MLI is a multilateral convention which is intended to prevent DTAs from being used to facilitate cross-border tax avoidance. The MLI amends a large number of each signatory’s DTAs at once, and so implements the OECD’s recommended DTA changes much faster than a succession of bilateral negotiations could. New Zealand signed the MLI on 7 June 2017.

4 This is because the general anti-avoidance rule applies only if an arrangement both uses the relevant provisions in a way that is outside of Parliament’s contemplation (the Parliamentary contemplation test) and has a more than merely incidental purpose of tax avoidance. The proposed PE avoidance rule would not incorporate the Parliamentary contemplation test. Instead it would only require that an arrangement had a more than merely incidental purpose of tax avoidance.

5 [No footnote content in report]

6 Submissions also received on BEPS – strengthening our interest limitation rules