Interest limitation rules

OVERVIEW

The use of related-party debt to fund New Zealand operations is one of the simplest BEPS strategies. Interest payments are generally tax deductible in New Zealand. This means a related-party cross-border loan between a foreign parent and its New Zealand borrowing subsidiary will reduce taxable profits and therefore taxes payable in New Zealand. Of particular concern is that some firms have borrowed from their foreign parents at high interest rates resulting in large interest deductions in New Zealand.

The Bill proposes new rules requiring related-party loans between a non-resident lender and a New Zealand-resident borrower to be priced using a restricted transfer pricing approach. Under this approach, specific rules and parameters are applied to inbound related-party loans to:

determine the credit rating of New Zealand borrowers at a high risk of BEPS, which will typically be one notch below the worldwide group’s credit rating; and
remove any features not typically found in third-party debt in order to calculate (in combination with the credit rating rule) the correct amount of interest that is deductible on the debt.

In response to submitters’ concerns officials recommend a number of changes including increasing the allowable difference between the New Zealand borrower’s credit rating and the worldwide group be extended to two notches provided the New Zealand borrower is a BBB- or higher credit rating.

The Bill also contains some amendments to New Zealand’s thin capitalisation rules so they are better at protecting the New Zealand tax base from BEPS techniques. Thin capitalisation rules limit the amount of debt for which a foreign-owned subsidiary can claim deductions for interest paid. Interest deductions are generally denied to the extent the debt exceeds 60% of the subsidiary’s assets.

Currently, debt percentages determined under the thin capitalisation rules are based on an entity’s debt relative to its gross assets. A significant issue is that the current treatment of non-debt liabilities allows companies to have higher levels of debt (and therefore higher interest deductions) relative to the capital invested in a company by its shareholders. For example, at present if a company purchases some inventory on deferred payment terms, the amount of debt allowed under the thin capitalisation rules will increase (because the new inventory has increased its assets but its interest bearing debts have stayed the same). The Bill proposes to change this, so that debt percentages are based on an entity’s assets net of its “non-debt liabilities”.

The Bill also provides an exemption to the thin capitalisation rules for infrastructure projects which have little risk of BEPS activity. The exemption allows all of an infrastructure project’s third party debt to be deductible even if the debt levels exceed the normal thin capitalisation limits, provided the debt only provides recourse against the assets associated with the infrastructure project and the income arising from those assets. The purpose of this rule is to improve the competitiveness in the bidding process for Public Private Partnership (PPP) procurement contracts by allowing investors that are subject to the thin capitalisation rules to make bids on a level playing field with investors that are not subject to the thin capitalisation rules.

The Bill’s other thin capitalisation changes are as follows:

introducing a de minimis in the inbound thin capitalisation rules;
reducing the ability for companies owned by a group of non-residents to use related-party debt;
new rules for when a company can use an asset valuation for thin capitalisation purposes that is different from what is used for financial reporting purposes;
an anti-avoidance rule that applies when a taxpayer substantially repays a loan just before the end of a year to circumvent the thin capitalisation rules; and
a minor remedial to clarify how the owner-linked debt rules apply when the borrower is a trust.

GENERAL

Issue: Support for rules

Submission

(Chartered Accountants Australia and New Zealand)

Chartered Accountants Australia and New Zealand in principle supports the proposals to prevent taxpayers from using excessive interest rates and debt volumes to shift profits out of New Zealand.

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: The restricted transfer pricing rule should not proceed

Submission

(Chartered Accountants Australia and New Zealand, KPMG, New Zealand Bankers’ Association, PwC)

CA ANZ fully supports adjustments to the price of interest via the transfer pricing provisions. However we have significant concerns about the restricted transfer pricing rule that seeks to limit New Zealand interest deductions by reference to the parent company’s actual or implied credit rating. This approach conflicts with recognised tax policy principles, is at odds with the approach in other jurisdictions, causes double taxation and is commercially difficult to apply. (Chartered Accountants Australia and New Zealand)

The restricted transfer pricing rule will be imposed unilaterally by New Zealand. It has not been implemented or otherwise proposed by any other jurisdiction. The restricted transfer pricing rule is a substantial departure from generally accepted orthodox transfer pricing approaches for determining arm’s length interest rates on related-party loans.

The potential effect of such a rule needs to be carefully considered from a national welfare perspective. New Zealand adopting a BEPS position that is outside of the norm may provoke countermeasures that may adversely impact NZ multinationals operating in those jurisdictions.

Comment

Officials consider the restricted transfer pricing rule is broadly consistent with both the arm’s length test and standard transfer pricing principles. These restrictions, which are limited to borrowers with a high BEPS risk, remove much of the subjectivity and manipulability from standard transfer pricing of related party interest but are still aimed at determining an appropriate interest rate that would apply if the New Zealand borrower had borrowed from a third party.

There is now widespread acceptance by most of New Zealand’s trading partners that standard transfer pricing does not result in appropriate outcomes when applied to related party debt. The OECD, as part of its BEPS Action 4 report recommended the introduction of an earnings before interest, tax depreciation and amortisation (EBITDA)-based rule to limit interest deductions. The European Union issued a directive in July 2016 that all member states must adopt an EBITDA-based interest limitation rule by 31 December 2018. The United Kingdom and United States are also introducing similar measures.

While the restricted transfer pricing rule is a departure from the OECD (EBITDA)-based rule officials consider the New Zealand approach avoids some of the issues inherent in the OECD’s recommendation. Many of the concerns raised by submitters, such as the risk of double taxation, are also present in an EBITDA rule and it would be more difficult to introduce many of the safeguards that the restricted transfer pricing rule includes. The European Union directive also allows member to states to introduce other targeted rules against intra-group debt financing.

Officials note that Australia has now proposed a rule denying an interest deduction where the lender is a related party in a low tax country and certain other conditions are met. Officials are yet to review this rule and consider its appropriateness in the New Zealand context. For the purposes of the current debate, it is significant in demonstrating that our nearest neighbour is also concerned about deductions for related party interest expense, and does not believe that its thin capitalisation and transfer pricing rules are themselves sufficient to deal with the base erosion issue such deductions present.

Recommendation

That the submission be declined.

Issue: Existing legislative tools are sufficient

Submission

(BNZ Bank, Corporate Taxpayers Group, Powerco, PwC, Russell McVeagh)

The existing legislative tools available to Inland Revenue (for example sections BG 1 and GB 2) can be applied to target those specific situations where excess deductions are being claimed. These targeted measures are less likely to have an adverse impact on compliant taxpayers, than the proposed restrictions on the existing transfer pricing rules. (BNZ)

The proposed rules achieve nothing that could not be more appropriately achieved through the proper application and enforcement of standard transfer pricing practice. In particular, with the introduction of reconstruction provisions as part of the Bill, the need for the proposed restrictions is unclear. (Corporate Taxpayers Group)

It would be strongly preferred to have guidelines rather than legislation which help taxpayers determine an appropriate rate (we understand this is the approach of the Australian Tax Office). (Powerco)

The rules are unnecessary to address Inland Revenue’s concerns about excessive interest expenditure deductions given other amendments being proposed in the Bill. (Russell McVeagh)

Comment

Relying on existing transfer pricing and anti-avoidance rules, as suggested by submitters, will not necessarily achieve the desired policy outcomes.

The international consensus is moving away from using ordinary transfer pricing to limit interest expenses in relation to related party debt. Concerns over highly-priced related party debt were part of what was behind the OECD’s recommended interest limitation rule based on EBITDA. Interest denial could result under an EBITDA rule even if the interest expense is determined by the arm’s length standard.

The detail of the transfer pricing rules are “soft law”. They are contained in the OECD transfer pricing guidelines to support the application of tax treaties. Most countries rely on them to solve transfer pricing issues even in the absence of a treaty. The transfer pricing guidelines take the form of guidance rather than set rules. Officials consider that, once amended as proposed in the Bill, the transfer pricing rules will work well for non-debt items. However, because of the significant BEPS risks associated with related-party interest payments, we consider that the rule for such payments needs to be stronger, less subjective, and less open to interpretation. We note, for example, the Australian Taxation Office has stated that the recent Chevron case in Australia had cost them $10 million in external experts (not taking into account the cost of their own staff) even though it involved related-party interest of about 9% when the parent had raised external funding at a rate of about 2%.

In addition, transfer pricing does not adequately take account of the fact that related-party debt financing is fundamentally different to third-party debt financing. For example, subordinated debt is less likely to be repaid compared to senior debt, and so carries a higher interest rate. This is appropriate in a third-party context: the higher interest rate compensates for the higher risk. However, in a related-party context, debt and equity are highly substitutable. The riskiness of a parent’s investment in a subsidiary does not change whether it invests through equity (which would generate no deduction) or debt. We do not consider that related-party debt being subordinate to other debt should justify a higher interest rate.

Recommendation

That the submission be declined.

Issue: Restricted transfer pricing rule may not meet the internationally accepted arm’s length principle

Submission

(ANZ, ASB Bank, BNZ, Chapman Tripp, Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, Deloitte, EY, KPMG, New Zealand Bankers’ Association, OliverShaw, Powerco, PwC, Russell McVeagh)

The proposed restricted transfer pricing rules will require taxpayers to alter the terms and conditions of certain related party lending, or the circumstances of the borrower itself, in determining an arm’s length price. Requiring taxpayers to step away from the actual terms and circumstances of a loan in determining the interest rate, raise a significant risk that the resulting pricing may not meet the internationally accepted arm’s length principle, which the overseas revenue authorities will generally be applying. This may result in the respective tax authorities being unable to agree on the appropriate pricing of the transaction, and therefore significantly increase the likelihood that taxpayers will face double taxation.

The interest limitation rule is inconsistent with our double tax treaties – OECD Model Convention Article 7 (as it was prior to the 2010 update that New Zealand has not adopted and Article 24 Non-discrimination). The result would be that the legislation would be overridden by the treaties and these proposals will not apply as the Treaties apply an arm’s length test.

Comment

We do not agree with the argument that the restricted transfer pricing rule is systematically inconsistent with the arm’s length standard. On the contrary, we consider the rule will generally be consistent with the standard because of the transfer pricing concept of “implicit parental support”. “Implicit parental support” is the notion that a foreign parent will stand behind a New Zealand subsidiary in the event of a default. That is, multinational groups generally do not let their local subsidiaries go under. “Implicit parental support” is a significant factor in transfer pricing analysis because it hypothesises that, as a commercial matter, it would affect what rate a third party lender would charge the New Zealand subsidiary and what that subsidiary would be prepared to pay. Accordingly, the credit rating of the foreign parent is a strong element in determining the credit rating of the New Zealand subsidiary.

Inland Revenue administers transfer pricing having regard to the concept of implicit parental support but some taxpayers do dispute it.

We acknowledge that there may be cases when the restricted transfer pricing rule would not produce an arm’s length interest rate because, for instance, the New Zealand subsidiary is in a completely different line of business from the rest of the multinational group and has a different risk profile. Nevertheless, we do not accept that in these cases the restricted transfer pricing rule would frequently result in double taxation. This is partly because the cap is not arbitrary (unlike the EBITDA test recommended by the OECD and rejected by submitters in previous consultation). Moreover, in our view, the shift in the international consensus makes it less clear that our treaty partners (especially Australia) would dispute the result of the restricted transfer pricing rule under a treaty.

Recommendation

That the submission be declined.

Issue: Rebuttable safe harbour

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, OliverShaw, PwC)

Many of the problems with the presumed credit rating rules could be more simply addressed by legislating the highest credit rating of a member of the borrower’s worldwide group as a rebuttable factor in determining the credit rating of the New Zealand borrower. Rules that force a presumed credit rating will have neither commerciality nor international acceptance.

Comment

Much of the problem with applying standard transfer pricing principles to related party debt is the subjectivity of the analysis which results in highly qualified experts on both sides debating the accuracy of opinion and not facts. Given the large size of some related party loans even small differences in an interest rate can make a large difference to the amount of deductible interest.

The problem with applying the restricted transfer pricing rule as a safe harbour, as the submitters has suggested, is it will result in the rule being ineffective as all taxpayers would argue they had commercial justification why the rule could not be applied to them.

Instead the rule has been designed with a number of exceptions to reflect commercial considerations. For example:

The rule will not apply to taxpayers with less than $10 million of related party cross-border loans
The rule will not apply to taxpayers that have less than 40% debt unless they are borrowing from a low tax jurisdiction
Recommendations elsewhere in this report will allow a high BEPS risk borrower to base their credit rating on third party debt. Related party debt can be up to four times the size of third party debt so only 20% of debt would have to be from a third party.

Officials are not aware, and submitters have not provided, any further objective measures that could be used to justify a departure from the proposed linkage to the parent’s credit rating for a high BEPS risk borrower with an identifiable parent.

Recommendation

That the submission be declined.

Issue: Rationale used to justify restricted transfer pricing approach

Submission

(EY)

The rationale used to justify the restricted transfer pricing rule, based on the parent company cost of funding is flawed.

The parent cost of funding does not typically represent the cost of borrowing to a standalone borrower in a group given the different asset and hence credit profiles. A third party would only lend to a standalone New Zealand subsidiary at the parent’s cost of borrowing where the New Zealand subsidiary is considered a core investment to the group (that is, full implicit support equates to an explicit guarantee) or the parent company gives an explicit guarantee.

In our experience, few New Zealand subsidiaries are considered core or highly strategic to a group and a guarantee from a stronger party comes at a cost to the borrower which takes the total cost of borrowing closer to a company’s standalone situation.

Comment

The restricted transfer pricing rule recognises that a New Zealand borrower (other than an insuring or lending person) will frequently have a slightly lower credit rating than their foreign parent. This is the reason the New Zealand borrower is allowed to be one notch lower than the group rating and why officials are recommending in this report that this be extended to two notches for borrower that have a rating of BBB- or higher.

While a New Zealand borrower, if it was a standalone entity, may have a credit rating that was lower than their foreign parent this would only be true if implicit parental support was not included. The international consensus, as demonstrated by the 2017 OECD transfer pricing guidelines, is that implicit parental support must be included. The relevant extracts of these guidelines were included in the Commentary to the Bill.

The restricted transfer pricing rule is attempting to codify implicit parental support and remove the subjectivity and manipulability inherent in the current approach to transfer pricing of related party debt.

Recommendation

That the submission be declined.

Issue: Group credit rating approach should not proceed

Submission

(KPMG)

If the restricted transfer pricing rule proceeds the group credit rating approach should not be used. Instead each New Zealand borrower should be able to apply either the standalone credit rating approach in proposed section GC 16(7) or the restricted credit rating approach in proposed section GC 16(8).

Using the ultimate parent’s credit rating as the starting point to derive an interest rate on New Zealand inbound debt is not economically correct. This implies that the New Zealand subsidiary has a similar business profile to that of the parent in all cases, which is often not the case with global multinationals. Typically, the New Zealand operations of foreign multinationals are often several multiples smaller and will typically comprise a single function or asset or, at the very most, a less diverse set of functions or assets when compared to the ultimate parent. Often entities within the group have much higher risk profiles than others, based on these features.

Instead, a transfer pricing approach for cross-border related party loans which starts with the borrower’s credit rating is more in line with the arm’s length principle. Not only does it have regard to the credit quality of the specific borrower, but it provides flexibility to notch the borrower’s standalone credit rating upwards to reflect the specific circumstances of that company and its position in relation to the wider group.

Comment

As discussed in other items in this report, officials accept that a New Zealand borrower, if they were truly standalone, may have a lower credit rating than their foreign parent. However, this does not acknowledge the implicit parental support provided by being a member of the worldwide group. This support has been internationally accepted, for example in the 2017 OECD transfer pricing guidelines. It is often the degree of this implicit support, and the effect that has on an interest rate, that is the subject of most of the disagreement between taxpayers and Inland Revenue. The restricted transfer pricing rule is designed to remove this subjectivity by inserting non-manipulable objective rules that reflect the level of support a worldwide group provides to their New Zealand subsidiary.

The restricted credit rating in GC 16(8) has been introduced for borrowers that are controlled by a co-ordinated group as there is no identifiable parent from which to benchmark an appropriate credit rating. This approach removes some of the problems with applying general transfer pricing rules to related party debt but would not resolve disagreements over the level of implicit support if it were applied more widely to borrowers with an identifiable parent.

Recommendation

That the submission be declined.

Issue: Defer until further OECD research completed

Submission

(Corporate Taxpayers Group)

It is premature for New Zealand to impose a restricted transfer pricing rule and it would be advisable to defer any action until we see the outcome of the OECD’s research and can review any multilateral recommendations.

Comment

The restricted transfer pricing rule has been developed in response to the OECD recommendations under Action 4 and attempts to achieve the purpose – to limit excessive related party interest deductions. Introducing the restricted transfer pricing rule at this time aligns with the Government’s other responses to BEPS included in this Bill. Officials will continue to monitor future developments whether further legislative or other responses are necessary but this does not negate the importance of the restricted transfer pricing rule.

Recommendation

That the submission be declined.

Issue: Inconsistency with the approach taken in Australia

Submission

(PwC)

The Government has noted that it is important that New Zealand’s transfer pricing rules are aligned with Australia’s to reduce compliance costs and the risk of double tax with one of our key trading partners. We agree that this is of critical importance, and are concerned that adopting the restricted transfer pricing (RTP) rule would be inconsistent with the approach taken by the ATO.

Consistent with international consensus, Australia adopts the arm’s length principle in its transfer pricing legislation and its approach to related party debt pricing. The ATO has supplemented the application of the arm’s length principle with a Practical Compliance Guideline (PCG 2017/4), which sets out a risk assessment framework for related party debt.

This approach is much more preferable to the RTP rule in a number of important respects:

The approach is not embedded in legislation, but rather is a supplementary guide that allows taxpayers to assess risk and enter into transactions accordingly;
Using appropriate third party comparables (including traceable third party debt) as a starting point is a key features the PCG;
The risk factors are similar to those identified by the Government, but are less stringent (e.g. leverage ratios of 60% rather than 40% representing high risk) and are able to be updated, as and when appropriate to reflect changing market conditions;
Credit ratings are not prescribed; and
The approach applies to both inbound and outbound debt.

Comment

The approach taken by Australia is, as noted by the submitter, broadly similar to New Zealand with the main difference being Australia’s approach is a risk assessment framework rather than legislation. The issue with the Australian approach is that if a taxpayer is brought to the ATO’s attention by failing their framework the ATO is still required to determine an appropriate price under standard transfer pricing rules – which have been internationally acknowledged as not being sufficient when applied to related party debt. While the Australian approach is likely to reduce the number of disputes, by taxpayers reducing their risk factors (as officials expect will also occur in New Zealand), for the disputes that do occur the PCG will not assist in reducing the uncertainty, cost and complexity of these disputes for either taxpayers or the ATO.

Some of the submitters’ other points, for example the use of third party comparables and the 40% debt percentage are addressed elsewhere in this report as they are not specific to a New Zealand and Australia comparison.

New Zealand officials have discussed the restricted transfer pricing rule with Australian officials and are confident that the two approaches are sufficiently similar that double taxation is highly unlikely to occur. As noted elsewhere in this report if this were to occur the Mutual Agreement Procedure will continue to be available for these limited number of cases.

Recommendation

That the submission be declined.

Issue: Withholding Tax

Submission

(BNZ Bank, KPMG, PwC)

There were varying submissions on this point.

As proposed in the Bill, New Zealand withholding tax (or Approved Issuer Levy) would continue to apply to the actual interest payments made, regardless of the amount of interest deemed to be deductible under the restricted transfer pricing rules. This position appears to be inconsistent – either the payment is interest, in which case it’s deductible, or it is something else and should be taxed accordingly.

A conceivable argument is that the amount by which the interest payment exceeds the deductible interest under the restricted transfer pricing rules is a transfer of value and is therefore a dividend. If this view is adopted, it is more appropriate for withholding tax to apply on the basis that the deduction that is available under the restricted transfer pricing rule is interest and should be subject to New Zealand tax (or levy) as appropriate. The excess interest payment is a dividend.

Taxpayers who have been denied a deduction for the interest payable under a financial arrangement ought to be able to treat that excess amount as a dividend, including attaching imputation credits and calculating withholding tax due under the dividend withholding tax rules. (BNZ)

We support the extension of the non-deductible treatment applied to interest denials under the current thin capitalisation rules. That is, any non-permissible (i.e. non-deductible) interest under the restricted transfer pricing rule should still constitute interest for non-resident withholding tax (NRWT) purposes and not a deemed dividend. (KPMG)

There is no reduction in NRWT where interest deductibility is denied. If an amount of interest is not deductible, and it is also subject to NRWT on the full interest rate, New Zealand is overtaking the interest. This outcome is not appropriate and will be detrimental to New Zealand’s reputation as a favourable place to invest. We strongly submit that NRWT is not imposed for interest deductions that are denied under the restricted transfer pricing rule. (PwC)

Comment

Related party interest payments are already subject to the transfer pricing rules and thin capitalisation rules therefore interest denial is already a feature of the existing rules. The restricted transfer pricing rules supplement the existing rules so that an adjustment may be required more frequently or to a larger amount but the denial of interest deductions in itself is not a new concept. The Income Tax Act 2007 already covers this issue in section GC 11 and GC 12 and officials consider this is the correct position.

Recommendation

That officials’ comments be noted.

Issue: Application date drafting

Submission

(EY, PwC)

Clauses 35(8) and 37(2) state that the interest limitation rules are to apply “on and after the first balance date of the person…on or after 1 July 2018, for a financial arrangement that the person enters before the first balance date…”.

For example, assuming a 31 December balance date and a loan entered into on 1 January 2017, the interest limitation rules would apply on 31 December 2018 and thereafter. It would not seem to be a policy intent that the interest limitation rules apply for the last day of an income year and thereafter.

Consistent with the broader changes to the transfer pricing rules, the application date proposed in clauses 35(7) and 36(6) is income years beginning on or after 1 July 2018. Therefore in the example above, the rules would apply from 1 January 2019 for the whole of the 2020 income year. (EY)

The term “balance date” is not defined and drafting is complicated. Drafting should refer to income years as per other sections dealing with enactment dates, or balance date defined to mean balance date for tax purposes. (PwC)

Comment

It was not intended that the interest limitation rules would apply to an existing loan from the last day of the income year finishing on or after 1 July 2018 and instead they should apply from the first day of the income year starting on or after 1 July 2018. Officials recommend drafting changes to these application clauses to ensure the rule apply correctly.

Recommendation

That the submission be accepted.

Issue: Restricted transfer pricing de minimis

Submission

(Chartered Accountants Australia and New Zealand)

We support the inclusion of a de minimis but submit that it should be increased from $10 million to $20 million. A de minimis should be set at a level where interest deductions are material and the associated compliance costs are warranted.

Comment

There are two de minimis thresholds in the proposed restricted transfer pricing rules in section GC 16(1)(a) for the credit rating adjustment and GC 18(1)(a) for the loan feature adjustment. Both of these are set at $10 million of cross-border related loans.

These thresholds are consistent with the existing administrative guidance issued by Inland Revenue in relation to transfer pricing of related party finance costs which endeavours to strike a balance between protecting the tax base and containing compliance costs. This guidance includes that “[F]or all loans in excess of the $10m guideline above, we expect far more science and benchmarking to support interest rates applied”.

This guidance is available at: http://www.ird.govt.nz/transfer-pricing/practice/transfer-pricing-practice-financing-costs.html

The $10 million threshold for administrative guidance was only set on 1 July 2014 and there has not been significant inflation since that time. Officials consider a $10 million de minimis appropriately balances compliance costs with the risk of excessive interest rates materially affecting taxable income.

Recommendation

That the submission be declined.

Issue: Grandparenting existing loans

Submission

(Chartered Accountants Australia and New Zealand, KPMG)

A grandparenting provision should apply to existing related party loans. A limited grandparenting provision is proposed for the worldwide debt threshold for non-residents acting together. Barring this exclusion, once the restricted transfer pricing rule takes effect it will apply to existing related party cross border financial arrangements. The exclusion of a grandparenting provision is contrary to stated policy on prospective and retrospective tax law changes and grandparenting.

All cross border related party loans in place as at the date of introduction of the Bill should be grandparented for a maximum period of five years from the date of the Bill. (KPMG)

Comment

The restricted transfer pricing rule is proposed to apply prospectively to income years starting on or after 1 July 2018. Although this will also apply to arrangements entered into before that date it will not change any tax positions for prior periods. The restricted transfer pricing rule will ensure that the price and features of related party debt is appropriate and it therefore is reasonable to apply this to borrowers that are a high BEPS risk on a prospective basis.

Recommendation

That the submission be declined.

Issue: Mutual Agreement Procedure

Submission

(Corporate Taxpayers Group)

In the event taxpayers are subject to double taxation they may seek resolution under our double tax agreements (DTAs) through the Mutual Agreement Procedure (MAP). Where a New Zealand taxpayer invokes the MAP provisions, Inland Revenue will bear the onus of proving to the corresponding jurisdiction that any transfer pricing adjustment ultimately made correctly applies the arm’s length principle. Accordingly, rather than the intended simplification suggested in the commentary to the Bill, diverting from the arm’s length principle will render this process more complicated and resource intensive, if not impossible to resolve double taxation imposed on taxpayers.

Comment

As noted by the submitter the Mutual Agreement Procedure will be available for any situations where New Zealand and other revenue jurisdictions do not arrive at the same price for related party debt. Officials do not expect this process to be frequently utilised as they expect taxpayers will remove unnecessary features and other terms that increase the interest rate on this debt so that the borrower is no longer a high BEPS risk or even where they are a high BEPS risk the restricted transfer pricing rules arrive at the same price as under standard transfer pricing principles. In this circumstance no risk of double taxation arises.

Recommendation

That the submission be noted.

Issue: Application to thin cap groups

Submission

(Corporate Taxpayers Group)

We are strongly of the view that the ambit of the restricted transfer pricing rule be limited to transactions subject to the transfer pricing rules (which is what the Bill does). We are pleased to see that a wider application, as originally proposed in the Discussion Document and which the Group opposed, has not proceeded. In our view this is appropriate as the rule has been suggested as a solution to over-priced debt which is a transfer pricing issue.

Comment

Officials do not agree that the restricted transfer pricing rule be limited to transactions subject to the transfer pricing rules as this would create the ability for certain groups to structure to avoid the restricted transfer pricing rules by varying the rights attached to shares and thereby charge inappropriate interest rates that are not comparable to those paid by more compliant businesses. Further discussion on this point is elsewhere in this report.

Recommendation

That the submission be declined.

Issue: Second transfer pricing analysis required

Submission

(EY)

A single transfer pricing analysis is currently undertaken to justify the arm’s length conditions of a loan for both borrower and lender. As the restricted transfer pricing rule does not necessarily comply with article 9 and OECD principles, a second transfer pricing analysis will likely be required for a lender to justify the amount the lender’s country would accept as arm’s length.

Comment

A large reason for these rules is the inclusion of extra debt, terms and features within related party debt that would not be present in third party debt. The exclusion of these features will result in a price that would arise in an arm’s length situation. Where the borrower is not a high BEPS risk and does not include disregarded features in the loan only a single transfer pricing analysis will be required to determine an arm’s length price appropriate to both jurisdictions. Officials expect taxpayers will reduce the level of debt and other features so that related party debt is more comparable with what could have been borrowed from a third party so that a single transfer pricing analysis will result in an arm’s length price acceptable to New Zealand and the lender jurisdiction’s revenue authority.

Recommendation

That the submission be declined.

Issue: Asymmetric application of rules to inbound and outbound loans

Submission

(EY, PwC)

The restricted transfer pricing rule is being applied asymmetrically to inbound loans and not to outbound loans. The same transfer pricing rules should apply in New Zealand to a transaction whether a New Zealand party is a supplier or recipient of goods, services or money.

There is no rationale for treating inbound debt and outbound debt differently. Australia’s transfer pricing rules do not differentiate between inbound and outbound related party dealings, and the ATO’s practice guidance on related party funding arrangements explicitly applies to both. New Zealand’s differentiating approach will result in a lack of understanding between Inland Revenue, offshore tax authorities and affected international groups (in turn resulting in increased uncertainty and compliance costs for all).

Comment

While the primary focus of the BEPS reforms is on foreign-owned businesses, similar base protection considerations can arise where New Zealand-owned businesses have offshore operations. For this reason New Zealand’s international base protection measures (such as the thin capitalisation rules and the transfer pricing rules) apply to both foreign-owned and domestically-owned businesses.

Officials are not aware of any concerns regarding the pricing of outbound related party loans that would require a similar restricted transfer pricing approach to be applied to outbound loans. Officials will continue to monitor the transfer pricing of outbound related party loans to identify whether any issues arise in the future. Any such proposals, if they were considered to be necessary, would need to be subject to consultation under the Generic Tax Policy Process.

Recommendation

That the submission be declined.

Issue: Pricing date for existing loans

Submission

(KPMG)

Interest rates on existing cross-border related party loans should be calculated by determining the interest rate that would have applied at the time the loan was entered into, had the new restricted transfer pricing rules been in force at the time.

Comment

Officials agree that the restricted transfer pricing rules, where they apply to existing loans, should reflect the fact that the cross-border related party loan was legally entered into at an earlier date and would have been priced based on the creditworthiness and market conditions in effect at that time.

While this was always the intention, officials agree that this position could be clearer in the legislation. Officials recommend drafting changes to clarify that existing loans should be priced at the date they were entered into rather than the date the new rules apply.

Recommendation

That the submission be accepted.

Issue: Interaction with wider transfer pricing rules

Submission

(Corporate Taxpayers Group, EY, PwC)

It is problematic that the Bill also proposes to give legislative force to the OECD guidelines. Proposed section GC 6(1B) will provide for New Zealand’s transfer pricing rules to apply “consistently” with the OECD guidelines, while proposed sections GC 15 to GC 18 will provide for a methodology in respect of certain related party debt that is inconsistent with the arm’s length standard on which those OECD guidelines are based. (Corporate Taxpayers Group)

The restricted transfer pricing rules are inconsistent with the general requirement to establish transfer prices based on arm’s length conditions. Section GC 13 requires the determination of an arm’s length amount of consideration for a transaction to produce a reliable measure of the amount that independent parties would have agreed upon under arm’s length conditions. The altered conditions on the related party loan brought about by sections GC 15 to GC 18 would not necessarily satisfy section GC 13(4) as being conditions that would be agreed between independent parties. The proposed legislation would also seem to require that any conditions imposed by sections GC 15 to GC 18 could in turn be replaced with other conditions that meet the arm’s length test in section GC 13. (EY)

The restricted transfer pricing rule had originally been included as part of the proposed changes to the thin capitalisation regime rather than the domestic transfer pricing regime. This was strongly opposed during the earlier submission phase and the drafting is not included as an addition to the general transfer pricing regime, which is the correct approach if specific legislative change is required in this area. However, given its origin, there has not been enough consideration to the application of the proposed general transfer pricing rules to be able to address the Government’s stated policy concerns and the likely impact of having the restricted transfer pricing rule as well – which is a unilateral and untested approach that other countries such as Australia are not adopting. (PwC)

Comment

During the earlier submission phase the equivalent of the restricted transfer pricing rule was proposed to be situated in the thin capitalisation rules as they both had the same purpose – to limit excessive interest deductions. However in determining the final version of the rules it was decided these interacted more closely with the general transfer pricing rules so should be situated accordingly. Officials note the submitter’s support for this decision.

The restricted transfer pricing rule has been developed in coordination with the transfer pricing changes that are also included in the BEPS Bill. Officials consider the restricted transfer pricing rule interacts consistently with the general transfer pricing rules. For example, if a New Zealand borrower structured to achieve a credit rating of BB but the restricted transfer pricing rule required this borrower to have a credit rating of BBB, the general transfer pricing rules would follow the same methodology and arrive at the same result as an otherwise identical borrower that had a credit rating of BBB without the application of the restricted transfer pricing rule.

Recommendation

That the submission be declined.

Issue: The drafting of the restricted transfer pricing rule is too complicated

Submission

(PwC)

There are a number of key aspects that should be simplified to reduce the scope for error in application, and to ensure the intended policy objectives are achieved. In particular:

when the rule applies (proposed sections GC 6(1C) and GC 15) particularly outside wholly-owned groups;
clarification of (a) the interaction of the concepts “acting together”, “acting in concert”, “control group” and “non-resident owning body” – each of these terms are complicated in their own right, let alone when they need to be interpreted together, and (b) how broad the terms are intended to be – for example clarity is needed as to whether securitisation vehicles, unit trusts and private equity funds are caught in the rules;
simplifying what the safe harbours/exceptions are and extending when they can be used;
determining the credit rating to be applied (including whether all members of a group must be credit rated); and
whether the term of a loan over 5 years needs to be adjusted for transfer pricing purposes.

Comment

Many of these themes were also raised by other submitters and have been addressed separately in this report. In response to the specific queries raised by the submitter:

Sections GC 6 and GC 15 are recommended to be partially redrafted, as supplied to submitters on 22 February 2018, which will remove and simplify the multiple application tests that are in the Bill as introduced. A copy of this proposed redrafting was provided to all submitters on the Bill with the opportunity for further submissions which have been incorporated into the final version recommended by officials.
The acting together, acting in concert, non-resident owning body terms have previously been used in either legislation or official’s documentation and the application to the interest limitation proposals is not intended to change this. Officials are recommending a number of changes to simplify this, particularly in relation to proposed sections GC 5 and GC 6 where the worldwide debt test for groups acting together has been reduced from 110% to 100%.
By removing the borrower’s credit rating proposed in section GC 16(7) of the Bill and returning these borrowers to the general transfer pricing rules this will simplify the use of the safe harbour.
It was never intended that a group be required to obtain a credit rating for all members of that group. Recommendations in this area will ensure this requirement only applies to a single member of that group.
Changes are recommended to the over 5 year term formula to ensure it operates correctly.

Recommendation

That the submission be noted.

Issue: Duplication of terms

Submission

(PwC)

A number of the requirements in section GC 15(3) are already covered by requiring that there be a “transfer pricing arrangement”, which in turn requires a “control group”. A transfer pricing arrangement may not include an indirect associated funding arrangement” so it is not clear how GC 15(3)(a)(ii) fits into the statutory scheme.

Comment

Officials agree that the drafting of section GC 15(3) as well as its interaction with GC 6(1C) and GC 15 created uncertainty through the use of multiple application tests with different terminology. Officials recommend a number of drafting changes to these sections to clarify their application as discussed elsewhere in this report.

Recommendation

That the submission be accepted.

Issue: Cross-border related party loans taxable within New Zealand

Submission

(Westpac)

Interest income on funding advanced to New Zealand subsidiaries via the New Zealand or foreign branches of an overseas entity is taxable in New Zealand either as part of attributed New Zealand branch profits or under the source rules in section YD 4(11)(b)).

It would be wrong to restrict the interest deductibility of cross-border loans where the corresponding income is also returned for New Zealand tax purposes given there is no loss to the New Zealand tax base. Under the proposed rules, a taxpayer would be required to rely on the Commissioner of Inland Revenue to exercise her discretion under section GC 11 to make a corresponding adjustment to the income recognised. A corresponding adjustment in these circumstances should be automatic and should not require discretionary approval, which would create uncertainty, delay and cost for the taxpayer.

Cross-border related party loans should be automatically exempted from the new rules where the corresponding loan interest income of the overseas loan holder is returned for New Zealand tax purposes.

Comment

The purpose of the restricted transfer pricing rules is to ensure an appropriate interest rate is charged on cross-border related-party loans. In a limited number of circumstances these interest deductions may be matched by income of the non-resident or branch. As noted by the submitter section GC 11 allows discretion for the corresponding income to receive a matching treatment.

Even where an adjustment to a cross-border deduction is matched by an adjustment to cross-border income, officials consider it is appropriate for this loan to be priced consistent with the restricted transfer pricing approach. For example ensuring this loan is priced appropriately supports the analysis as to whether an anti-avoidance rule may apply to the overall arrangement.

Where a cross-border related party loan is back-to-back or reflected in the features of a third party loan the restricted transfer pricing rules contain a number of provisions to ensure this pricing is reflected in the related-party transaction.

Officials consider it is appropriate for the restricted transfer pricing rules, including the various exemptions, to apply to cross-border related party loans where the corresponding interest income of the overseas loan holder is returned for New Zealand tax purposes.

Recommendation

That the submission be declined.

Issue: Minor drafting issues

#	Section	Submitter	Submission	Response
1.	FE 6(3)(e)(iiib)	Corporate Taxpayers Group	Clause 19(4) inserts proposed section FE 6(3)(e)(iiib). This should be FE 6(3)(e)(iv).	Section FE 6(3)(e)(iv) already exists. Proposed section FE 6(3)(e)(iiib) is not a replacement of this section and is intended to be inserted between existing sections FE 6(3)(e)(iii) and (iv) therefore the numbering in the Bill is correct.
2.	FZ 8(5)	Corporate Taxpayers Group	The transition period for the grandparenting of the worldwide group acting together test applies for a period beginning from the first balance date after 1 July 2018. Is this intended to be the end of the first full income year beginning after 1 July 2018? i.e. if a taxpayer had a September balance date, you would count from 30 September 2019 not 30 September 2018.	The amendments to the worldwide group test for acting together apply from the first income year after 1 July 2018 so the transition period should match this. For a taxpayer with a September balance date the transition period should start on 1 October 2018. Officials recommend drafting changes to achieve this.
3.	GC 6(1C)	PwC	Section GC 6(1C) is unnecessary and should be removed as it is poorly drafted.	Elsewhere in this report officials recommend the removal of proposed section GC 6(1C) as part of a package of amendments that clarifies the application of the restricted transfer pricing rules.
4.	GC 15(1)	PwC	Section GC 15(1) is unclear in its reference to providing funds to a “group of persons”.	While a group of persons can provide funds to a person it would not be normal for a person or group of persons to provide funds to a group of persons. In this instance each borrower would be treated as having a separate loan. Officials recommend this reference is removed.
5.	GC 16(1)(b)(i)	PwC	The cross reference in section GC 16(1)(b)(i) seems wrong.	Section GC 16(1)(b) applies only to co-ordinated groups and subsection (i) requires that section GC 16(1)(d) does not apply. As GC 16(1)(d) applies to borrowers not controlled by co-ordinated groups subsections (b) and (d) can never apply to the same borrower anyway. Therefore the cross-reference to paragraph (d) can be omitted.
6.	GC 16(1)(b)(iii) and (e)(iii)	PwC	It is unclear why sections GC 16(1)(b)(iii) and (e)(iii) references to “each lender”. If one lender does not meet criteria, is the borrower a high BEPS risk for all of its related-party debt? It should just be for that loan.	This wording was included to provide for a situation where a single loan is provided by more than one person and they are in different jurisdictions. This wording is not intended to extend the scope to other related-party loans which would continue be individually considered.
7.	GC 16(1)(b)(iii) and (e)(iii)	Corporate Taxpayers Group	The reference to the lender’s “ultimate parent” in section GC 16(1)(b)(iii) and (e)(iii) is not adequately defined. This is linking to FE 34 which applies only to registered banks.	Officials agree that the definition of “ultimate parent” in section YA 1, which references to FE 34, applies only to registered banks. However, subparagraph (2) of that section provides a definition that could be usefully repurposed more generally. This would be a company – That has an ownership interest in the lender of 50% or more; and In which no other company that has an ownership interest in the lender of 50% or more has an ownership interest. Officials recommend drafting changes to clarify this treatment.
8.	GC 16(1)(c)	Corporate Taxpayers Group	Should “controlled by a co-ordinated group” in section GC 16(1)(c) be a defined term?	This is already defined as a nonce term in section GC 16(1)(b).
9.	GC 16(1)(d)(ii)	EY, PwC	Section GC 16(1)(d)(ii) should refer to a borrower electing to use a credit rating given by GC 16(9) rather than GC 16(7)	Officials agree. However, this issue will be superseded by the recommended changes to the borrower’s credit rating recommended elsewhere in this report.
10.	GC 16(1)(d)(ii)	Corporate Taxpayers Group	The use of “made when or before” in section GC 16(1)(d)(ii) is unusual wording.	The intention of this provision is a borrower who is above the $10m de minimis can elect to use the group credit rating. This can be done in the return of income that includes the interest deduction or at an earlier date when they determine the interest rate on the loan. The current wording does not achieve this and officials recommend drafting changes to correct this.
11.	GC 16(1)(e)(ii)	PwC	It is not clear what is required by GC 16(1)(e)(ii). If a borrower has a 60% ratio but only 100% worldwide debt percentage, is the requirement met? We assume this is the intention.	It is only necessary for a borrower to meet either the 40% debt percentage or be within 110% of the worldwide group. In the submitters example as the borrower is less than 110% of their worldwide group they would still meet the requirement despite having a 60% debt percentage. Officials consider no amendment is necessary to achieve this.
12.	YA 1	PwC	The amended definition of related-party debt is unclear – the only definition of related-party debt is in RF 12H(1) so beginning words of paragraph (a) and para (b) should be removed.	Although RF 12(1) provides the definition of related-party debt this is restricted to not apply to banks by section RF 12(2). This restriction is necessary for the NRWT rules but not the restricted transfer pricing rules therefore the amended definition ensures that banks can have related-party debt for the purpose of the restricted transfer pricing rules. No further amendment is necessary.

Recommendation

That the officials’ recommendations, as shown above, be accepted.

APPLICATION TO THE SAME GROUPS AS THIN CAPITALISATION

Overview

Officials have become aware that as the Bill is drafted, the use of the general transfer pricing ownership threshold means this rule does not apply as widely as was intended. Officials recommend that in line with the stated policy intention, the ownership threshold is changed to align with that in the thin capitalisation rules, which also deal with the issue of interest deductibility.

In short, as the Bill is drafted, the restricted transfer pricing rule applies where a person or group holds 50% or more of the voting interests in a New Zealand company. Voting interests are the average percentage a person holds of four shareholder decision-making rights in a company. Officials recommend that ownership should be calculated using the same approach taken in determining whether the thin capitalisation rule applies. This means it will be calculated:

a) taking into account of non-voting shares and rights to receive distributions of income or capital, as well as voting interest;

b) in determining a person’s voting interests, it is the highest of the shareholder decision making rights that will be relevant, not the average.

This change will have no impact in the usual case where shareholders all hold ordinary voting shares.

From a policy perspective, the restricted transfer pricing rule is closely related to the thin capitalisation rule. Both rules deal with the very significant BEPS issue of interest deductions on cross border related party debt. In an EBITDA rule, as recommended by the OECD in BEPS Action 4, thin capitalisation and transfer pricing are effectively combined, and thus necessarily subject to the same ownership threshold. What has now been developed as the restricted transfer pricing rule was initially seen by officials as sitting appropriately in the thin capitalisation subpart of the Act.

However, it was decided the restricted transfer pricing rule should sit within the transfer pricing rules than the thin capitalisation rules. This was because:

the restricted transfer pricing rule restricts certain terms and features of related-party debt but still relies on the general transfer pricing rules to determine the final price of that debt;
embedding the restricted transfer pricing rule in the general transfer pricing rules allows certain related adjustments to be made, e.g. the adjustment for the payee in section GC 11.

An unintended consequence of locating the restricted transfer pricing rules outside the thin capitalisation rules is that the ownership test in the thin capitalisation rules does not automatically apply. As the restricted transfer pricing rule, as included in the BEPS Bill, does not include a provision to determine ownership in accordance with the thin capitalisation rules, it does not apply as widely as was intended.

Given the close policy connection between the thin capitalisation rules and the restricted transfer pricing rule, it is not appropriate for the two to have different ownership thresholds.

This information was publicly released and provided to submitters on this Bill on 15 February 2018 and proposed legislation was released on 22 February 2018. The Finance and Expenditure Committee called for further submissions on this issue and a summary of these and officials’ comments are shown below.

Issue: Officials’ proposal should not be accepted

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, Te Kakano Investment Limited Partnership)

The change to align the “related party” rules with the thin capitalisation rules is more than a drafting amendment and is fundamental to the rule. A change to the “related party” rules by way of an Officials’ submission is inappropriate and undermines the generic tax policy process. Such an approach assumes that Officials’ view of the policy intent is correct and increase the risk that the rule will be enacted based on ill-considered policy principles. (Chartered Accountants Australia and New Zealand)

The matter raised in the letter is not a drafting error. It is a significant policy change which should not proceed. The proposal contained in the letter received from officials is a significant change that would be introduced outside of the generic tax policy process. Further this has been introduced without any accompanying commentary or explanatory note. The normal accompanying commentary/ explanatory note is a vital resource for both the taxpayer and Inland Revenue that will be missing in relation to the restricted transfer pricing rule. (Corporate Taxpayers Group)

Any proposals along these lines could only be justified if the investments are clearly foreign controlled and interest rates are set on non-commercial terms. The measures being considered seem to go well beyond this and should be rejected. (Te Kakano Investment Limited Partnership)

Comment

The error is a minor change in the scope of the proposed rule. Officials acknowledge that this change may not be minor for an affected taxpayer as they will become subject to the rules when they would not be without this change. However, any such taxpayer, or their advisors, that has been following the policy development process would have been aware they were intended to be covered by the proposed rule as this was included in the Government discussion document as well as subsequent consultation documents and discussions with officials. This error does not change the operation of the rule for any other taxpayers.

While the commentary to the Bill and explanatory note will not refer to this error, taxpayers and Inland Revenue officials interpreting the application of this rule in the future will be able to find guidance in this officials’ report and other documentation including a Tax Information Bulletin item published shortly after the enactment of the Bill.

Recommendation

That the submission be declined.

Issue: Policy development

Submission

(Chartered Accountants Australia and New Zealand, KPMG)

In original consultation on the proposals, it was suggested that the new interest limitation rules form part of the wider thin capitalisation regime. We and others submitted that it would be more appropriate that the new rules formed part of the transfer pricing regime because the rules concerned the pricing of offshore related party debt and not the volume of debt. The Cabinet Paper issued in August 2017 stated that the Government had agreed to replace the interest rate cap proposal with a restricted transfer pricing rule. The Cabinet Paper noted submitters concerns that the proposed rule would be difficult to apply and could result in double taxation. It stated that the new rule would be more flexible. We believed that the change to a transfer pricing based rule was in response to submissions. The change to a transfer pricing based rule was confirmed when the Bill was introduced. (CA ANZ)

The officials’ letter of 15 February does refer to the February 2017 Discussion Document which did take a thin capitalisation based approach. However, that release was for consultation purposes. It was not supported by the August 2017 Cabinet Paper, which seemingly accepted the submissions made on the approach to be taken.

As we read the August 2017 Cabinet Paper, at best, it is silent on the exact related party test to be applied. However, in the context of a restricted transfer pricing rule, it is reasonable to assume that the transfer pricing “related party” test would apply in the absence of a clear statement to the contrary.

We are therefore not convinced by officials’ reasoning that there is a clear policy intent to the contrary and the Bill and Commentary positions are a drafting error. (KPMG)

Comment

Officials have been consistent in their intention that the restricted transfer pricing rule (and the interest rate cap as was earlier proposed) should apply to the same entities as thin capitalisation. For example on 8 September 2017 officials provided a note to all submitters on the March 2017 BEPS – Strengthening our interest limitation rules Government discussion document which set out the current proposals. This document included at paragraph 34 that:

“We propose that a financial arrangement will be subject to the restricted transfer pricing rule if:

the borrower is an entity that is subject to the thin capitalisation rules (the inbound rules, the outbound rules, and the banking rules); and ….”

Both before and after the release of this note officials held meetings with interested parties to develop the restricted transfer pricing rule. As part of these discussions officials discussed their continued intention to apply the restricted transfer pricing rule to entities that are subject to the thin capitalisation rules.

Recommendation

That the submission be declined.

Issue: Effect on iwi

Submission

(Te Kakano Investment Limited Partnership, CNI Iwi Holdings, Teresa Farac)

These changes need considered and detailed analysis and full consultation with Iwi that has not occurred.

To develop the potential of our assets, Iwi often find it productive to enter into joint venture arrangements sometimes with foreign investor participation. To protect our kaitiatanga these joint ventures often involve varying voting rights and rights to income. In this way Iwi can ensure sustainable development and fulfil our obligations as stewards of Iwi assets.

The interest limitation proposal is likely to intrude on these endeavours by imposing an artificial interest limitation on the funding provided by joint venture partners. Our investments have varying degrees of debt and equity balancing the interests of the different parties. All prices and returns (including interest rates) are set at market prices so that all parties are fairly remunerated. Foreign investors do not have control of these investments.

Iwi often want debt funding because this best protects our stewardship of Iwi resources. If enacted an interest rate cap will undermine existing investment partnerships and hinder the development of new partnerships.

The proposed amended policy by officials seems especially directed at varying voting and income rights. The officials’ letter of 14 February notes that in the usual case shareholders do not have varying rights. That may be true but is false with respect to Iwi ventures. The officials’ proposal, to us, appears to be a direct attack on the commercial interests of Iwi. (Te Kakano)

As Maori entities, it is important that we maintain our right of Rangatiratanga over our resources going forward. It is also important that we protect our rights on whom we partner with along with the structure or financing that is used in creating these partnerships going forward. We see that this rule may hinder potential debt structured entities from claiming all of the interest costs that is charged on loans with foreign partners. This will likely decrease some of the value that these partnerships with foreign partners will have if this limitation is adopted into the BEPS bill. (CNI Iwi Holdings)

The proposal will penalise NZ tax preferred investors such as Iwi and charities in non-foreign controlled companies by imposing artificial credit ratings, credit support and debt terms to otherwise commercially negotiated loan terms.

Comment

The proposal will have no direct impact on the tax implications for Iwi and other New Zealand investors as the restricted transfer pricing rules only apply to interest paid to non-residents. However, there are two indirect impacts that may arise. First, foreign investors may limit their investment or reduce the price they are prepared to pay for New Zealand assets due to a lower after tax return. But this is true of the impact of all the BEPS measures that address strategies to exploit weaknesses in those laws applying to in-bound investment. Second, New Zealand residents, including Iwi, may be shareholders in a company which has non-resident shareholders and which is subject to interest disallowance under the restricted transfer pricing rule. This outcome could only be avoided if the restricted transfer pricing rule applied only to 100% foreign owned companies. Such a limitation would allow the rule to be circumvented by including a small element of New Zealand ownership. This would significantly reduce its effectiveness. Furthermore, this kind of effect can already arise for New Zealand investors in a company subject to the thin capitalisation regime. As already referred to in this report, the restricted transfer pricing rule is designed to buttress the thin capitalisation regime. Furthermore, if the New Zealand borrower does not adopt a tax-aggressive capital structure, there will be no interest disallowance.

Recommendation

That the submission be declined.

Issue: More than 100% combined interests

Submission

(Teresa Farac)

Taking the highest percentage is artificial as it has the capacity to attribute more than 100% shareholding interests to a company and can result in there being more than one parent (which renders the credit rating provisions unworkable).

Comment

The restricted transfer pricing rules do not require calculating the aggregate interests of all shareholders therefore the situation where total interests total to more than 100% does not cause any concerns. There is a possibility for more than one shareholder to have interests of 50% or higher but the only consequence of this is the restricted transfer pricing rules could apply to loans from both shareholders if the other requirements (such as being non-resident) were met. Officials are not proposing the highest percentage interest approach be extended beyond the restricted transfer pricing rules so general concerns, such as a company being required to be a member of two separate groups, do not arise.

Recommendation

That the submission be declined.

Issue: Lack of control over New Zealand entity

Submission

(Corporate Taxpayers Group, KPMG, Teresa Farac)

The scope of the rule should be limited to those with the ability to control a New Zealand entity, such that it is in a position to artificially inflate the price of debt or “shift” profits. (Corporate Taxpayers Group)

The further from the transfer pricing “related party” test the boundary is drawn, the more likely it is that the restricted transfer pricing rule will inappropriately apply. The transfer pricing rules apply because of an assumed ability to influence pricing to achieve a “non arms-length” result. The extended (i.e. thin capitalisation-based) “related party” test would also apply in circumstances where there is no or limited ability to influence to achieve a non arms-length result. (KPMG)

Parent status will be attributed to a foreign company that from a commercial, accounting and economic perspective does not control the NZ entity. The foreign company will be deemed to have provided credit support and mirror credit rating, which is factually incorrect. The proposal is inconsistent with the intent of the debt pricing rules in having the capacity to apply to transactions without the requisite control and ability to use artificial or commercially irrational debt pricing. It is unnecessary given the breadth of the current provisions in the Bill. (Teresa Farac)

Comment

Officials agree that the intent of the restricted transfer pricing rules is to apply them to related party transactions where the lender has the requisite control and ability to use artificial or commercially irrational debt pricing.

Without this proposal certain taxpayers will be able to structure around the restricted transfer pricing rules by having varying shareholder rights so that their average rights are less than 50% while still having control over debt pricing.

Recommendation

That the submission be declined.

Issue: Consistency with CFC and thin capitalisation rules

Submission

(Teresa Farac)

The use of the highest shareholding interest test is based on an incorrect interpretation by Inland Revenue of the international tax rules. The CFC rules recognise that it is not sensible to attribute >100% shareholding interests to a company and they reduce shareholder interest on a pro rata basis for the purpose of the taxing provision which attributes income of the CFC to the NZ taxpayer. In particular, section EX 12 of the Income Tax Act 2007 applies where the total interests exceed 100% interest and operate to reduce a person’s income interest by the formula: income interest before reduction x 100 ÷ total income interests before reduction.

The thin capitalisation rules use the highest percentage interest which a shareholder has in the 4 categories listed in section EX 5(1) which are the percentage of shares held, shareholder decision-making rights, rights to distribution of net income and distribution of net assets. Section FE 39 only refers to “…categories listed in EX 5(1)…” and does not expressly refer to EX 5(4) (which takes the highest shareholder decision-making rights).

Comment

Section EX 12 is necessary for the purpose of the CFC rules as it would be inappropriate to attribute more than 100% of the income as this could result in double taxation. There is no need for an equivalent provision in the restricted transfer pricing rule as no double taxation can arise from potentially having more than one investor with 50% control. As noted in the item above, in this situation the only consequence is two or more investors, subject to meeting or failing other criteria, may be prohibited from including terms and features that would not be present in third party debt from affecting the pricing of related party debt.

Section FE 39 refers to EX 5(1) as a list of direct control interests existing prior to the introduction of the thin capitalisation rules. Although there is no direct reference from subpart FE to EX 5(4) this is not necessary as section FE 39 already takes the highest percentage of shares or rights held by the investor.

Recommendation

That the submission be declined.

Issue: Non-voting preference shares

Submission

(Teresa Farac, OliverShaw)

The proposal will penalise companies with material non-voting preference shareholders that have borrowed from otherwise third parties on commercial terms (noting that in distress situations it is not uncommon for lenders to hold preference shares). The rules would deem parental support when factually none exists and disregard features that are commercial and entered into by third parties operating at arm’s length. (Teresa Farac)

Where a lender has minority voting interests but high income interests (because, for example, it has provided non-voting preference share funding to assist a New Zealand entity in financial difficulty) it will be deemed to control the New Zealand borrowing entity whose interest rate should be capped at the lender’s borrowing rate. (OliverShaw)

Comment

In general preference shares should not be excluded from the restricted transfer pricing tests, particularly when they are held by shareholders who also hold ordinary shares, as they can be used as part of wider investment decisions by owners of the business.

Officials accept that in a case of financial difficulty a lender may convert debt to preference shares and that implicit parental support is unlikely to be provided in such a situation as the lender is more interested in getting some or all of their investment returned rather than operating the business. Officials recommend preference shares are removed from the control tests in the restricted transfer pricing rule provided that:

the preference shares were issued in satisfaction of or as a replacement for outstanding debt;
that outstanding debt was provided by the lender in the ordinary course of their business of providing funds to unrelated parties; and
the debt was originally issued by the same entity that now holds the preference shares.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Appointment of directors

Submission

(OliverShaw)

A shareholder with majority rights to elect just one of say 12 directors with otherwise only a minority shareholding interest will be deemed to control the borrower and the borrower’s interest rate should be capped at the lender’s borrowing rate. For example, it is not uncommon for a cornerstone foreign investor to provide debt funding. If the cornerstone shareholder owns say 20% of the company but has the majority rights to appoint a director, the tax deductible interest will not be the market or arm’s length interest rate, rather it will only be deductible based on the rate determined based on the credit rating of that foreign shareholder.

Comment

The shareholder decision-making right the submitter refers to is the right to participate in any decision-making concerning the appointment of a director of the company. A cornerstone shareholder with the right to appoint a single director out of 12 would have a 100% right to appoint one director but a 0% right to appoint the other 11. This would only provide that shareholder with an 8.3% right to appoint the directors of the company.

In addition to the definition in section YA 1 the term “shareholder decision-making right” appears in 8 other existing sections. To apply the interpretation suggested by the submitter, would result in each of these sections not operating correctly.

The example provided by the submitter shows that a shareholder who owns 20% of the company, even if all other rights were held equal to their ownership would only hold a 17.1% average interest because their rights to appoint a director were less than proportional to their ownership. In contrast, the approach proposed by officials would result in the shareholder having a 20% interest for the purpose of the restricted transfer pricing rules, as they already do for the thin capitalisation rules.

Recommendation

That the submission be declined.

Issue: Scope of draft legislation

Submission

(EY)

The draft legislation provided to submitters on 22 February 2018 is unnecessarily complex. A number of aspects previously included in clause 37 appear to have been shifted into clause 35, with various wording and definitional changes made to both clauses. This complexity, along with the short period allowed for submissions on the proposed changes, makes it hard to determine whether the draft legislation goes further than indicated in the Note. We would have strong concerns around any extension to the scope of the restricted transfer pricing rule or to the transfer pricing rules in general beyond that indicated in the Note.

Comment

Officials provided all submitters on the BEPS bill with proposed draft legislation to rectify the error discussed above. In preparing this draft legislation officials, and other submitters as noted elsewhere in this report, recognised the complexity in proposed sections GC 6 and GC 15 of the Bill. As well as addressing the identified error this draft legislation reduced this complexity by removing a number of the entrance tests to the restricted transfer pricing rule. These changes were not intended to change the scope, rather they intended to clarify the rule’s application. These changes were consistent with the changes that would normally be recommended by officials as part of the revision tracked version of the bill provided to the Finance and Expenditure Committee.

Recommendation

That the submission be declined.

Issue: Application of thin capitalisation threshold to restricted transfer pricing

Submission

(EY)

Provided the recasting of the ownership interest test for the purposes of the restricted transfer pricing rule is not seen as further increasing the scope of the restricted transfer pricing rule or the transfer pricing concept across the board (which we would strongly oppose), we consider the proposed change will have limited application. That is, it will only apply to relatively rare situation where the various rights differ. As stated at [3] of the Note, “this change will have no impact in the usual case where shareholders do not have varying rights”.

In our view, the proposed extension to the restricted transfer pricing rule is unnecessary and is not justified given that it will cover, at most, a handful of rare situations which would not in themselves appear to carry a high risk of base erosion and profit shifting. We find it hard to envisage taxpayers restructuring their affairs in such a way as to change ownership and dilute control simply to escape the restricted transfer pricing rule.

Comment

Officials agree that this change will not affect many taxpayers. However, officials believe that it is appropriate for the restricted transfer pricing rule to be able to apply to loans between a New Zealand borrower and any non-resident lender where that lender is part of the group whose control of the borrower has triggered application of the thin capitalisation rules to the borrower’s interest deductions.

Recommendation

That the submission be declined.

Issue: Public perception of integrity of Parliamentary process

Submission

(EY, KPMG)

Submitters should be able to rely on the Bill as introduced as representing the Government’s policy intent in all but the most extreme situations.

The Committee itself should be able to scrutinise the Bill and the recommended changes safe in the knowledge that officials will not seek to introduce alternative measures on which the Committee is unable to deliberate. Supporting the change recommended by officials risks weakening public perception of the integrity of the Committee’s processes.

The fact that draft legislation was not released at the same time as the Note which called for submissions on the proposed change is also likely to have a negative impact on public perception, especially as the draft legislation appears to go further than originally indicated in the Note. (EY)

The officials’ letter of 15 February description of the proposed extension – that the restricted transfer pricing rule should apply if the highest, rather than the average of, the four types of ownership interest, is 50% or greater – is inconsistent with the proposed legislative re-draft. The new wording imports the full thin capitalisation “related party test”. This includes counting rights arising from options and similar pursuant to current section EX 6. This goes beyond the officials’ letter description of the policy intent. It gives effect to the September 2017 note and not to the second paragraph of the officials’ letter. This further reinforces our submissions. The policy intent either remains unclear, is shifting, or the drafting does not achieve the intent. (KPMG)

Comment

While officials and the Government make every effort to ensure a Bill as introduced represents the final position it is common for officials to make a submission on items that are not correctly reflected in a Bill but have not been reflected in external submissions. These items are usually identified as such and considered by the Committee, and by the independent advisor to the Finance and Expenditure Committee, without calling for further public submissions.

Officials recognised that in the case of this error, notwithstanding that it affects a very small group of taxpayers, there would be public interest in this correction. While there was less time for submitters to consider this issue, officials note that this is a discrete issue that would be expected to be less resource intensive to review than the Bill as a whole.

Officials released the note in advance of the draft legislation as they sought to provide all submitters with the maximum time to consider this issue while the necessary draft legislation was still being prepared. The draft legislation, as could be expected, provides greater detail than the note on the operation of the proposed correction; however both are consistent in their intent that the restricted transfer pricing rule should apply to the same taxpayers that the thin capitalisation rules apply to and as the submitter notes this is consistent with the September 2017 note.

Recommendation

That the submission be declined.

ADJUSTED CREDIT RATINGS

Overview of three policy recommendations

In response to concerns by submitters, officials are recommending three particular changes to the credit rating adjustments in proposed section GC 16. These changes will reduce compliance costs and go some way to addressing submitters’ concerns regarding double taxation but do not fundamentally alter the operation of the credit rating adjustment portion of the restricted transfer pricing rule. In order to provide for a better understanding of the intended policy they are summarised in the following paragraphs.

These three changes are:

Removing the income-interest ratio high BEPS risk test;
Allowing credit ratings to be implied from significant third party debt; and
Allowing the borrower’s credit rating to be within two notches of the group credit rating provided the borrower has a credit rating of BBB- or above.

Income-interest test

The Bill as introduced had three high BEPS risks tests where the borrower must apply a restricted credit rating or group credit rating if any of these three tests are not passed.

Two of these tests were aimed at similar features being the amount of debt the business was sustaining. The first test in section GC 16(1)(b)(ii) and (e)(ii) calculates the borrower’s debt a proportion of net assets. The second test in section GC 16(1)(b)(iv) and (e)(iv) calculates the borrower interest expenditure as a proportion of their earnings before interest, tax, depreciation and amortisation (EBITDA).

When considering all businesses, debt will be highly correlated with interest. Net assets will be highly correlated with EBITDA. While this relationship will not always hold, for example where a business is making losses or is paying a very low rate of interest, the majority of businesses that fail one test would also fail the other test. Where a borrower passes the debt/net assets test and fails the EBITDA test this may be because of reasons that do not necessarily indicate they are a high BEPS risk.

The debt as a percentage of assets test will show that borrowers who are sustaining a high level of debt (either as a proportion of assets or compared to their worldwide group) are a high BEPS risk. A borrower that has a lower percentage of debt would only fail the income-interest test if they had a smaller amount of higher priced debt. The standard transfer pricing rules, including the amendments in this Bill, would continue to apply.

Accordingly, officials recommend that the income-interest test be removed and whether a borrower is a high BEPS risk is determined on the remaining two tests. This will reduce compliance costs and remove a number of borrowers from the high BEPS risk credit rating who were not the intended target of the proposals.

Third party credit ratings

The part of the restricted transfer pricing rule relating to disregarded features allows a borrower to include a feature in pricing if it is a feature in significant third party debt. That an unrelated borrower would provide debt with this feature provides objective support that the feature is included for commercial reasons and the requirement that the debt be significant prevents a borrower obtaining a small amount of expensive uncommercial debt to justify high priced related party debt.

The Bill as introduced does not include an equivalent feature for credit ratings. A borrower who is not a high BEPS risk would be able to continue to use third party debt to imply a credit rating for related party debt under a standard transfer pricing approach but this would not be available to a borrower that was a high BEPS risk.

Officials recommend the third party test is extended to the credit rating of high BEPS risk borrowers. This is for the same reason as the disregarded features third party test, that if an unrelated lender is willing to lend at a particular implied credit rate this is objective evidence of the credit rating of the New Zealand borrower.

In order to apply this rule a number of requirements must be met:

The related party debt must be no more than four times the amount of the relevant third party debt – this is the same requirement in the disregarded features third party test and prevents a borrower having a very high level of related party debt to appear riskier then borrowing a small amount of third party debt to justify the higher interest rate.
The third party debt must be unsubordinated. As with the disregarded test, subordinating debt will increase its interest rate as there is a reduced chance of the lender getting their money back if the borrower gets into financial difficulty.
The third party debt must be unsecured. Unsecured debt better represents the risk of the borrower whereas secured debt represents the credit risk attached to the secured asset and can also be influenced by the size of the borrowing compared to the value of the secured asset.

Two notches below group credit rating

The group credit rating, for borrowers that are a high BEPS risk and have an identifiable parent, limits the borrower’s credit rating to one notch below their group’s credit rating[4]. Officials now recommend that this spread be increased to two notches provided the New Zealand borrower’s credit rating is BBB- or higher. This is aimed at addressing submitters’ concerns about double taxation. This change means a borrower using the group credit rating will be less likely to suffer double taxation and reduce the need for reliance on the Mutual Agreement Procedure in Double Tax Agreements.

BBB- is the lowest grade investment credit rating. Below this rating the difference in interest rates from a single notch movement becomes much larger so officials consider the original proposal for a single notch spread should be retained.

The following table sets out the lowest available credit rating for a high BEPS risk borrower with an identifiable parent for certain credit ratings:

Group credit rating	Borrower’s credit rating	Maximum spread
A	BBB+	2 notches
A-	BBB	2 notches
BBB+	BBB-	2 notches
BBB	BBB-	1 notch
BBB-	BB+	1 notch
BB+	BB	1 notch

These changes are incorporated into the following flowchart which provides a high level picture of whether a credit rating adjustment would be required under the proposed rules.

Restricted Transfer Pricing: Process for determining NZ borrower's credit rating

Source: SVG version

Issue: Linkage with parent’s credit rating

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, KPMG, PwC)

Submitters do not support a restricted transfer pricing rule deeming a subsidiary to have a credit rating one notch below its parent. They consider this approach:

Is unique, and an inappropriate departure from the internationally accepted norm of an arm’s length transfer pricing principle;
Arbitrarily and inappropriately disregards the facts and circumstances of the New Zealand taxpayer, and will yield unfair and inconsistent results;
Will almost certainly give rise to double taxation in circumstances where the credit ratings are more than one notch different;
Is inconsistent with New Zealand’s Double Tax Agreements;
Will create significant complexity and increase compliance and administration costs; and
Could limit flexibility in raising capital.

The proposals do not adequately take into account significant differences among businesses in terms of their scale, their credit ratings, the industries and countries in which they operate, different levels of external leverage, where the entity is at in terms of its business life cycle, its risk weighting and different policy considerations that countries may have in determining how related party debt should be priced. All of those factors are reflected in the business’ operating results and it seems illogical to totally disregard them when pricing the cost of debt.

Comment

Applying a credit rating is already standard practice for transfer pricing. Applying the restricted transfer pricing limits on a high BEPS risk borrower is a straight-forward, simple and non-manipulable way of pricing related-party debt. The rules have been designed to limit compliance costs and have minimal application for borrowers that have similar debt structures for the New Zealand borrower and their worldwide group. Officials expect many of the businesses that would currently be subject to the restricted transfer pricing rules will adopt more orthodox structures so they pay appropriate amounts of interest and are not at risk of double taxation.

However, as noted above, officials recommend the maximum allowable gap between the group credit rating and the borrower’s credit rating be increased from one notch to two notches. This will alleviate some submitters’ concerns regarding double taxation.

Recommendation

That the submission be declined and officials’ recommendation to extend the group credit rating to two notches for borrowers at BBB- or above be noted.

Issue: Flat 40% gearing ratio limit

Submission

(Chapman Tripp)

We are uncertain whether a flat 40% gearing percentage is an appropriate measure for determining whether a taxpayer is a high BEPS risk across all industries. Consideration should be given to whether a fixed percentage is appropriate.

Comment

While a 40% limit will be easier for businesses in some industries to meet than others officials consider it is an appropriate indication of where a borrower represents a higher risk of BEPS. For borrowers that have higher than 40% debt percentages they will still be able to rely on a having a debt percentage within 110% of the worldwide group. Officials note that even where a borrower has a debt percentage putting them at high BEPS risk so the restricted transfer pricing rules apply; provided debt levels are appropriate so that the credit rating of the New Zealand group is within one notch (or two notches as recommended elsewhere in this report) of the parent (and there are no disregarded features), deductions for interest will not be adjusted under the restricted transfer pricing rule.

Recommendation

That the submission be declined.

Issue: 40% gearing limit is too low

Submission

(Chartered Accountants Australia and New Zealand, PwC)

The thin capitalisation threshold at which a New Zealand borrower is considered to have an excessive level of debt (taking into account all three factors) is too low. As we understand it, the 40% threshold is based on a median thin capitalisation ratio for New Zealand borrowers that were significant foreign owned enterprises. On this basis, the rule is designed to ensure that half of significant foreign enterprises are within its scope.

This appears arbitrary, and given the anti-avoidance nature of the rule the threshold should reflect the 60% thin capitalisation debt threshold rather than seeking to further constrain. (Chartered Accountants Australia and New Zealand)

The 40% leverage ratio is significantly lower than the risk factor ratio of 60% referenced in the ATO’s guidance. It is also significantly lower than that allowable under our thin capitalisation regime, which is surprising given the Government has recently stated that it is satisfied with the current thin capitalisation ratio. While the proposed legislation does allow taxpayers the alternative of referencing the 110% worldwide leverage ratio, this is too difficult for taxpayers to calculate (as is the case under existing thin capitalisation rules where this ratio is rarely applied in practice).

Comment

Officials consider the high BEPS risk debt percentage and the thin capitalisation debt percentage should not to aligned. The high BEPS risk test is only an initial indication that the borrower may be structuring to achieve higher than commercial debt levels but does not necessarily result in any adjustment to interest deductions. Whereas, the thin capitalisation limit sits at a higher level as when debt is above this percentage interest starts being not fully deductible. As the consequences of breaching the thin capitalisation limit are more severe than breaching the high BEPS risk test it is appropriate that they are set at different levels.

The 40% threshold was set on the basis that the median debt percentage for significant foreign owned enterprises is between 30-40%. Data for the 2016 income year showed that 89 of 320 New Zealand groups (approximately 28%) had a debt percentage higher than 40%. The Bill also tightens the debt percentage calculation, so this percentage can be expected to increase (assuming no behavioural change). A reasonable estimate of the percentage of groups over the 40% threshold once this change is made might be a 10% increase, to 38%.

However, this is almost certainly an over-estimate of the groups who will need to consider whether the RTP rules may apply, because it does not take into account the fact that some New Zealand groups will:

be able to establish that they are within 110% of their worldwide debt percentage
have significant third party debt which they can use to determine their credit rating
reduce their New Zealand debt percentage below 40%

Officials also considered the relative debt levels of NZX listed companies. While New Zealand is a relatively small market and thus suffers from the problem of having relatively fewer independent companies compared to other developed economies it is still possible to compare financial data of NZX listed companies to significant foreign owned enterprises. On the considered metrics of Debt/Equity, Debt/Assets and Net Finance Costs/EBITDA all three showed foreign-owned groups are biased toward holding more debt and paying more interest than NZX listed companies.

Recommendation

That the submission be declined.

Issue: Location of 40% gearing ratio

Submission

(Chartered Accountants Australia and New Zealand)

It would be preferable if the gearing rule were moved to the debt pricing section of the analysis, rather than the credit rating section so that taxpayers are simply required to price the debt assuming an arm’s length amount of debt is in place. This would allow them to use their own debt credit rating adjusted for implicit parent support rather than being forced to the highest group member less one notch.

Comment

Applying a 40% maximum gearing ratio to a high BEPS risk borrower does not remove the potential for disagreement over the level of implicit parental support. For the majority of borrowers who do have an identifiable parent, linking the borrower’s credit rating to their worldwide group’s better reflects the implicit support available. While a maximum 40% gearing approach is proposed for borrowers who do not have an identifiable parent this was designed because there is no foreign parent to calculate the appropriate rate off. Officials consider basing the credit rating off the foreign group’s rating more accurately reflects an appropriate credit rating when such an approach is achievable.

Recommendation

That the submission be declined.

Issue: Income-interest test

Submission

(Corporate Taxpayers Group)

The requirement to satisfy a 3.3 EBITDA / interest ratio will adversely impact a number of businesses operating in a completely commercial manner. Businesses with trading losses are likely to automatically fail the test. Start-up businesses are likely to fail the test (for example a petroleum miner undertaking exploration activities may have a long lead time between incurring expenses exploring for petroleum and finding fuel and generating income). Likewise, taxpayers who have a year in which there are one-off major transactions which result in a lower trading profit will be impacted (for example, an acquisition or merger may result in higher than normal expenses).

Comment

Officials accept that there are problems with an EBITDA test such as the inability to accurately forecast earnings in advance and the variability of such a measure. The bill as introduced attempts to mitigate these concerns by allowing an average EBITDA over up to three years. Upon further consideration officials consider the EBITDA test can be removed with its effect being covered by the 40% debt percentage test which has a broadly similar outcome of identifying taxpayers with higher levels of borrowing.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Annual income-interest test

Submission

(Chapman Tripp)

It is undesirable that a taxpayer be treated as a high BEPS risk simply because their income is less than 3.3 times their interest expense in any particular year – this test could cause taxpayers to drop in and out of the rules, greatly increasing compliance costs. We submit that this test should be dropped, or measured on a rolling three year average.

Comment

The proposals in the bill already applied an average of up to three years and only at the time each loan was entered into or repriced. However, as discussed in the item above officials recommend the EBITDA test is removed.

Recommendation

That the submission be accepted.

Issue: Income-interest ratio in a first year of operation

Submission

(PwC)

A company within its first year of operation will not be able to calculate an income-interest ratio as it will have less than 4 quarters/12 months of data. It should be clarified that the calculation can be done in a start-up year.

Comment

The bill proposal did not address this issue, however as officials now recommend the EBITDA test is removed this is no longer a concern.

Recommendation

That the submission be noted.

Issue: Income-interest ratio of 3.3 is too high

Submission

(PwC)

The income-interest ratio of 3.3 is too high. We appreciate that this is consistent with OECD BEPS interest limitations recommendations (which have not been adopted in our draft thin capitalisation legislation), but note that the ATO’s guidance has recently relaxed its risk rating assessment in relation to this ratio (presumably in recognition of the requirement being too stringent).

Comment

As noted by the submitter this ratio is consistent with the 30% OECD recommendation. The OECD also suggested alternative ratios between 10% and 30% which would have made the equivalent income-interest ratio higher than 3.3. However as officials now recommend the income-interest test is removed this is no longer a concern.

Recommendation

That the submission be noted.

Issue: Residence test – having the usual tax status of a company

Submission

(Corporate Taxpayer’s Group)

The reference to “or would be for a company having the usual tax status of a company” in section GC 16(1)(b)(iii) and (e)(iii) is not clear enough from the legislation what this is intended to cover.

Comment

The purpose of this wording is to cover situations where the lender is not subject to a tax rate of 15% or more because of a policy decision in the lender’s jurisdiction other than the general company tax rate being below 15%. For example the lender may be a sovereign wealth fund that is tax exempt even though a company from the same jurisdiction would be subject to a higher rate than 15%. Entities that are not subject to tax because of a policy decision are not intended to be covered by the high BEPS risk tests.

Recommendation

That the submission be noted.

Issue: Two out of three high BEPS risk tests

Submission

(Corporate Taxpayers Group)

It should not be necessary to meet all three high BEPS risk tests in order to be a low BEPS risk. Two of the three tests should be sufficient.

Comment

As discussed above officials recommend that the EBITDA test be removed. Of the two remaining tests, officials consider if either of these is not met it is appropriate for the borrower to be considered to have a high BEPS risk.

Recommendation

That the submission be declined.

Issue: Credit rating adjustment when third party loan with similar features exists

Submission

(Chapman Tripp, EY, KPMG, PwC)

The credit rating adjustment is unnecessary where a taxpayer has an existing loan with a third party with broadly similar features to the taxpayer’s related party debt. In this situation, a third party lender has already priced the implicit support that a New Zealand subsidiary will receive from its offshore parent. There is no need to adjust the credit rating of the subsidiary to account for implicit support.

This approach is also consistent with both achieving the Government’s objectives (as the best evidence of a borrower’s cost of funds is actual third party financing) and long-established international transfer pricing practice. This principle is fundamental to the approach taken by the ATO in addressing the same policy concerns the Government is seeking to address.

Comment

Officials agree that it would be appropriate for a borrower, including one that is a high BEPS risk or is an insuring or lending person, to apply a credit rating on related party debt that is equal to the credit rating implied from third party debt. To prevent abuse of this position officials recommend a third party credit rating should be subject to the following restrictions:

The credit rating should be equal to the highest implied credit rating on outstanding long term senior unsecured debt that is not related party debt.
The amount of related party debt cannot be more than four times the principal of the third party debt with that credit rating – this restriction is consistent with the third party debt exception applying for disregarded features.

Officials have considered whether it would be appropriate for a credit rating to be implied from cross-border related party lending that is back-to-back with a third party loan. This would require further consideration and consultation to ensure it was appropriately targeted at genuine back-to-back arrangements. This is not possible as part of the current select committee process so officials do not recommend the introduction of such a provision as part of this Bill.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Highest group credit rating

Submission

(Corporate Taxpayers Group, EY, KPMG, Powerco, PwC)

The legislation and commentary are inconsistent in describing what credit rating should be used with a taxpayer who is required to use the “group credit rating”. The commentary (and previous Government announcements) states that the group rating is “the higher of the parent’s credit rating minus one notch or the borrower’s own rating”, however, proposed section GC 16(9) states the credit rating is the higher of the borrower’s credit rating and the highest credit rating of a member of the borrower’s worldwide group. As drafted, the legislation is impractical and will not be able to be complied with. Worldwide groups can include hundreds of entities in different countries and they will not all have credit ratings (as they are not required or necessary). Section GC 16(9) should be amended to reflect the position in the commentary. (Corporate Taxpayers Group)

Using a rating that is the highest rating for any member of a borrower’s worldwide group could result in an interest rate below that which could realistically be obtained by the group from external borrowing to fund New Zealand operations. The interest limitation rule should reflect the most common scenario under which a parent (or special purpose finance entity) will issue group debt instruments. Further the pricing of related party debt to a New Zealand entity should reflect the group credit rating (less one notch) for its particular business operations and not some higher rating applying to a completely different type of business activity in the group. (EY)

Where borrowers are required, or choose, to apply the worldwide group approach for pricing their cross-border related party loans, they should only have to have regard to the credit rating of the ultimate parent of the worldwide group. (KPMG)

Comment

The group credit rating is intended to be the rating for long term senior unsecured debt of the highest rated member of the group. In most cases this highest rated member would be the ultimate parent company of the group, but this will not always be the case. For example, the ultimate parent company may be a holding company that holds shares in a higher-rated operating company.

It was not intended that a group would have to consider the credit rating of every company within the worldwide group. It is not obvious how to create a simple rule that identifies the relevant foreign operating parent in all situations and it is not appropriate to rely on the ultimate parent as this, as noted above, could be a lower rated holding company. Instead officials recommend this rule be amended so that the group credit rating is based on the member of the worldwide group that has the highest level of unsecured third party debt. In most instances officials expect this entity would be either the main operating entity or a treasury function for the group.

Recommendation

That the submission be declined and officials’ alternate suggestion to apply the credit rating of the member with the highest unsecured third party lending be accepted.

Issue: Access to credit ratings

Submission

(Corporate Taxpayers Group, EY, KPMG, PwC, Westpac)

“Credit rating” is itself not clearly defined in the Bill but appears to refer to a formal credit rating provided by a credit rating agency authorised by the Reserve Bank of New Zealand. This approach is problematic, as it is predicated on the borrower (and every member of its worldwide group) maintaining a credit rating provided by one of the authorised rating agencies.

Where a formal credit rating is not maintained, the transfer pricing rules provide an avenue through which a rating can be deduced, through the preparation of a credit scoring analysis typically following industry standards and commentary provided by the authorised rating agencies. The resulting indicative credit score provides a reasonably reliable approximation of the borrower’s credit rating for the purposes of applying the transfer pricing principles.

The term “credit rating” must be defined in the legislation so as to provide taxpayers with the ability to determine their indicative credit score, and should not enforce unnecessary compliance costs by requiring taxpayers to maintain a formal credit rating. This must also apply to situations under which the proposed rules require the use of a rating of any offshore related party. (Corporate Taxpayers Group)

Sections GC 16(7) and (9) should be clarified to state that the rating used should be the most recent actual rating at the most recent calculation date, or a rating that a rating agency would have given at that date where no actual rating exists. (EY)

Where no published long-term senior unsecured credit rating exists for the worldwide group (or a group member) preparation of a traditional transfer pricing interest rate benchmarking analysis for the consolidated worldwide group (using credit rating models published by organisations like Moody’s or S&P, or similar) should be acceptable for determining the worldwide group’s credit rating.

Credit rating should be a defined term, and should permit the taxpayer to use a credit rating from any major credit rating agency (for instance Moody’s Standard & Poor’s, and Fitch) for the purposes of GC 16 and GC 17. (Westpac)

Comment

Officials agree that taxpayers covered by the restricted transfer pricing rules should not have to obtain a formal credit rating from a credit rating agency if they do not already have one. Where a credit rating for an entity is not available it is common transfer pricing practice to imply one from external borrowing. Where there is no external borrowing transfer pricing will following additional steps to estimate an appropriate credit rating.

While the restricted transfer pricing rule will place some limits on an appropriate credit rating for a New Zealand borrower it is not intended to change the methodology used to calculate a credit rating where one is not available. Officials recommend drafting changes to reflect implied credit ratings and other transfer pricing methodology for estimating a credit rating when an entity does not have a formal credit rating. This is in addition to the use of published credit ratings by major rating agencies which are already available in the Bill as introduced.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Calculation date for interest rate on existing loans

Submission

(EY)

We understand the policy intent in relation to existing loans is to recalculate interest as if the new rules had been in place from the commencement of the loan, with any interest restriction applying only for interest relating to an income year commencing on or after 1 July 2018.

Section GC 16(4) provides for a calculation date that is the first balance date of the borrower on or after 1 July 2018 where the loan exists before 1 July 2018. For example, if the borrower has a 30 June balance date, the calculation date would be 30 June 2019 for a loan that commenced, say, 1 April 2017.

The debt percentage measurement in GC 16(1)(b)(ii) or (e)(ii) would appear to be determined based on a 30 June 2019 balance sheet rather than a 1 April 2017 balance sheet.

If a loan exists before 1 July 2018 the borrower’s credit profile and factors in determining that profile need to be determined at the time the loan was entered into and not some subsequent date. The borrower’s credit profile may have subsequently deteriorated. The balance sheet at the balance date proceeding 1 April 2017 would be more appropriate in determining whether the borrower’s debt percentage is less than 40% or within 110% of the worldwide debt percentage.

The calculation date in GC 16(4) applying to existing loans under GC 16(1)(b)(ii) or (e)(ii) should be the balance date immediately before the loan was entered into. This approach is better aligned to arm’s length conditions and takes into account taxpayers’ ease of compliance and the need for certainty at the commencement of the interest limitation rules for existing loans.

Comment

The submitter is correct that GC 16(4), as included in the Bill, would set the calculation date for a borrower with an existing loan and a 30 June balance date as 30 June 2019 by which time the interest limitation rules would have applied to that loan for an entire year. In contrast a borrower with any other balance date which is the last day before the interest limitation rules apply, for example a balance date of 31 July would have a calculation date under GC 16(4) of 31 July 2018. It was not intended that the rules would apply differently for borrowers with a 30 June balance date than all other borrowers.

In considering whether to apply the high BEPS risks tests at the time the loan was entered into or immediately before the new rules applied it was decided to apply the latter. The reason for this is (1) the calculation would be on more recent data and therefore may be easier for taxpayers to calculate; (2) taxpayers who currently do not pose a high BEPS risk would not be required to apply the restrictions; and (3) taxpayers aware that the restricted transfer pricing rules would apply from their first balance date after 1 July 2018 would have a limited period to restructure related party debt so they were no longer a high BEPS risk.

However, officials accept that if the interest limitation rules were in place when the borrower entered into an earlier loan, say 1 April 2017, then a borrower who was not a high BEPS risk at that date would not be required to recalculate this at any later point including the point the restricted transfer pricing rules will apply to that borrower.

To address this issue officials recommend section GC 16(4) is extended so the calculation date for a borrower with a loan that exists before 1 July 2018 should be, at the option of the borrower, either of:

the day before the restricted transfer pricing rules apply; and
the day the arrangement was entered into.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Calculation periods for income-interest ratio

Submission

(EY)

For loans existing before 1 July 2018, the four periods in section GC 16(6) should be taken with reference to the quarters ended or balance dates immediately prior to the dates the loans were originally entered into.

Comment

This issue is very similar conceptually to the calculation dates for existing loans under GC 16(4) which is discussed above. For the same reasons GC 16(6), as included in the Bill, calculates the income-interest ratio for the most recent periods before the calculations are completed.

However, the recommendation elsewhere in this report to remove the EBITDA test makes this submission redundant as the calculation period will no longer be necessary.

Recommendation

That the submission be noted.

Issue: Data for the income-interest ratio calculation

Submission

(Corporate Taxpayers Group)

The calculation period in GC 16(6) refers to data being available “when the calculation is performed”. When is a calculation performed? Who performs the calculation? Is this just when the calculation is first performed or can it be reperformed?

It should be clarified whether this means the calculation includes periods in the subsequent income year if the calculation is being undertaken as part of the tax return process.

Comment

As officials have recommended removing the income-interest test this calculation will no longer need to be performed.

Recommendation

That the submission be noted.

Issue: 12 month period

Submission

(Corporate Taxpayers Group)

The calculation period in GC 16(6)(b) is for the 12-month period ending with the most recent balance date. How does this apply to taxpayers with more than 12 months to prepare a tax return?

Comment

As officials have recommended removing the EBITDA test this calculation will no longer need to be performed.

Recommendation

That the submission be noted.

Issue: Control by co-ordinated group

Submission

(Russell McVeagh)

The Bill should be amended to clarify that where a New Zealand entity is controlled by a co-ordinated group, but one of the members of the co-ordinated group holds more than 50% of the New Zealand entity on its own (and so the entity does have a worldwide thin capitalisation group), then the rule for co-ordinated groups in section GC 16(8) does not apply. Instead, the group credit rating rule in section GC 16(9) should apply (assuming the entity is high BEPS risk).

Comment

The purpose of the co-ordinated group approach is where the New Zealand entity does not have an identifiable foreign parent. The submitter is correct that when a single entity holds more than 50% of the New Zealand entity it is unnecessary for the co-ordinated group rules to apply. In this circumstance a New Zealand borrower with a high BEPS risk should apply the group credit rating in GC 16(9) rather than the restricted credit rating in GC 16(8).

Recommendation

That the submission be accepted.

Issue: Co-ordinated group rules should apply to interposed special purpose vehicles

Submission

(KPMG)

Where a New Zealand borrower is held by an offshore special purpose vehicle (SPV) that is in turn owned by a “co-ordinated group”, the SPV will be considered to be the parent company of the New Zealand borrower rather than the co-ordinated group.

Where a New Zealand borrower is owned by a co-ordinated group via offshore SPVs that do nothing else other than hold shares in the New Zealand borrower (or other entities in New Zealand that are associated with the New Zealand borrower) it should be able to calculate the interest rate on its cross-border related party loans using its standalone credit rating in proposed section GC 16(7) or the restricted credit rating rule in proposed section GC 16(8).

Comment

Officials agree that requiring a New Zealand borrower to use a credit rating based on their worldwide group would be ineffective if the direct owner of the New Zealand borrower is a non-resident SPV (or holding company) owned by a coordinated group and that SPV has no business activity other than owning the New Zealand operations. In this circumstance it would be appropriate to apply the restricted credit rating rather than the group credit rating if the New Zealand borrower was a high BEPS risk.

Recommendation

That the submission be accepted.

Issue: Foreign government shareholders

Submission

(KPMG)

Where a foreign Government is the ultimate shareholder of the borrower, the sovereign credit rating should take preference over published credit ratings of interposed companies in the worldwide group.

Comment

Officials refer to the submission elsewhere in this report that the appropriate entity to base a group credit rating off is the group member with the highest level of external borrowings. Where a foreign Government is the ultimate parent of a New Zealand borrower it is likely that they would also have the highest external borrowings compared with any interposed companies. In this instance the sovereign credit rating minus two notches (assuming the sovereign credit rating is BBB+ or higher) would be the appropriate group credit rating. Officials agree with the submission but do not consider specific legislation is required to achieve this.

Recommendation

That the submission be accepted.

Issue: Local published credit ratings

Submission

(KPMG)

There should be specific allowance for use of a published credit rating of the borrower, where available for the restricted credit rating rule (proposed section GC 16(8) or group credit rating rule (proposed section GC 16(9). This should supersede the need to apply the restricted credit rating or group credit rating rules.

The use of such a credit rating in the limited circumstances where a New Zealand borrower may have one would appear to be consistent with the overall purpose and intent of the Bill which is to reduce the level of subjectivity in the existing rules. Such a credit rating would also arguably achieve a more arm’s length result as it has regard to the credit metrics of the borrower, as considered by third parties. It therefore naturally flows that a borrower’s published credit rating should be acceptable and preferable to that of the worldwide group.

Comment

The use of a published credit rating for the NZ borrower is already included in proposed sections GC 16(8)(a) – subject to a BBB- minimum – and GC 16(9)(a). The availability of such a rating does not remove the necessity of considering whether GC 16(8)(b) or GC 16(9)(b) may arrive at a higher rating. Although use of a published credit rating would remove a level of subjectivity it would not remove the ability of a group to manipulate the New Zealand borrower’s credit rating, for example by including more debt; particularly when that New Zealand borrower had not used that credit rating to obtain third party debt.

Recommendation

That the submission be declined.

Issue: Safe harbour for loans under $10 million

Submission

(KPMG, PwC)

As drafted, the Bill requires taxpayers with cross-border related party loans to adopt the credit rating that the taxpayer has for long-term senior unsecured debt. This applies regardless of the loan value, including for “low value” loans of $10 million or less. (KPMG)

To reduce compliance costs, Inland Revenue currently has a safe harbour that allows inbound related party loans under $10 million to be priced with reference to a relevant base indicator plus a margin. The safe harbour is widely applied and should be maintained. As currently drafted the restricted transfer pricing rule would require even loans under $10 million to be priced by reference to a credit rating. (PwC)

Comment

The intention of the “borrower’s credit rating” safe harbour in proposed section GC 16(7) was that the restricted transfer pricing credit adjustment would not apply when a borrower does not have a high BEPS risk. This was not intended to change the approach of a low BEPS risk taxpayer which is already required to follow transfer pricing analysis, or the administrative method for loans under $10 million. As noted elsewhere in this report this can more efficiently be achieved by removing the borrower’s credit rating provision and instead relying on the standard transfer pricing rules in sections GC 7 to GC 14. This should provide the clarification, and continued access to the administrative method, requested by the submitters.

Recommendation

That the submission be accepted.

Issue: Application of $10 million de minimis

Submission

(Corporate Taxpayers Group)

Does the $10 million de minimis in section GC 16 (1)(a) apply to individual loans or all loans? Do taxpayers assess a loan under this section at the time the loan is taken out and they do not have to revisit for the life of the loan if this rule is satisfied?

Comment

The de minimis applies to all loans at the calculation date for the particular loan being taken out. This prevents a borrower being able to structure around the rules by taking out multiple loans each of less than $10 million.

As the interest rate, or the margin, is set at the time the loan is entered into; if the de minimis applies at the time the pricing is set, this de minimis will apply for the term of the loan. If a subsequent loan is entered into without repaying the original loan such that total loans now exceed the $10 million threshold this would result in the de minimis applying to the original loan but not the subsequent loan.

Recommendation

That the submission be noted.

Issue: Compliance costs

Submission

(EY, PwC)

Before traditional pricing analysis commences, analysis will be required to determine whether a company is at a high risk of BEPS. Further detailed analysis of external group debt instruments is required if a loan has a term of more than five years or other features that may be disregarded. This will increase compliance costs for companies with inbound loans over $10 million. (EY)

Taxpayers should be given the option of a low compliance cost approach (e.g. allowing a default credit rating of BBB- with plain vanilla loan terms). The so-called “safe harbour” provided in the Bill as drafted is not a true safe-harbour – it effectively requires all members of the multinational group to be credit scored so is not a low cost compliance approach. (PwC)

Comment

The BBB- minimum rating is not available outside of high BEPS risk coordinated groups as in most instances this will not be an appropriate rating for a New Zealand group. Even for co-ordinated groups this is a minimum rating that the borrower cannot go below rather than a rating available to all borrowers irrespective of their circumstances. For higher rated borrowers allowing a safe harbour of BBB- could result in a significantly lower credit rating than their standalone or group credit rating.

While these proposals will result in some taxpayers incurring higher compliance costs this will not always be the case and the rules have been developed to minimise compliance costs where possible. For example the three high BEPS risks tests (which are recommended to be reduced to two elsewhere in this report) requires data that should easily obtainable and necessary for other purposes. For the majority of borrowers who are not a high BEPS risk no credit rating adjustment will be required. Where taxpayers do not include disregarded features in their loans these will not need to be considered in determining the appropriate price which will decrease compliance costs. Where taxpayers are a high BEPS risk the use of the group credit rating will eliminate much of the subjectivity, and potential for dispute, around the degree of implicit support which is also likely to reduce compliance costs.

A number of other recommendations in this report also lower compliance costs compared with the introduced Bill. These include:

Where related party loans are less than $10 million the existing administrative method will continue to be available.
Where a taxpayer is a low BEPS risk the amendments recommended in this report will allow them to continue to use their existing transfer pricing methodology; and
Where a taxpayer is a high BEPS risk and has to consider the group credit rating the recommendations in this report would limit this to considering only the rating of the entity within their worldwide group that had the highest amount of third party debt.

Recommendation

That the submission be declined.

Issue: Calculation dates

Submission

(PwC)

It is unclear why the calculation date in section GC 16(4) is only specified for limited subsections – all of the criteria in (a) to (e) need to be tested at a particular date and should have a calculation date specified.

Comment

The calculation date in GC 16(4) applies for calculations under section GC 16(1)(a), (b)(ii) or (e)(ii). These three provisions are whether the de minimis applies, and the debt percentage for borrowers controlled or not controlled by a co-ordinated group respectively.

For subsection GC 16(1)(b)(iii) and (e)(iii), which calculate the residence of the lender, there is currently no reference to a calculation date therefore listing these provisions within GC 16(4) would not in itself be effective. However, there is a possibility that these provisions could be met on one date and not met at a different date therefore a calculation date should be provided. This will require amendments to both of GC 16(1)(b)(iii) and (e)(iii) to refer to a calculation date in GC 16(4) as well as including these provisions in GC 16(4).

As the EBITDA test is recommended to be removed a calculation date will not be required for section GC 16(1)(b)(iv) and (e)(iv).

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Restricted credit rating debt percentage

Submission

(PwC)

The reference to the debt percentage of the New Zealand group in section GC 16(8)(b) should be removed as the debt percentage of the group must be above 40% for subsection (8) to be relevant.

Comment

The restricted credit rating in GC 16(8) applies to a borrower that is a high BEPS risk and controlled by a co-ordinated group. There are three tests to be a high BEPS risk in the Bill (and officials recommend reducing this to two) with only one of these being a debt percentage over 40%. As a borrower will be a high BEPS risk if they do not meet the residence of the lender (or under the bill the EBITDA) test it is possible for a borrower to be a high BEPS risk despite having a debt percentage below 40%.

The submitter is correct that if a borrower controlled by a co-ordinated group has a debt percentage above 40% then 40% will always be lower than their actual percentage. However, if a borrower controlled by a co-ordinated group is a high BEPS risk but has a debt percentage below 40% the current drafting will ensure their actual debt percentage is taken into account for the purpose of section GC 16(8)(b).

Recommendation

That the submission be declined.

Issue: Borrower’s credit rating

Submission

(Corporate Taxpayers Group)

The borrower’s credit rating in GC 16(7) is inconsistent with the Commentary (page 11). This should be a reference to the parent credit rating not the highest rating within a worldwide group. This is an impractical test which many taxpayers will not be able to comply with as not all members of a worldwide group will have credit ratings (or may be impractical to identify the highest rating in the entire group).

Comment

The borrower’s credit rating reflects the credit rating of the New Zealand borrower rather than the worldwide group. This is the rating that can apply when the borrower is a low BEPS risk or is below the $10 million de minimis. This is consistent with page 11 of the Commentary, specifically the second bullet point.

The submitter’s comments in relation to identifying the highest rating in the entire group only apply to the group credit rating and have been addressed elsewhere in this report.

Recommendation

That the submission be declined.

DISREGARDED FEATURES

Issue: List of features

Submission

(Chartered Accountants Australia and New Zealand, PwC)

Proposed section GC 18(2) should be excluded. It is not correct to say that the features listed in proposed section GC 18(2) always exist for tax reasons. There may be genuine commercial reasons for these terms. If Government wishes to include these elements in legislation they should be part of a rebuttable presumption only and should not be determinative of an outcome. However, it is our strong preference that these factors are present in official guidance only and not part of the legislation. (CA ANZ)

The factors in section GC 18(2) should not be hard coded and specified in legislation because:

it is unnecessary to do so if the general transfer pricing regime is strengthened as anticipated (which will in future require non-commercial terms to be disregarded);
legislating for these factors does not allow for flexibility to take account of the individual commercial circumstances of the taxpayers or changes to general market conditions;
legislating for a 5 year term is inappropriate (and loan term is not a factor the ATO takes into account) – there are many situations where it is entirely appropriate for debt to remain outstanding for longer than 5 years, such as where funding is required to match an investment profile of longer than 5 years (for example, in a fund or private equity context where assets are held for long-term investment, or in the forestry sector) but where there will not be other external debt;
inclusion of other factors such as payment-in-kind interest have not been properly justified. These types of loan features are incorporated in situations where investment is riskier and cashflow may be used for other things in early years (e.g. expansion), whether the lenders are related parties or third parties. (PwC)

Comment

Officials recommend that subparagraphs GC 18(2)(a) and (b) are removed as set out elsewhere in this report. However, the remainder of the features should continue to be included. Without providing legislative guidance on the type of features that may be disregarded taxpayers will have less certainty over the type of features this section is targeted at. These features have been collated based on the practical experience of implementing the existing law and what are commonly observed features of third party and related party loans.

Recommendation

That the submission be declined.

Issue: Long-term loans with shorter term interest rate resets

Submission

(ANZ Bank, Corporate Taxpayers Group, New Zealand Bankers’ Association)

Loans with a term of greater than 5 years but have interest rates re-priced more regularly should not be limited to have interest set every five years. Such debt should not be caught by the 5 year proposals as regular re-pricing removes any risk that the long-term loan presents.

Comment

It is possible, although less common, for loans to have a term exceeding five years but for the interest rate to be set at more frequent intervals, for example annually. The submitter is querying how the proposed rules will apply in this circumstance.

Money lent with a long, or perpetual, term with an annual interest rate reset faces an identical price risk to a one year loan that is expected to be replaced by a new one year loan upon its maturity. This is because any shift in the market rate of interest during that year can be factored into the applicable interest rate either upon the reset or as part of the new loan. Provided this reset happens within a period of five years or less there should be suitable comparables for this price risk to be accurately incorporated into transfer pricing calculations. Accordingly the proposals do not adjust this calculation.

However, depending on the terms of the instrument, the lender on a long term instrument may be subject to other risks, such as increased credit risk. For example after the expiry of a one year loan a lender may decide the borrower’s credit risk has deteriorated and choose not to renew the loan whereas if the loan was perpetual this would only be available to the lender if the terms allowed them to demand repayment. To the extent the term of the loan was greater than five years, even though the interest rate was reset more frequently, and this feature increased the interest rate beyond that which would apply to a loan with a five year term with equivalent interest rate resets this feature should be disregarded.

If the term and the period the interest rate are fixed for are both beyond five years, section GC 18(2) will adjust both. If only the term is beyond five years but the interest rate is fixed for a shorter period section GC 18(2) will adjust the term without forcing the interest rate to be fixed for five years. Officials consider the drafting already achieves this.

Recommendation

That the submission be declined.

Issue: Total debt terminology

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, PwC)

Proposed section GC 18(6)(c) uses the term “total debt”. From a financial perspective this generally means for an entity “the sum of current liabilities and long term liabilities”. However, while it is not entirely clear, as a consequence of the interaction of GC 18(4)(a) it seems “total debt” in proposed section GC 18(6)(c) means total debt that has a term of more than five years.

The drafting of sections GC 18(4) to (7) is too complicated and it is very difficult to work out whether the term of a loan needs to be adjusted to five years. The formula in subsection (5) does not appear to give the intended result without the inclusion of the “threshold fraction”. (PwC)

Comment

While the definition of total debt currently refers to total debt rather than total debt over five years, as submitters have noted, the formula would not operate correctly unless the denominator was total debt with a term of more than five years. Officials consider the current drafting already achieves this as the threshold term in section GC 18(4)(a) already limits the calculation to financial arrangements having a term of more than five years. However, as a number of submitters have noted that this is unclear officials recommend the drafting be clarified to ensure taxpayers calculate total debt only for arrangements with a term more than five years.

As noted elsewhere in this report, the third party exception for terms over five years is more complicated than the third party exception for other disregarded features at it needs to take into account that all loans with a term over five years cannot be considered homogenous. Officials consider the approach in the bill, once the clarification above is included, strikes an appropriate balance between accuracy and complexity.

Recommendation

That the submission to amend the definition of total debt be accepted.

Issue: Commissioner’s discretion

Submission

(Chartered Accountants Australia and New Zealand, EY)

The rules should allow for the Commissioner to recognise otherwise disregarded terms and conditions of loans to the extent she believes that recognition is consistent with the purpose and intent of the rules which is to allow deductions for interest on loans with arm’s length terms and conditions.

Comment

The rules include a third party exception so a borrower who has a commercial reason to include certain features can also include these features in related party debt. The requirement that the related party debt can be up to four times the size of third party debt with that feature means a borrower would only need 20% of their total debt to be from third parties and this debt would be equivalently priced to related party debt.

Allowing features to be included in pricing when there was no equivalent third party debt but a claimed commercial reason to do so would result in legislation with a very similar effect to the current position where the inclusion or not of certain features is very subjective and borrowers will often claim commercial justifications for tax driven decisions.

Recommendation

That the submission be declined.

Issue: Terms greater than five years

Submission

(Chartered Accountants Australia and New Zealand)

The requirement that the loan term be less than five years (or is assumed to be) is unduly restrictive. There are examples of commercial loans in the current market with terms of up to ten years. It should be open to a taxpayer to demonstrate that an independent person in similar circumstances would have had a loan term longer than five years. That is, the loan term should form part of a rebuttable presumption only.

In addition, a five year term ignores specific industries. For example, forestry and mining industries typically require long term loans of more than five years.

Comment

Where an independent third party has provided loans with a term more than five years this will be available to allow the borrower to also have related party debt with these terms. This is much less ambiguous than a view that a third party would provide funding for greater than five years even though none have done so. Where a borrower is funded for more than five years through related party debt and there is no longer term third party debt, the related party debt starts to more closely resemble the qualities of equity. The proposals do not prevent a borrower having related party debt for greater than five years but the tax deductions will more closely reflect a more likely commercial outcome of loans entered into for five years then rolled over for a further period.

Recommendation

That the submission be declined.

Issue: Terms greater than five years related party exception

Submission

(Chartered Accountants Australia and New Zealand)

The third party exception in proposed section GC 18 should be extended to include a single related party loan with a term of more than five years that has an interest rate reflecting the proportionate global funding mix. This is instead of the current proposal that requires related party debt to be structured in separate ‘less than five year’ and ‘greater than five year’ tranches matching the external debt to avoid all of the debt being forced to five years. Any adjustment to force to five years should only be on the balance that is disproportionate.

Comment

If the global funding mix is a mixture of loans above and below five years it would not be equivalent to allow a single loan to New Zealand with a term over five years. Although this could be given a price that reflected the worldwide cost of funds this would be more arbitrary given it would have to reflect the cost of funds and the makeup of funds at a particular point in time rather than being based on the term and risk of the funding actually provided.

Recommendation

That the submission be declined.

Issue: Exception from five year limit

Submission

(Powerco)

The exception from the five year limit for related party loans provided for groups with external loans with longer terms is a positive step although the exception is complex for groups with large multinational structures.

Comment

Officials note the support. There is an unavoidable increase in complexity for the small number of borrowers who will rely on the five year exception for third party loans of the worldwide group. However, these rules (subject to the amendments recommended in this report) have been designed to minimise this to the extent possible. Without knowing the amount and proportion of third party debt with terms over five years it would not be possible to determine whether the amount of related party debt with terms over five years is appropriate.

Recommendation

That the submission be noted.

Issue: Proportionality requirements in exotic terms

Submission

(Russell McVeagh)

Further work is needed on the operation of the proportionality requirements for exotic terms as the current “threshold” approach can produce anomalous outcomes. For example, currently it appears that a loan with an eight year term must (assuming 50% of third party debt has an eight year term and 50% a five year term) be priced assuming just a five year term, because the proportionality threshold has been breached. It would be better if a weighted average loan term could be used (eg, 6.5 years) or if 50% of the related-party loan could be priced at 8 years. There would be merit in having a regulation-making power and determination-making power to deal with outcomes such as these.

Comment

The proportionality threshold is necessary so that a worldwide group does not borrow third party debt with terms that makes it more expensive then onlend a portion of this debt to New Zealand that is greater than the New Zealand borrower’s proportion of the overall business.

In the situation raised by the submitter, the borrower would not be able to include the eight year term in the pricing of the loan as 100% of their related party lending would be for a term more than five years whereas only 50% of their third party debt would be for over five years. While it would be possible to amend the relevant provisions to allow for such a scenario this would introduce an additional layer of complexity to provisions that have already been raised by other submitters as being complex. In this situation, as the loan is between related parties it would presumably be possible for the New Zealand borrower to have two loans, one for five years and a second for eight years. This approach would more closely reflect the underlying borrowing from third parties.

Recommendation

That the submission be declined.

Issue: Payment-in-kind

Submission

(PwC)

Section GC 18(2)(a) and (b) does not appear to catch “payment-in-kind” or deferred interest in all cases because capitalised interest is treated as “paid” under the Income Tax Act. Intention should be clarified.

Comment

Officials agree that the wide definition of “pay” in the Income Tax Act makes section GC 18(2)(a) not entirely effective. Capitalised interest is not the focus of this provision as it is not practical to distinguish between capitalised interest and new borrowing used to pay interest. Paying interest by capitalising interest does not necessarily increase the risk to the borrower other than by the increased exposure from the value of the loan increasing. The intention of this provision is to restrict an increase in risk of a loan by a borrower paying interest or repaying principal other than in money – for example by providing shares in the borrower. Officials recommend this provision is amended so that the provision of value must be in money, excluding money’s worth.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Deferral of interest

Submission

(Corporate Taxpayers Group)

What time period does the deferral in section GC 18(2)(b) refer to?

Comment

Payment of interest in arrears is a common feature of many loans. It is significantly less common for this deferral to exceed 12 months. Officials agree that section GC 18(2)(b) should refer to a time period and consider this period should be a deferral of a liability to pay interest of more than 12 months.

Recommendation

That the submission be noted and officials’ recommendation be accepted.

Issue: Calculation dates

Submission

(PwC)

The definition of calculation date in section GC 18(4), which refers to the calculation date for terms longer than five years, is not consistent with the definition in GC 16(4) which refers to the calculation date for the high BEPS risk tests.

Comment

The definition of calculation date in GC 16(4) is referenced to in some of the high BEPS risk tests in GC 16(1). The use of calculation date in GC 18(4) is a nonce term used only in the definition of term debt in GC 18(6)(b). There is no requirement for these two uses of the same term to be identical and each usage arrives at the intended policy outcome.

Recommendation

That the submission be declined.

Issue: Term of the loan approach

Submission

(PwC)

Term of the loan should not be included in subsection GC 18(2)(f) as it is dealt with separately in subsection (3), and it is not clear which set of rules takes priority. Alternatively, subsection (3) and related provisions could be deleted in favour of subsections 2(f) and (8)

Comment

Subsections 2(f) and (8) would not reach the correct outcome without section GC 18(3) to (7) as these set out the third party exemption to the five year term. Without these provisions all loans over five years would be treated equally. For example, if a borrower had a third party loan with a six year term that met the necessary criteria this could be used to justify a related party loan with a term of 60 years.

Including the term of the loan within GC 18(2) reduces complexity for the majority of borrowers who do not have third party loans with a term of over five years. It is also consistent with the approach for the other disregarded features which are included in GC 18(2) unless a third party loan with that feature qualifies for an exception under GC 18(8).

Recommendation

That the submission be declined.

Issue: Third party exception for acting together

Submission

(PwC)

The third party exception in section GC 18(8) does not appear to apply to loans advanced by parties “acting together”/a “non-resident owning body” (rather than associated) which presumably is not intended.

Comment

The third party exception allows a disregarded feature to not be disregarded if the New Zealand borrower or worldwide group has significant third party debt with this feature. When a New Zealand borrower is controlled by parties acting together or a non-resident owning body there is no identifiable parent and therefore no worldwide group.

This is the same problem encountered with the 110% debt percentage test in the thin capitalisation and high BEPS risk tests in the restricted transfer pricing rules. The thin capitalisation rules approach this by treating the New Zealand group as the worldwide group in section FE 31D. This approach would not be effective for the third party exception as there is already a New Zealand group test. In the restricted transfer pricing rules a worldwide group test is omitted from section GC 16(1)(b)(ii) which applies to co-ordinated groups compared with its inclusion in the equivalent section GC 16(1)(e)(ii) that applies to non-co-ordinated groups.

As there is no worldwide group it is correct that a worldwide group third party exception cannot apply to co-ordinated groups.

Recommendation

That the submission be declined.

INSURING OR LENDING PERSONS

Issue: General support for rules

Submission

(IAG)

The Bill treats insurers as a special case when applying the restricted transfer pricing rules to limit interest deductibility. We agree that the insurance industry has special features and that a bespoke rule is necessary.

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: Insuring or lending persons applying the same credit rating as their parent

Submission

(ANZ, ASB Bank, New Zealand Bankers’ Association, PwC, Westpac)

We understand from discussions with officials that the intention of the proposals is not to override observable market interest rates relevant to the New Zealand borrower with the foreign parent’s borrowing cost. However, to mitigate this potential inconsistency and resulting uncertainty in application of the proposals, it would be beneficial for guidance to be issued confirming that the appropriate transfer price will not automatically be the foreign parent’s borrowing cost but rather should reflect an appropriate comparable borrowing cost for a New Zealand borrower to unrelated parties (albeit after taking into account the effect of the interest limitation proposals). (ANZ)

There is market observable evidence that where New Zealand banks issue debt into the same market that their ultimate parent is issuing similar debt into, that the New Zealand bank must pay a higher rate of return to attract investment. This suggests that the market itself does not assume that the New Zealand bank has the same creditworthiness as its parent. This would appear to contradict the approach taken in the Bill to require the New Zealand bank to adopt the parent rating. It is notable that in a number of cases the New Zealand banks have their own credit ratings that are one notch below their parent ratings. (ASB)

Not all members of the NZBA have the same credit rating as their offshore parents. Restricting the ability for such banks to use, for example, their own credit rating would be unfair compared to other taxpayers who can use their own credit rating and does not reflect the market position prevailing for such banks (where they borrow from the market at rates reflecting a different credit rating than that of their parent). (NZBA)

The requirement for an “insuring or lending person” to base its credit rating on the highest rating within the group does not make sense given a number of these financial institutions will have their own formal credit rating. Alternatively, in instances where the New Zealand borrower does not have a formal credit rating, they should be able to apply a one notch downgrade to the group rating, consistent with the approach applied to other corporate taxpayers. (PwC)

Credit rating agencies will sometimes give subsidiary financial institutions in New Zealand a lower credit rating that that of its parent. This is relevant because, in the banking context, the prudential and regulatory framework under which banks operate does not permit parental support. Section GC 17 should therefore reflect this market reality and allow banks to use the same alternatives available to other taxpayers. (Westpac)

Comment

The proposals in section GC 17 require an insuring or lending person to apply the same credit rating as their parent but do not necessarily require them to have the same credit worthiness. This difference arises as a credit rating provides a range of possible prices, bounded by that credit rating, rather than a single appropriate price. Officials agree that there is market observable data of NZ banks issuing similar debt at a higher margin than their foreign parents despite both entities having the same credit rating. These differences in pricing will continue to be available to insuring or lending persons, within the bounds of a single credit rating.

Officials note that the majority of New Zealand banks – including the four Australian banks that provide the majority of lending in New Zealand’s banking sector – all have the same credit rating as their foreign parent in which case section GC 17 will impose no practical restriction for borrowing by the registered bank. Requiring an insuring or lending person to have the same credit rating as its parent reflects the different structure of financial institutions where the New Zealand operations are more integral to their worldwide group. It is highly unlikely a worldwide group would choose to let their New Zealand financial institution subsidiary fail as to do so would reflect on the reputation of the entire group even more so than in a non-financial institution.

However, as noted elsewhere in this report, officials agree with submitters that a credit rating should be able to be set with reference to the implied credit rating of significant senior third party debt. This proposal should apply equally to insuring or lending persons which should resolve some of the submitters concerns on this issue.

Recommendation

That the submission be declined.

Issue: Third party tests for insuring or lending persons

Submission

(ASB Bank, New Zealand Bankers’ Association, Westpac)

The interest limitation proposals inappropriately treat the financial services industry differently from other taxpayers by limiting the ability for this industry to apply certain of the exceptions within the proposals.

The proposed restricted transfer pricing rules require certain contractual terms or features to be ignored in determining the arm’s length price. There are exceptions to this in proposed sections GC 18(8) and (9). Paragraph 8 applies to general taxpayers while paragraph (9) applies to banks and insurance companies and similar. While we agree with the direction of the paragraph 9 exceptions which acknowledge the fact that often the contractual terms may be driven by regulatory capital requirements, it is conceptually unclear why banks cannot also use the paragraph 8 exceptions where the terms of the related party debt reflect third party terms. Related party debt may be used to provide general funding, not just regulatory capital funding, and where the terms of non-regulatory capital funding provided by related parties nevertheless reflects third party funding terms, the paragraph 8 exceptions should be available to banks also.

Comment

Third party debt of a bank is much more likely to be senior debt without exotic features – as it is the cheapest form of funding – or regulatory capital – as it meets Reserve Bank requirements and reduces the risk of senior lenders not receiving their full investment return. In general there would be little commercial benefit in borrowing money at a higher interest rate than necessary without receiving any regulatory capital recognition. If a New Zealand bank had no non-regulatory capital third party debt with exotic features, there would be no benefit in section GC 18(8) also applying to them. This was part of the trade-off considered when including the exceptions in GC 18(9) of which there is no equivalent for non-insuring or lending persons.

Recommendation

That the submission be declined.

Issue: Definition of “insuring or lending person” is unclear

Submission

(PwC, Corporate Taxpayers Group)

The proposed definition of “insuring or lending person” (proposed section GC 15(2)) is too wide, leading to complexity on application in determining whether certain clients will have “a main business activity of providing funds” (for example the leasing of aircraft, commercial property and motor vehicles).

Paragraphs (d) and (e) are not drafted consistently with the Bill Commentary and appear to catch all group members rather than a member whose main activity is lending to non-associates.

Comment

The definition of “insuring or lending person” is necessary to apply the different treatment under the proposals. The submitter is specifically querying subsections (d) and (e) of the proposed definition. These subsections are aimed at groups and individual taxpayers respectively that have a main business activity of providing funds to non-associated persons. For example a person who was a commercial property developer who also arranged financing so their customers could purchase a property(s) after they had been completed would need to consider whether the majority of their time and profits came from the property development or from the financing. The group would only be an insuring or lending person under subsection (d) if their main activity was financing of commercial property rather than developing commercial property. If the financing was undertaken by a separate subsidiary within the same group this financing subsidiary would be an insuring or lending person under subsection (e) but the rest of the group would not be unless that financing subsidiary was the main business activity of the entire group.

If a borrower meets paragraph (e), and does not meet paragraph (d), then that borrower will be an insuring or lending person but the rest of the group will not be. Officials recommend a drafting change to paragraph (e) to ensure this is clear.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Definition of insuring or lending person – securitisation trusts

Submission

(Corporate Taxpayers Group)

Separate consideration should be given to structured finance entities (including securitisation trusts). It is not clear how to measure the ‘group of persons’ where a trust is involved. There should be a separate rule to carve out structured finance entities.

Comment

When a group operates a securitisation trust as part of its wider business it would be appropriate for this to be included within a group of persons as this term is used in section GC 15(2). If this were not the case a securitisation trust would be unlikely to be treated as an insuring or lending person, even when their wider group was, as they will be raising funds for their own group rather than for non-associated parties.

The majority of securitisation will be with third parties and even where it is with related parties this will often be able to rely on the third party exemptions in the proposed rules or be between two New Zealand entities so that the restricted transfer pricing rules will have no effect. Developing a comprehensive rule for securitisation trusts that was appropriately targeted without excluding unintended entities would require further consideration and consultation and cannot be completed within the select committee process for this Bill.

Recommendation

That the submission be declined.

Issue: Long-term credit rating in section GC 17

Submission

(ANZ Bank, Corporate Taxpayers Group, New Zealand Banker’s Association, Westpac)

Proposed section GC 17 should be amended to require the highest credit rating for “long term” senior unsecured debt of a member of the borrower’s worldwide group to be applied to the New Zealand borrower.

Comment

This section currently refers to the highest credit rating for senior unsecured debt of a member of the borrower’s worldwide group. There will frequently be a different credit rating for long term compared to short term debt. Adding the words “long term” would make this test consistent with the other equivalent tests in the restricted transfer pricing rules.

Recommendation

That the submission be accepted.

Issue: Credit rating of insurers

Submission

(EY, IAG)

An insurance operating company will typically have a higher issuer credit rating than a non-operating holding company in the same group. This difference reflects the structural subordination of the creditors in a holding company to those in a regulated insurance operating company. Where group funding is through a holding company of a special purpose finance company, the relevant cost of borrowing would reflect the holding company credit rating, not the operating company rating. A New Zealand insurer using a higher credit rating relating to an insurance operating company in the group would result in the New Zealand insurer paying an interest rate on related party funding that would be below the group cost of borrowing.

Comment

Officials note the recommendation elsewhere in this report that the reference to the highest rated member of the group be replaced by the entity with the highest amount of unsecured third party funding. In most circumstances a group will choose to borrow through the highest rated member of the group – to minimise their external borrowing costs – but where this does not happen, for example for regulatory reasons, the group rating would be determined by the rating of that funding entity rather than the higher rated operating company.

An insurer with significant third party debt in their New Zealand group would also be able to rely on the implied credit rating from this debt, provided it was senior-ranking and unsecured, as recommended elsewhere in this report.

Recommendation

That the submission be noted.

Issue: Reserve Bank capital requirements

Submission

(ANZ Bank, ASB Bank, BNZ, Corporate Taxpayers Group, New Zealand Bankers’ Association, Westpac)

Proposed section GC 18(9) should be amended to refer to features of bank regulatory capital imposed by regulations set by the Reserve Bank of New Zealand as opposed to the conditions of registration under the Reserve Bank of New Zealand Act 1989

Comment

This section allows certain features to be included in pricing a loan if those features are included to meet regulatory capital requirements imposed by the Reserve Bank. As noted by submitters, these requirements are currently imposed by the relevant Reserve Bank Regulations which are currently BS2A (Capital Adequacy Framework (Standardised Approach)); and BS2B (Capital Adequacy Framework (Internal Models Approach)). The intention of the wording in the bill was to capture these regulations without specifically referring to them so the legislation continued to apply if and when these regulations were replaced.

Officials agree that the submitters’ proposed change will improve the clarity of this provision.

Recommendation

That the submission be accepted.

Issue: Regulatory capital should be exempt from the rules

Submission

(Westpac)

What can be classified as bank regulatory capital is determined by the Reserve Bank of New Zealand (RBNZ). Given the characteristics of such instruments, the market for bank regulatory capital issued by New Zealand banks to third parties is limited in capacity. This is not the case for non-bank taxpayers, who have greater choice in how their debt capital is structured and/or sourced. In addition, non-bank taxpayers have much more flexibility to restructure their existing debt in response to the new rules, whereas the RBNZ imposes substantial limitations on a bank’s ability to restructure its regulatory capital (which would also be prohibitively expensive). The proposed rules therefore have a discriminatory effect against banks who are tied into long term regulatory capital structures by the RBNZ.

Bank regulatory capital should be exempt from the new rules. Failing that, we submit that existing issues of bank regulatory capital should be grand-parented in the same manner as those under the new Hybrid mismatch rules in the Bill.

Comment

The rules include exemptions for features included in regulatory capital, including instruments that were issued as regulatory capital but no longer qualify, as well as back-to-back instruments used to fund regulatory capital. Officials are not aware of any current regulatory capital instruments that would have their price altered as a result of the restricted transfer pricing rules but if this was the case it would be due to either the credit rating being lower than either the group rating or rating on third party debt or where terms were included that were not required for the instrument to qualify as regulatory capital. If this was the case officials consider it would be appropriate for the price of these related party instruments to be adjusted consistent with what would apply to the equivalent non-regulatory capital instruments in a similar situation.

Recommendation

That the submission be declined.

Issue: Regulatory capital that is an excepted financial arrangement

Submission

(ANZ Bank, Corporate Taxpayers Group, New Zealand Bankers’ Association, Westpac)

Proposed section GC 18(9)(b) should be amended so that the features of a funding arrangement are not disregarded or adjusted if that funding arrangement provides funds in regulatory capital which may include an excepted financial arrangement

Comment

Section GC 18(9)(b) allows a banking group to include a feature in a cross-border loan from a related party if those features are also included in a back-to-back loan that is regulatory capital of the New Zealand bank. This is currently restricted to both arrangements being financial arrangements. The submitters point out that a cross-border loan can be used to finance the New Zealand entity investing preference shares into the New Zealand bank.

It is relatively common for a New Zealand bank to have a related party loan to its New Zealand holding company which then invests in equity of the New Zealand bank. The Reserve Bank imposes regulatory requirements on the New Zealand bank and its subsidiaries but not on its holding company.

Officials agree that, for the purpose of the bank-to-back provisions a cross-border loan that is used to finance preference shares should be treated consistently with a cross-border loan used to finance a financial arrangement.

Recommendation

That the submission be accepted.

Issue: Matching features of a back-to-back loan

Submission

(ANZ Bank, ASB Bank, Corporate Taxpayers Group, New Zealand Bankers’ Association)

Clarification is required on proposed section GC 18(9)(b)(ii) and GC(b)(iii) on what is required for the features of funding arrangement to “reflect” the features of the funded arrangement as well as the requirements for regulatory capital. The features of the loan should not need to be completely identical, it should be sufficient that the relevant terms are similar to the regulatory capital terms.

Comment

Officials note that this submission should apply equally to sections GC 18(9)(d)(ii) and (iii) for insurers and GC 18(9)(f)(ii) and (iii) for non-bank deposit takers.

The purpose of the back-to-back provisions in proposed section GC 18(9)(b), (d) and (f) is so an interposed resident entity can have similar terms on a loan (referred to as the funding arrangement) to the terms imposed by the Reserve Bank on the groups regulatory (or solvency) capital (referred to as the funded arrangement). Officials acknowledge that the features of two separate arrangements, albeit entered into for the same wider purpose, may not have identical features. This will particularly be the case in the scenario discussed above where a New Zealand company borrows to invest in preference shares.

For a particular feature that doesn’t match between the funding and funded arrangement, this raises three possibilities:

A feature is present in the funded arrangement but not in the funding arrangement – the feature is not present in the cross-border loan so no further action is necessary. The absence of the feature from the funding arrangement should not disqualify the entire loan from being a back-to-back loan for the purpose of considering other features.
A feature is present in the funding arrangement but not in the funded arrangement – the feature is not necessary to meet a regulatory capital requirement so should not be included in considering the pricing of the funding arrangement. Again the absence of the feature from the funded arrangement should not disqualify the entire loan from being a back-to-back loan for the purpose of considering other features.
A feature does not exactly match between the funding and funded arrangements – the example provided by the submitter is the funded arrangement has a conversion to equity feature while the funding arrangement has a mandatory repayment feature.

It is the third of these possibilities that could cause confusion. Officials consider there should not need to be perfect mirroring between features of the funding and funded arrangements but this should not provide the opportunity for carte blanche inclusion of any features in the funding arrangement under the pretext of imperfect mirroring. We consider an appropriate approach would be to include a feature in a funding arrangement, that does not exactly match a feature in a funded arrangement, only if the feature provides protection to the banking/insuring/deposit taker associate in a similar circumstance to that faced under the funding arrangement. Furthermore the increase in price due to the feature in the funding arrangement should not exceed the increase in price due to the feature in the funded arrangement.

Recommendation

That the submission be noted.

Issue: De minimis

Submission

(Deloitte)

The $10 million de minimis for non-insuring or lending persons in section GC 16 and disregarded features in section GC 18 should also be available for insuring or lending persons in section GC 17.

Comment

There is no restriction on entity types applying the existing $10 million administrative practice as the same compliance cost compared with risk of excess interest rate trade-off applies. Officials agree that this approach should continue to be available for insuring or lending persons and this can be achieved by including a $10 million de minimis for related-party cross-border lending in section GC 17.

Recommendation

That the submission be accepted.

THIN CAPITALISATION – EXCLUSION OF NON-DEBT LIABILITIES

(Clause 24, 25, 28)

Issue: Should the proposal to subtract non-debt liabilities from value of assets in calculating debt percentages for the non-bank thin capitalisation rules proceed?

Submissions

(Chapman Tripp, Chartered Accountants Australia New Zealand, EY)

Some submitters (Chapman Tripp) said that the proposal to subtract the value of non-debt liabilities from a firm’s asset value for purposes of the thin capitalisation rules should not proceed. It is unnecessary, has no connection with BEPS, and the current “no subtraction” approach does not currently result in taxpayers exceeding commercial levels of debt. Along with other measures, it will have the effect of materially reducing the 60% safe harbour threshold.

Other submitters supported the proposal in principle (Chartered Accountants Australia New Zealand, EY)

Comment

Submitters are correct that this proposal is not part of the OECD’s BEPS recommendations. However, those recommendations do include rules to limit interest deductions of a multinational group, which is exactly what this proposal is concerned with.

The proposal ensures that the thin capitalisation rules better measure the extent to which a group has used debt funding rather than equity funding to fund its New Zealand operations. Currently, the rules treat funding by way of non-debt liabilities as though it were equity. This is not the appropriate treatment.

Officials add that ever since the inception of Australia’s thin capitalisation rules, those rules have subtracted non-debt liabilities in the same way as proposed in the Bill.

Recommendation

That the submissions that the proposal should not proceed be declined. That the submissions in support of the proposal be noted.

Issue: Should the exclusion of non-debt liabilities reflect how they would be treated by a third party lender

Submissions

(PwC)

PwC submitted that the treatment of non-debt liabilities proposal should better reflect how such liabilities would be treated by a third party lender. For example, they said that third party lenders would not take deferred tax liabilities (DTLs) or remediation provision made by a mining company into account in determining a borrower’s creditworthiness, and therefore those items should be treated in the same way as shareholders’ funds in determining a New Zealand group’s maximum level of deductible debt under the thin capitalisation rules.

PwC commented that the thin capitalisation rules are not trying to evaluate the true equity investment by a multinational group into its New Zealand business.

Comment

Officials agree that one of the objectives of the thin capitalisation rules is to limit a company to interest deductions on a commercial level of debt, and this was recognised at paragraph 4.7 of the March 2017 Discussion Document BEPS – Strengthening our interest limitation rules. However, experience around the world has shown that determining a commercial level of debt for a part of a multinational enterprise is very difficult. The approach of the thin capitalisation rules in New Zealand has always been to avoid doing this. Instead we provide a reasonable safe harbour, and in addition allow a group if it wishes to do so to take the metrics of the worldwide group as an objective measure on which to judge the New Zealand part of the group. So, we allow a New Zealand group to deduct the interest on up to 110% of its worldwide group debt percentage, with a reasonable safe harbour to reduce compliance costs. The 2017 Discussion Document recognised this approach when it listed, as another objective of the thin capitalisation regime, to ensure that only a reasonable portion of a multinational’s worldwide debt is allocated to New Zealand.

The proposed exclusion of non-debt liabilities from the denominator in the debt percentage calculation is consistent with this approach. The exclusion applies to both the New Zealand and the worldwide debt percentages. It uses an apportionment approach based on the debt and net assets of the worldwide group to determine the amount of deductible debt that should be allocated to New Zealand.

The benefit of the non-debt liability exclusion is that it ensures that the amount of deductible debt that should be allocated to New Zealand is not able to be inflated because non-debt liabilities are treated in the same way as equity. For example, the current rules allow a group whose New Zealand operation has relatively high non-debt liabilities to operate with a much higher debt/shareholder’s equity (i.e. net assets) ratio in New Zealand than it has on a worldwide basis.

An attempt to discriminate between liabilities which are and are not taken into account by creditors would be complex and difficult to administer. For example:

there are some non-debt liabilities, such as trade payables or most derivative obligations, that other creditors would often take into account in determining a company’s creditworthiness. However, not all creditors will treat these equally, eg they may be irrelevant to secured lenders;
treating non-debt liabilities as equity allows a company to increase the amount of deductible interest bearing debt it is allowed by increasing the amount of its non-debt liabilities. While there will frequently be commercial constraints on such activity, there are other cases where these constraints do not apply.

Officials also note that whether or not non-debt liabilities are taken into account by lenders, they do have at least one very important commercial consequence. A company must take them into account in determining its solvency for corporate law purposes. They are thus an important constraint in determining the amount of debt which a company can commercially borrow on a stand alone basis.

Recommendation

That the submission be declined.

Issue: Should deferred tax liabilities be included in non-debt liabilities

Submissions

(Chartered Accountants Australia New Zealand, Corporate Taxpayers Group, EY, KPMG, Powerco, PwC)

These submitters submitted that deferred tax liabilities (DTLs) should be excluded altogether from non-debt liabilities. Reasons included:

DTLs are often the tax-effected expectation of future profits (eg deferred tax on building revaluations). (Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, EY, Powerco) They are relevantly different from other liabilities such as trade payables where the business transaction giving rise to the liability has already occurred
DTLs may be volatile where an asset is valued at fair value, and the movements do not necessarily reflect future tax to pay (Chartered Accountants Australia and New Zealand, EY)
lenders look to after tax cash flows of a business to determine a client’s ability to service debts, and that is reflected in balance sheet debt levels. They do not look to adjust for deferred tax because that would be double counting the tax liability. The non-debt liability proposal should better reflect this perspective (Corporate Taxpayers Group, EY, PwC)
calculating DTLs is complex and expensive, and DTLs are typically calculated at most once or twice a year. This means that requiring their inclusion in non-debt liabilities will impose a significant compliance burden on companies choosing to determine their thin capitalisation position on a quarterly or daily basis. (Corporate Taxpayers Group)
It is not possible or else it is impractical and unnecessarily costly to split up a DTL into a portion which is included in non-debt liabilities and a portion that is not, in the way proposed by the Bill. (Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, Powerco, PwC)
DTLs are excluded from non-debt liabilities in the Australian rules. (Chartered Accountants Australia and New Zealand, KPMG, Powerco, PwC)
DTLs should not be taken into account because they do not fund a company’s assets.

Comment

In principle, officials do not consider it is appropriate to exclude DTLs from non-debt liabilities. For example, where an asset held on revenue account at a balance date has a higher financial statement value than its tax value, the future sale of that asset for its financial statement value will generate a tax liability. The financial statement value of the asset will be included in the measurement of equity under the thin capitalisation rules. It is therefore unarguable that this tax liability should also be taken into account in determining shareholders’ equity in the financial statements. That is an example of why DTLs exist. For thin capitalisation purposes, this liability should not be treated as if it were equity provided by the shareholder of the company, as it currently is. Accordingly, to exclude DTLs from non-debt liabilities altogether would not be inappropriate in principle.

It is not relevant that the transaction giving rise to the DTL (eg the earning of income, or sale of the relevant asset) has not occurred at the time the DTL is created. The DTL is necessary to give an appropriate picture of the financial position of the company. Financial statements routinely include asset valuations which can only be justified on the basis of assumptions about future cash flows. For example, when assets are fair valued, this is often on the basis of an estimate of future cash flows. It is not appropriate for those kinds of assumptions to be made when valuing gross assets for thin capitalisation purposes, and then ignored when valuing associated tax liabilities.

Volatility of DTLs in general terms should support the proposal in the Bill. Volatility in deferred tax is generally caused by volatility in the value of assets (or sometimes liabilities) where the tax effect is deferred until sale. The two movements are inversely correlated and including them both in the thin capitalisation calculation rather than only including one will therefore decrease volatility. For example, if the financial statement value of a revenue account asset increases by $100, this will increase the DTL by $28. The overall increase in the denominator for the thin capitalisation calculation will be $72 rather than $100. Including the DTL has reduced volatility. Officials note that the role of DTLs in reducing volatility was not acknowledged by submitters.

Officials note that a re-categorisation of an asset from held for use to held for sale, or vice versa, can cause a change in the DTL associated with that asset. This volatility does not seem to be important for this issue. The change in the company’s intentions with respect to the asset, if genuine, should be reflected in a change in the DTL, and that change in turn should be reflected in the amount of the company’s net equity. Officials also note that submitters did not quantify this effect, and it seems likely to be minor.

Officials accept that the calculation of DTLs is complex and technical. Officials do not consider that this is a basis for their exclusion from non-debt liabilities. The calculation must be undertaken for financial reporting purposes, and no changes to the amount so calculated are required by the Bill (though we do propose an option to exclude certain DTL amounts from non-debt liabilities).

The approach taken by actual or hypothetical lenders to a borrower’s DTLs is also not relevant.

The thin capitalisation rules are not primarily concerned with creditworthiness as assessed by third party lenders. They are concerned with the extent to which a New Zealand group is funded by its shareholders rather than by external creditors.
Lenders will often have priority over income tax liabilities, and may also consider that if a company is in trouble, its tax losses may well offset any tax liabilities. Their approach therefore may say little about the reality of DTLs in measuring shareholder equity.
The submission that lenders look to a company’s after tax cash flows to assess its creditworthiness leads logically to the conclusion that we should not be restricting interest deductions by reference to the balance sheet but instead by reference to earnings before interest, tax depreciation and amortisation (EBITDA). It does not support a submission that DTLs in particular should be omitted from non-debt liabilities in a balance sheet based test.

Submitters did not explain why, if apportionment is not possible, it would be better to exclude DTLs from non-debt liabilities, rather than to include them.

Officials agree that DTLs arising from building revaluations, and any other revaluations of assets where the valuation already recognizes the owner’s inability, or limited ability, to claim a deduction (including by way of depreciation) for the financial reporting value of the asset should be able to be excluded from non-debt liabilities. The reason for this exclusion is that in that case, the DTL should not be taken into account in determining the shareholder’s equity in the company. That equity has already been accurately measured through the valuation of the asset. For instance, in the case of a building which is acquired or revalued after buildings became non-depreciable, the value of the building already recognizes that it is not depreciable. This outcome is intended to be achieved by proposed section FE 16B(1)(e)(iii), discussed in more detail below.

Officials note that there is no intention to exclude from non-debt liabilities a provision for depreciation on a building which was acquired before buildings became non-depreciable, if the building has not been revalued in the financial statements since that time. In that case the value of the building in the financial statements reflects an assumption of depreciability, so there is insufficient reason to exclude the DTL that was created when the building became non-depreciable.

The submission that requiring the inclusion of DTLs in non-debt liabilities will impose additional cost on entities choosing to calculate their thin capitalisation position quarterly or daily seems correct, but such entities are already accepting the imposition of what is presumably a much larger additional cost, being the cost of preparing a complete balance sheet for each quarter or day when the company would not otherwise do so. As a practical matter, officials expect that in a business as usual situation, a company wanting to use quarterly valuation might be able to make some simplifying assumptions to reduce the cost of calculating its DTL.

It is true that Australia does not include DTLs in the definition of a non-debt liability. When Australia introduced its thin capitalisation rules, DTLs were included in non-debt liabilities. They were only removed in 2008, in response to the change to the current accounting standards, which officials understand are materially the same as those applying in New Zealand. The published reasons for the removal are the same as those given by submitters in relation to our proposal. As set out above, officials consider those reasons do not warrant a carve out for deferred tax liabilities.

Recommendation

That the submissions in favour of excluding DTLs altogether from non-debt liabilities be declined.

Issue: The exclusion from non-debt liabilities for certain deferred tax liabilities should be clarified

Submissions

(Chartered Accountants Australia New Zealand, KPMG, Powerco, EY)

Some submitters (Chartered Accountants Australia New Zealand, KPMG) submitted that the exclusion from non-debt liabilities for certain deferred tax liabilities should be clarified.

Other submitters (eg Powerco) supported the intention of the exclusion from DTLs for certain amounts, as contained in proposed section FE 16B(1)(e), but were concerned that it was not workable.

Comment

Some submitters (Chartered Accountants Australia New Zealand, KPMG) submitted that proposed subsection (e) of the exclusion which is provided for certain DTLs is unclear.

KPMG submitted that as the provision is drafted, it would not necessarily allow a DTL arising in connection with a building to be excluded. That is because the deferred tax liability in that case might arise due to the fact that the depreciation rate for buildings is 0%, rather than because of a difference between the financial statement value and the adjusted tax value.
Chartered Accountants Australia New Zealand, along with a number of other submitters, were concerned that paragraph (iii) of the exclusion might only apply if the asset is valued in the financial statements at the value it has for tax purposes. This is not the purpose of the provision. The paragraph is intended to limit the exclusion to the situation where the difference between the tax and financial reporting value of an asset has already been taken into account in the valuation or cost exercise.

With respect to the submission that paragraph (iii) of the proposed exclusion from DTLs would be difficult to apply, officials consider that the concern may be overstated in many cases. Once taxpayers adjust to its existence, they will find it relatively easy to undertake the exclusion. Officials accept that it might be desirable for the legislation to be amended to make it explicitly optional for taxpayers to exclude any amount from DTLs. (EY)

With respect to the submission that a DTL on a building might not be able to be excluded under paragraph (e), officials agree that a DTL arising because an asset is depreciable at zero percent should be able to come within paragraph (e), and that a clarification is appropriate.

Contrary to PwC’s submission, officials do not consider that proposed section FE 16B(1)(e)(ii) assumes that a gain on sale of an asset will remain consistently taxable or non-taxable. If the tax status changes, that may result in any DTL associated with that asset being in or out of proposed section FE 16B(1)(e)(ii). That does not seem problematic.

Recommendation

Officials recommend that the exclusion from non-debt liabilities for certain deferred tax liabilities proposed in the Bill should be made explicitly optional, and that this exclusion should clearly be able to apply to buildings which are depreciable at 0%.

Issue: Should derivative positions be included in non-debt liabilities

Submissions

(Chartered Accountants Australia New Zealand, Corporate Taxpayers Group, Powerco, Russell McVeagh)

Some submitters submitted that derivative liabilities (eg an “out of the money” forward contract to exchange US$ for NZ$) should not be included in the calculation of total assets or non-debt liabilities. (Chartered Accountants Australia and New Zealand)

Such liabilities should be excluded to the extent that they have not been used to fund the taxpayer’s balance sheet. (Corporate Taxpayers Group, Powerco)

Taxpayers should be able to elect to exclude certain classes of derivatives from the calculation of assets and non-debt liabilities. The submitter gave the example of a business with an interest rate swap which becomes out of the money. The interest rate swap hedges a floating rate loan (which officials understand cannot be fair valued). The non-debt liability proposal means that the derivative going out of the money would increase the business’s thin capitalisation ratio under the proposed exclusion for non-debt liabilities. (Russell McVeagh)

Comment

Under current law, a derivative position which is recorded in the financial statements as an asset is treated for thin capitalisation purposes as an asset, whereas a derivative position which is recorded in the financial statements as a liability is treated in the same way as shareholders’ funds. It is not treated as debt because it does not bear interest. This asymmetry is not appropriate, and there were no submissions in favour of its retention.

Where a derivative is a hedge of an item which is fair valued for purposes of the thin capitalisation calculations, then it is appropriate to include the derivative in assets (if it is in the money) or non-debt liabilities (if it is not). This treatment will be symmetric for the derivative itself, and will match the thin capitalisation treatment of the hedged item. This will reduce volatility in the thin capitalisation calculation.

Where the derivative is a hedge of an item which is not fair valued for thin capitalisation calculation purposes, or which is not included in the thin capitalisation calculations at all, officials consider there may be an argument for excluding the derivative from the measurement of assets and non-debt liabilities. There may also be cases when it should be taken into account. Take the case given by Russell McVeagh and described above. If the swap becomes out of the money, that does seem to be appropriately reflected in an increased debt percentage, since the value of the shareholder’s funds has been diminished, which will be recognised by way of a credit to the cash flow hedge reserve. Officials consider that the information provided in submissions does not provide a sufficient basis for excluding derivatives from non-debt liabilities.

Recommendation

That the submissions supporting excluding all or some derivatives from assets and/or non-debt liabilities be declined.

Issue: Liabilities of retirement village operators under occupational licences should be grandparented or excluded from non-debt liabilities

Submissions

(Corporate Taxpayers Group, PwC)

The non-debt liability proposal will have a significant impact on the retirement village industry, because of the substantial non-debt liabilities that are on their balance sheets from occupational right agreement. Either:

a grandparenting provision should be provided for agreements entered into before enactment of the Bill (Corporate Taxpayers Group); or
such provisions should not be treated as non-debt liabilities (PwC).

Comment

The Bill does not generally provide any grandparenting against the effect of the proposals relating to non-debt liabilities. The fact that the change will have a significant impact is not of itself a sufficient reason for grandparenting.

Recommendation

That the submission that retirement village operators be entitled to some form of grandparenting from the effect of the non-debt liability proposal in relation to liabilities from occupational right agreements be declined.

Issue: Should there be industry specific carve outs for certain non-debt liabilities

Submissions

(PwC )

A blanket approach to non-debt liabilities will severely adversely affect certain industries, including the securitisation, securities lending, retirement village and forestry industries.

Comment

The extent of such severe adverse effects has not been quantified by submitters during the policy formation or submissions process, and so is difficult to determine. Officials note that taxpayers are able to deduct interest on debt which does not exceed 110% of their worldwide debt percentage. Accordingly, if there are severe adverse effects, this would suggest that multinationals are putting much more debt, proportionately, into their New Zealand businesses than their non-New Zealand businesses. That is the practice which the thin capitalisation rules are trying to address.

Recommendation

That the submission be declined.

Issue: Related party trade payables should be excluded from non-debt liabilities

Submissions

(Russell McVeagh)

Trade payables owing to shareholders or members of a shareholder group should be excluded from non-debt liabilities. This is on the same basis as the exclusion for other forms of non-interest bearing financing provided by shareholders, eg an interest free loan.

Comment

Related party trade payables are not the same as an interest free loan from a shareholder. If a foreign company sells goods to a New Zealand group member on deferred payment terms, the price charged by the related party will be higher than the price that would be charged for a cash sale. The New Zealand purchaser’s cost of goods deduction is in a sense a composite of a deduction for pure cost of goods and an interest charge. If the related party arrangement were changed so that the goods were sold for cash, and the purchase price obligation were funded by a loan, in order for the arrangement to produce the same tax result as the pre-change arrangement, the loan would have to bear interest. It would then be a financial arrangement, and included in the denominator of the debt percentage calculation.

The proposed thin capitalisation treatment (treating the related party trade payable as a non-debt liability) is more favourable to taxpayers than alternate treatment of it as a financial arrangement providing funds. There was no consultation on that alternative, and accordingly we do not recommend it here. It does demonstrate though why this submission should be declined.

Recommendation

That the submission be declined.

Issue: Inconsistent measurement approaches in determining non-debt liabilities

Submissions

(Corporate Taxpayers Group, Powerco )

Corporate Taxpayers Group and Powerco submitted that there is a measurement inconsistency in the drafting of proposed section FE 16B. Whereas “total liabilities” in the opening words of proposed subsection (1) are measured by reference to the outstanding balances shown in the financial statements, the reductions for liabilities which should not be treated as non-debt liabilities do not have a clear measurement test. It could, for example, refer to the outstanding balance for income tax purposes. This would not be desirable.

Powerco also submitted that there is a measurement inconsistency when a taxpayer uses spot exchange rates to calculate income or expenditure from hedged foreign currency financial arrangements.

Comment

Officials agree that consistency in this regard is important.

Recommendation

That the submissions in favour of consistency be accepted, and that the subtractions listed in section FE 16B(1) be clearly measured by reference to the outstanding balance of the items as shown in the financial statements. Similar amendment may also be required to section FE 16B(2).

Issue: Other exclusions from non-debt liabilities

Submission

(Chartered Accountants Australia New Zealand, PwC)

Chartered Accountants Australia New Zealand supported the exclusion from the definition of non-debt liabilities of interest free loans, preference shares and provisions for dividends. They also submitted that the exclusion from the definition of non-debt liabilities for preference shares should be expanded so it applies to preference shares that are not pro rata with shareholding. They pointed out that determining whether preference shares are a liability or not for financial statement purposes can be complex.

PwC supported the exclusion from non-debt liabilities for certain shareholder funding arrangements.

Comment

An amount will only be a non-debt liability if it is a liability in the financial statements. Generally preference shares will be a liability in the financial statements if they are redeemable in cash at the option of the holder or at a specified time. In that case, they are in a commercial sense a liability, and there is prima facie no reason not to treat them as such under the non-debt liability proposal. Non-debt liability treatment should only be reversed where there are other funding arrangements between the company and the shareholder that support treating the preference shares as equity. The nature of these arrangements should be the same as if the preference shares were an interest free loan. That is the effect of the Bill as introduced.

The complexity of determining whether preference shares are a liability or not for financial statement purposes does not seem relevant to the issue. The cost of making this determination has to be incurred for financial statement purposes, so there is no additional cost involved in using that determination under the thin capitalisation rules. At the margin, any complexity may mean that there are differences of opinion between advisors or over time. These kinds of issues are difficult to avoid, and not a strong argument in favour of abandoning a distinction which should generally be predictable enough.

Recommendation

That the submission in support of these exclusions be noted. That the submission in favour of excluding from non-debt liabilities preference shares which are not pro rata with shareholding be declined.

INFRASTRUCTURE PROJECT FINANCE

(Clauses 5, 17, 20, and 43(18))

Issue: Support for concessionary rule for infrastructure projects

Submission

(Corporate Taxpayers Group, Chartered Accountants Australia and New Zealand, Chapman Tripp, PwC)

Most submitters supported a concessionary rule that allows certain infrastructure projects that produce a public benefit to have more third party, limited recourse debt than would be allowed under the ordinary thin capitalisation rules.

Investment in infrastructure is essential for New Zealand’s growth. (Chartered Accountants Australia and New Zealand)

Current tax rules unduly penalise participants in government infrastructure projects, which can result in increased costs for government and public sector entities. The infrastructure project finance exemption will help prevent project participants with a commercial level of debt from being penalised. This will help minimise infrastructure costs for the Government. (Corporate Taxpayers Group)

Comment

Officials note the support of the majority of submitters for a concessionary thin capitalisation rule for certain infrastructure projects.

Recommendation

That the submission be noted.

Issue: Opposition to concessionary rule for infrastructure projects

Submission

(New Zealand Council of Trade Unions)

One submitter said that the proposed concessionary rule for certain infrastructure projects is an unnecessary subsidy for those projects. This submitter said that the infrastructure projects that would receive the benefit of this rule are an expensive way to fund public infrastructure because private debt is more expensive than government debt.

This submitter considers that any competitiveness issue in the bidding process that results from the application of the thin capitalisation rules can be addressed by specifying maximum levels of debt in all relevant infrastructure projects.

This submitter also points to the failings of these infrastructure projects in New Zealand and overseas as a reason for why they should not be subject to a concessionary thin capitalisation rule.

Comment

Officials note that the Crown weighs up a number of factors when deciding how to fund an asset, but whether a specific asset could be funded in a more efficient way should not influence whether or not a concessionary rule should be provided.

Under current tax settings, infrastructure project participants using some structures are able to receive a deduction for all of their third party debt, while investors using other structures are limited in the amount of debt they can have by the thin capitalisation rules. This means that potential investors in an eligible infrastructure project are not always able to bid on a level playing field.

Officials consider that the proposed exemption for public project debt in eligible infrastructure projects is the best way to deal with this competitiveness issue. The alternative of limiting all eligible infrastructure projects to a lower proportion of limited recourse third party debt would force project participants to take on less debt funding than would be commercially efficient, which would in turn lead to an increase in the cost of the infrastructure project.

Recommendation

That the submission be declined.

Issue: Scope of rule – Central Government

Submission

(Chapman Tripp, Corporate Taxpayers Group, PwC)

The rule as drafted would apply only to infrastructure projects entered into with the Crown or with a public authority. The rule should be amended so as to apply to a wider range of infrastructure projects, including those entered into with local authorities, council controlled organisations, schedule 4A companies, and social housing providers.

Comment

Officials consider that the current definition of public project assets is sufficient. If, in the future, it becomes clear that some infrastructure projects are established that should receive the benefit of this rule but are not covered by the definition, the definition can be reviewed.

Recommendation

That the submission be declined.

Issue: Scope of rule – Determination power

Submission

(Corporate Taxpayers Group, PwC)

If the rule is not expanded so as to include infrastructure project contracts concluded with local authorities, council controlled organisations and social housing providers, Inland Revenue should have a determination power that would give them the flexibility to allow these infrastructure projects to still receive the benefit of this rule, provided that the infrastructure project was in the public interest. (Corporate Taxpayers Group, PwC)

Alternatively, such a discretionary power could instead be held by the Treasury. (Corporate Taxpayers Group)

Comment

Officials do not consider that a determination power is necessary. Officials consider that the scope of this concessionary rule should be fixed by Parliament. If changes to the scope of this rule are necessary, those changes should be required to go through the ordinary parliamentary process.

Recommendation

That the submission be declined.

Issue: Scope of rule – management phase

Submission

(Chapman Tripp, Corporate Taxpayers Group)

The rule as drafted requires the persons performing the contract to both construct or upgrade the public project assets, and operate or maintain the assets in New Zealand. Some infrastructure projects entered into by the Government comprise only a construction phase. These infrastructure projects should also get the benefit of this rule. (Chapman Tripp, Corporate Taxpayers Group)

Consideration should also be given to the fact that Government may be shifting to a delivery model for infrastructure projects that only includes a construction phase, meaning that future projects are unlikely to have a management phase. (Chapman Tripp)

If these submissions are not accepted, the legislation should be amended so that contracts that require the persons performing the contract to manage the assets in New Zealand are eligible, in addition to contracts that require the assets to be operated or maintained. (Corporate Taxpayers Group)

Comment

The rule is only intended to apply to infrastructure projects of at least 10 years in length. Any infrastructure project that is only a contract to construct an asset is not intended to receive the benefit of this rule. Officials consider that short term projects that only require the construction of a public asset do not require a concessionary thin capitalisation rule.

Recommendation

That the submission be declined

Issue: Public project debt - security requirement

Submission

(Corporate Taxpayers Group)

The current definition of public project debt requires the debt to be secured against either public project assets, or income derived from public project assets. The definition of “public project debt” should not include any reference to the security provided for the debt. Instead, any debt that has a nexus with an eligible infrastructure project should be included in the special thin capitalisation calculation.

If this is not accepted, Inland Revenue needs to clarify how a person involved in an eligible infrastructure project needs to treat debt that is not public project debt for the purpose of the thin capitalisation rules.

Comment

The submitter is correct that the definition of public project debt is intended to be a gateway that captures all debt related to the eligible infrastructure project. Officials considered that all debt that was connected with an eligible infrastructure project would have some recourse against the project.

Based on the submission that it is possible that debt that will be applied to an eligible infrastructure project can, in certain circumstances, not have any recourse against the project, officials consider that the rule should be amended so that the requirement that public project debt must have recourse that is related to the project is removed.

Widening the definition of what constitutes public project debt will ensure that the rule works as intended. All debt that is intended to receive the benefit of this rule would have been included in the definition of public project debt as originally drafted. However some debt that is used as a part of an eligible project, but is not intended to receive the benefit of the rule, would not have come within this definition. Widening the scope will allow this debt to be appropriately dealt with by the special infrastructure rule, as opposed to the ordinary thin capitalisation rules.

Recommendation

That the submission be accepted.

Issue: Limited recourse debt

Submission

(Corporate Taxpayers Group)

All public project debt that is not recourse debt or public project participant debt should be limited recourse debt.

The current definition of limited recourse debt only includes debt that has security against either interests in public project assets, or income derived from these public project assets. There are two problems with this:

Public project assets are generally owned by the Crown, so the taxpayer applying the rule will not have an interest in the assets.
Debt should be allowed that has security against income derived from the project, not the public project assets.

The concept of limited recourse debt should be more widely defined so that it only excludes public project debt that has recourse that does not relate to the infrastructure project. The current definition of what constitutes “limited recourse” is too narrow and would exclude debt that in effect only has recourse against assets and income related to the project.

The submitter has provided some suggested drafting that would put this into effect.

rec[course] debt is the amount of all public project debt that is not public project participant debt and for which any security provided to the creditor does not relate only to the project:

limit[ed recourse] debt is the amount of all public project debt that is not public project participant debt and is not rec debt:

Alternatively, the definition of “limited recourse debt” should be amended so as to allow for all forms of security that are commonly provided in relation to eligible infrastructure projects. The common security interests that are not covered by the rule as drafted include assets and equity interests in a project Special Purpose Vehicle (SPV), where the activities of the SPV only relate to the project.

In addition, the reference to income derived from a public project asset in the definition of “limited recourse debt” should instead refer to income derived from the project.

Comment

Officials agree with the submitter’s point and consider that their alternative definition of recourse debt has merit. Any debt that has a nexus with an eligible infrastructure project, but has security that is not related to that project, should not receive the benefit of the proposed concessionary rule. If debt that had recourse against assets that do not relate to the project was allowed to get the benefit of this rule, then it is possible that the project participants could over-allocate debt into the New Zealand-based infrastructure project.

Officials therefore consider that the definition of recourse debt be amended in a manner similar to what is proposed by the submitter

If a definition similar to what is proposed by the submitter is adopted, there will no longer be a specific reference to income derived from a public project asset in the definition of recourse and limited recourse debt.

Recommendation

That the submission be accepted subject to officials’ comments.

Issue: Excess debt

Submission

(Corporate Taxpayers Group)

The restriction on how much debt can receive the benefit of the proposed exemption should be removed so that all debt related to an infrastructure project covered by this rule would be deductible, unless it is either on non-limited recourse terms, or is public project participant debt.

In addition, a safe harbour should be introduced that would allow a taxpayer to avoid the apportionment formula in the rule, provided that the project is funded solely by limited recourse debt that is not public project participant debt. The submitter also recommends granting the Commissioner of Inland Revenue a determination-making power to approve certain project debt facilities so that no deductions are denied.

This is because the value of public project assets as defined does not include all relevant assets. Instead, the effect of the rule is to effectively compare a legal amount of debt with an accounting value of a financial asset that is an estimate of the present value of the cash flows that are expected to be derived from the project. This accounting valuation of the financial asset associated with the project can underrepresent the real value of the assets associated with the infrastructure project.

If the submission to remove excess debt from the rule is not accepted, then the definition of what constitutes excess debt should be amended to mean one of:

Debt that exceeds the aggregate estimated construction costs of the public project assets; or
Debt that exceeds 100% of the total assets held by the contractor at the relevant time for the purpose of the project.

Comment

Officials accept the submitter’s submission that all debt that is on limited recourse terms and is not public project participant debt should be deductible. As such, officials consider that a limit on how much debt can be deductible under this proposed rule should be removed.

If excess debt is removed from the apportionment formula, officials consider that a safe harbour that would allow a taxpayer to avoid the apportionment formula if the only type of debt they have is limited recourse debt that is not project participant debt is unnecessary. If this is the only type of debt that a taxpayer has, then the apportionment formula will result in zero interest deductions being denied.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Public project participant debt (FE 4B (3))

Submission

(Chapman Tripp, Corporate Taxpayers Group)

The definition of public project participant debt is unnecessarily complex and should be simplified so that it can be more easily understood and applied by taxpayers. In addition, a detailed example should be provided in the commentary that clearly illustrates when an investor will be considered to have lent money in its capacity as a third party lender. (Corporate Taxpayers Group)

Further clarity is required around what form of related party debt will be deductible, as the rule as currently drafted is uncertain in its application. At a minimum, non-proportionate shareholder debt should be deductible. (Chapman Tripp)

Comment

Officials agree that complexity is to be avoided and aim to achieve this. The definition was drafted with the intention of ensuring that taxpayers could not abuse the rule and treat debt that is effectively a substitute for equity as if it were third party, limited recourse for the purpose of the thin capitalisation rules. Officials believe that the definition of public project participant debt can be redrafted in a way so as to be easier to understand and still accomplishes this purpose.

Officials consider that, in general, non-proportionate shareholder debt should be deductible under this rule. However, there may be some specific situations where some project participants are able to provide debt that is not in proportion to their interest in the project, but that is still effectively a substitute for equity. Officials consider that such debt should still some within the definition of “public project participant debt”. A clear example will be provided in the Tax Information Bulletin (TIB) in order to illustrate what sort of participant debt will be considered a substitute for equity.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Debt applied to assets used in the project

Submission

(Corporate Taxpayers Group)

The definition of “public project debt” in s FE(4)(2)(a) requires that the debt is applied to the project to give rise to public project assets. Officials should confirm that this definition does not prevent debt from being applied to other assets required to carry out the project.

In addition, the definition of “public project debt” should be amended so that it only looks at the purpose for which the debt is drawn down for. All debt that is drawn down for the purpose of the project should be considered “public project debt”. This would prevent the need for a prescriptive definition that appropriately captures all relevant debt.

Comment

Officials consider that the definition of “public project debt” already allows for debt that is applied to assets that are used to carry out an infrastructure project. The debt must simply be used in a way that gives rise to public project assets, even if that way is in an indirect fashion.

Officials consider that the test for whether debt is “public project debt” should not be based solely on the purpose for which the debt is drawn down for. The various requirements in the definition of “public project debt” are necessary in order to ensure the integrity of the proposed rule.

The Specialist Advisor to the Finance and Expenditure Select Committee (Specialist Advisor) has suggested that the definition of “public project debt” be amended to allow for debt that is applied to the project so as to give rise to income derived from the project. Officials agree with this suggestion.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Debt refinancing

Submission

(Corporate Taxpayers Group)

It is possible that the definition of public project debt does not cover the refinancing of debt that was originally valid public project debt. The submitter states that it is common for senior bank debt facility refinancing to occur every 5 to 7 years for relevant infrastructure projects and that this debt should be included in the definition of “public project debt”.

Comment

Officials agree that the definition of “public project debt” should be amended so that debt used to refinance existing “public project debt” is also included in the definition.

Recommendation

That the submission be accepted.

Issue: On-lending funds to persons not associated with the project

Submission

(Corporate Taxpayers Group)

The restriction on “public project debt” not being lent to a person not associated with the project is too restrictive. It is possible for third party funds to be drawn down on pre-set dates, but the timing of the construction does not necessitate the payment of those funds immediately.

As a result, there will be a legitimate business reason for funds that will eventually be applied to a relevant project to be on-lent to a third party, such as a financial institution, before the funds are eventually applied directly to the project.

Comment

Officials consider that such on-lending of funds that will eventually be applied to the infrastructure project should not invalidate that debt from being public project debt.

Officials consider that the provision should be amended so as to explicitly carve out on lent funds that are deposited with a registered bank, except for funds that are not intended to be used in performing the project.

Recommendation

That the submission be accepted.

Issue: Rule should be optional

Submission

(Corporate Taxpayers Group)

As the rule is intended to be concessionary to taxpayers investing in eligible infrastructure projects, these taxpayers should have the option of applying the ordinary thin capitalisation rules if they want. The rule as currently drafted could result in a greater level of interest deductions denied than under the ordinary thin capitalisation rules in certain circumstances.

Comment

Officials consider that the exemption should never result in a worse outcome for eligible taxpayers when compared to the ordinary thin capitalisation rules. However, officials consider that providing an election for an infrastructure project as to whether the proposed concessionary rule or the ordinary thin capitalisation rules will apply is acceptable, provided that the election is a one-off decision for each excess debt entity that can only be taken before the first thin capitalisation calculation is undertaken for the infrastructure project by that entity.

For existing infrastructure projects, the election will need to be exercised at the time of the first thin capitalisation calculation after the rule comes into effect.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Allow investor to divest interest to a separate investor

Submission

(Corporate Taxpayers Group)

The rule as drafted prevents the investor from disposing of the public project assets within 10 years from the beginning of the contract, except to the public authority, or to another person performing the contract.

It should be clarified in a TIB example that the legislation allows one investor to sell their interest in an infrastructure project to a new investor.

Comment

Officials consider that the Bill already allows for an investor to dispose of their interest in the project to another investor. The meaning of “another person performing the project” includes new investors who will be performing the project in the place of the participant selling their interest in the project.

However, officials do consider that the wording in the draft legislation should be amended to clarify that public project assets can also be disposed of to the Crown, in addition to a public authority.

Recommendation

That the submission be declined.

Issue: Taxpayers within scope (FE 7B(1)(a))

Submission

(PwC)

The rule as drafted would only apply to a company controlled by a group of non-residents acting together. This restrictive requirement should be relaxed.

Comment

The rule as drafted would not apply to a New Zealand company controlled by a non-resident owning body. Instead, the rule will only apply to excess debt entities that are not controlled by a non-resident owning body.

All persons who will foreseeably be involved in eligible infrastructure projects are covered by the rule. The rule as drafted would only apply to entities that are controlled by a single non-resident, partnerships entered into by non-residents, and New Zealand resident entities subject to the outbound thin capitalisation rules.

Recommendation

That the submission be declined.

Issue: Exclude interest expenditure on public project debt from ordinary thin capitalisation rules (FE 7B(6))

Submission

(Corporate Taxpayers Group)

The rule as drafted excludes public project debt and public project assets from being taken into account in a separate thin capitalisation calculation. This exclusion needs to be extended to the interest expenditure on that debt as well.

Comment

Officials consider that the carve out of public project debt from use in other thin capitalisation calculations means that any interest related to that debt is similarly carved out.

For clarity, officials consider that the legislation should be amended to specifically carve out interest on public project debt from other thin capitalisation calculations.

Recommendation

That the submission be accepted.

Issue: Drafting issues

Submission

(Corporate Taxpayers Group)

The “Meaning of public project participation debt” title in section FE 4B should instead read “Meaning of public project participant debt”.

“Is not a source of funds” in section FE 4B(2)(c) should be “provides funds” in order to be consistent with the other thin cap legislation.

The section FE 4B (2) definition of “public project debt” refers to an excess debt entity’s proportion of a loan. This is unnecessary as section HG 2 already deals with a partner in a partnerships proportion of a loan. Using the wording contained in FE 4B(2) of “the excess debt entity’s proportion of the loan” instead of relying on section HG 2 is inconsistent with the rest of the thin capitalisation legislation.

Officials note that s FE 7B(2)(b) should also be amended so as to remove the reference to an excess debt entity’s proportional interest in the public project assets, so as to be more consistent with section HG 2 and the rest of the thin capitalisation rules.

Comment

Officials agree that all of these drafting issues should be addressed by amending the legislation.

Recommendation

That the submission be accepted, subject to officials’ comments.

Issue: Threshold debt amount

Submission

(Matter raised by officials)

The “threshold debt amount” of an excess debt entity, which is defined in s FE 7B(4)(c)(i) should be altered so as to be consistent with the definition of the entity’s debt percentage, as provided by FE 7B(2). This will allow assets used in performing the project to also be used in determining how much debt would be denied under the special infrastructure project rule.

Recommendation

That the submission be accepted.

Issue: Non-debt liabilities

Submission

(Matter raised by officials)

All of the non-debt liabilities of an excess debt entity that relate to a public project should be included in the calculation of that excess debt entity’s debt percentage. In addition, all of those non-debt liabilities should also be excluded under s FE 7B(6) from being taken into account for the rest of the thin capitalisation rules.

Recommendation

That the submission be accepted.

Issue: Disposing of asset to the Crown

Submission

(Matter raised by officials)

Section FE 4B(1)(b) allows a public project asset to be disposed of within 10 years from the beginning of the project to a public authority. Officials consider this should be amended to also include “the Sovereign in right of New Zealand”.

Recommendation

That the submission be accepted.

OTHER THIN CAPITALISATION

Issue: Non-proportionate shareholder debt

Submission

(Chapman Tripp)

The Bill contains an amendment to reduce the 110% worldwide debt threshold to 100% for an entity controlled by a group of non-residents acting together. Under the proposed amendment, if such an entity exceeds the 60% safe harbour, any owner-linked debt will be non-deductible.

We do not agree with the proposed amendment in respect of non-proportionate shareholder debt. In this scenario the shareholder is effectively taking on the role of third party lender. Shareholder debt in this situation should be considered analogous to third party debt and should remain deductible (even above 60% gearing). Any arm’s length debt should remain deductible above the 60% safe harbour on the basis that these investors will not have the benefit of a worldwide group test to reflect an appropriate industry debt level.

Comment

This position was set when these rules were introduced as the same issue can arise with a 110% threshold, albeit it arises earlier with a 100% threshold. While holding debt in proportion to shareholding is an indication that this debt may be in substitution for equity this is not the only way to achieve this effect. Allowing a borrower to have over 60% debt and the ability to have debt from shareholders would increase the risk of a company having high levels of debt in substitute for equity.

Recommendation

That the submission be declined.

Issue: De minimis threshold for inbound thin capitalisation rules should include owner-linked debt

Submission

(Chapman Tripp, Corporate Taxpayers Group)

We support the extension of the de minimis threshold so that it applies to inbound entities. However, we submit that the de minimis threshold should include owner-linked debt as otherwise the de minimis threshold would be very limited in application. The de minimis threshold should instead apply to all cases where the inbound thin capitalisation rules apply. This would be consistent with Australia’s de minimis threshold which is a flat A$2 million regardless of whether any lending is with a related party. (Chapman Tripp)

Section FE 6(3)(ac) provides a de minimis for a person subject to the outbound thin capitalisation rules. The Bill proposes to extend this to persons subject to the inbound thin capitalisation rules but only if they have no “owner-linked debt”. The proposed related party debt restriction is likely to make the de minimis very limited in application. It would be appropriate to extend the de minimis to inbound investment without the owner-linked debt restriction on compliance cost saving grounds. In Australia a flat $2 million de minimis applies, regardless of whether any lending is related party debt. (Corporate Taxpayers Group)

Comment

The de minimis applies consistently with the existing outbound de minimis in that it fully applies up to $1 million then diminishes up to $2m million. The purpose of this is to remove the disincentive to have only a small amount of interest above the $1 million threshold and therefore have the de minimis not apply at all.

Unlike the $10 million de minimis in sections GC 16(1)(a) and GC 18(1)(a) which apply to the loan principal this de minimis applies to interest. At an interest rate of 5% the amount of the loan(s) may be up to $20 million. Officials consider at this level of related party lending it is appropriate for the thin capitalisation rules to apply as the compliance costs of applying the rules will become smaller relative to the potential tax effect of the rules applying.

This de minimis does not apply to the inbound rules when the group has owner-linked debt. The thin capitalisation rules consider the level of debt rather than the price of that debt. When a group has owner-linked debt this increases the risk that a group has included higher levels of debt for tax purposes compared with a group that has entirely third party debt. As the de minimis does not currently apply at all to the inbound rules the proposals in the Bill result in a more taxpayer favourable treatment than under the current legislation.

Recommendation

That the submission be declined.

Issue: Non-residents acting in concert: related–party debt

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group)

We support the proposed amendement, sections FE 5(1)(ab) and FE 6(3)(e)(iii), to amend the rules for entities controlled by a group of non-residents acting together. (CA ANZ)

The group is supportive of the proposal to amend the way the thin capitalisation rules apply to “non-resident owning bodies” so if such a firm exceeds the 60% safe harbour, any owner-linked debt in excess of 60% will be non-deductible. (Corporate Taxpayers Group)

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: Group acting in concert

Submission

(Corporate Taxpayers Group)

The group does not support the inclusion of new section FE 5(6) for the following reasons:

It is an inherently uncertain rule. What does it mean to “act in concert”? “Act in concert” is not a defined term. The concept of acting in concert was rejected when the thin capitalisation rules were amended in 2012. Without further guidance it is completely meaningless and drives unnecessary complexity into the rules.
Perhaps more importantly section FE 5(6) is not required. As noted above, the issue to be addressed only arises when the membership of the worldwide group is determined under section FE 31D. Section 31D only applies when the “ordinary” thin capitalisation rules do not apply. It is only entities controlled by a non-resident owning body (i.e. entities that fall into section FE 2(1)(cb) that use FE 31D to determine the make-up of their worldwide group, and then it is only those entities that should be subject to this proposals.

Comment

“Act in concert” is an existing concept used in the thin capitalisation rules in sections FE 1(1)(a)(iii), FE 2(1)(d)(iii) and FE 26(7)(b) which each relates to the use of trusts. This was explained in Tax Information Bulletin Volume 26, Number 7 August 2014 as:

As with companies, the thin capitalisation rules apply to trusts settled by entities acting in concert. This is important to ensure the rules cannot be easily circumvented through the use of trusts. However, the rules for determining when a group of entities appear to be acting together used for companies (described in the section non-resident owning body) cannot be used for trusts. Instead, the rules apply to a trust settled by a group “acting in concert”. This is because, for example, it is not sensible to refer to settlements made in proportion to debt extended to a trust because rights to income from a trust generally do not depend on the amount a person has settled on it.

The proposed changes to sections FE 5 and FE 6 separate taxpayers who are controlled by a group acting in concert from other taxpayers covered by the thin capitalisation rules so that a 100% worldwide debt percentage can be applied instead of the typical 110%.

The changes to sections FE 5 and FE 6 were not intended to change the scope of who is acting in concert for the purpose of the interest apportionment rules, merely to apply a 100% percentage to taxpayers who were already treating their New Zealand group as their worldwide group. The provision that achieves this is section FE 31D and officials recommend the reduction in debt percentage to 100% of the worldwide group should cross-reference to this provision to ensure that the scope is unchanged.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Grandparenting of non-residents acting in concert

Submission

(Chartered Accountants Australia and New Zealand, Corporate Taxpayers Group, Powerco)

Submitters support the five year grandparenting of existing arrangements under proposed section FZ 8 on the basis that they were agreed under the existing tax regime. This will result in a fairer and more equitable outcome for taxpayers.

However, some submitters considered the grandparenting provision should include the option to assess the grandparented debt percentage up until the effective date of the changes. This would allow the preservation of the positions being taken immediately before the law is actually passed by Parliament in its final form, which gives effect to the intention of the grandparenting.

By contrast, only allowing assessment of the debt percentage on or prior to the date of introduction to Parliament appears to assume that taxpayers are likely to change their debt levels following introduction of the Bill. Given that the Bill is still subject to further changes as part of the Parliamentary process, the Group does not consider that this is a realistic assumption. Further, the existing and newly proposed specific anti-avoidance rules in section FE 11 and section GB 51B should deal with any avoidance concerns that Officials might have with regard to grandparenting positions taken between the introduction of the Bill and the date of enactment. (Corporate Taxpayers Group)

Without legislation taxpayers have not been in a position to reduce their debt to within the new reduced threshold. It is consistent with previous legislative reform in this area that taxpayers are given an opportunity to reset their gearing levels to fall within the new rules. The transition date should apply from the first measurement date after the corresponding Act is passed or the measurement date preceding the application date. (PowerCo)

Comment

The Bill includes, at the option of the borrower, two measurement dates for the grandparenting provision: either the introduction date of the Bill (6 December 2017) or the borrower’s thin cap calculation date immediately before this date. A measurement date for the date of enactment of the Bill was not included as this would create an incentive for any borrower covered by the proposed law change to increase their level of related party borrowing, up to the current 110% limit, prior to enactment of the Bill so they could maintain this higher level of debt for the next five years.

Officials consider the two possible measurement dates should accurately reflect the related party debt proportion of borrowers affected by the proposed law change.

Recommendation

That submitters’ support for the grandparenting be noted but that the submission to amend the transition date be declined.

Issue: Asset valuation

Submission

(Chartered Accountants Australia and New Zealand)

We support the proposal to allow taxpayers to use the net current valuation method when a valuation from an independent valuer has been received or the valuation methodology, assumptions and data have been approved by an independent valuer.

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: Assets not recognised for financial reporting purposes

Submission

(Corporate Taxpayers Group)

The asset measurement rules should be amended to allow assets that are not recognised for financial reporting purposes to be included in total assets for thin capitalisation purposes, where an independent valuation of that asset can and has been obtained. Specifically, we are advocating for the recognition of intangible assets that are not able to be recognised under existing financial reporting standards.

Business models and operators have changed significantly in the last few decades, such that traditional accounting measures of assets do not necessarily reflect the value of a company or provide an indication of likely cash flow to be generated. The reality is that a third party lender would look to more than just the balance sheet of a business when determining how much to lend, such as forecasted future income and “off balance sheet” assets like licences. As such, it would be appropriate to include the value of such assets for the purposes of the thin capitalisation rules.

If this is not accepted, then at a minimum interest paid on non-recourse third party debt supporting assets where a third party valuation is available to support should be allowed. What the thin capitalisation regime is seeking to do is ensure that the level of debt borne by the New Zealand operations is appropriate and not excessive. If a third party would be prepared to lend against an asset then such debt is not excessive as it is an arm’s length amount. We should not be denying interest on such funding depending on the accounting treatment.

Comment

There can be considerable subjectivity in the valuation of intangible property that is not able to be recognised under financial reporting standards, even when an expert valuer is involved. Allowing the inclusion of such assets when they are not recognised under existing financial reporting standards would create an incentive for taxpayers to (sometimes substantially) increase the value of intangible assets when they were approaching or over the thin capitalisation threshold (or the equivalent test in the restricted transfer pricing rules).

Recommendation

That the submission be declined.

Issue: Drafting of thin capitalisation rules

Submission

(Corporate Taxpayers Group)

Following numerous rounds of amendments to the regime, the thin capitalisation rules have now become extremely complicated and difficult to understand. Subpart FE of the Act should be renumbered and rewritten to allow the rules to be more easily understood.

Comment

The submitter’s suggestion would require significant resources that cannot be committed as part of the select committee process of this Bill. Officials will consider this submission for future inclusion on the tax policy work programme.

Recommendation

That the submission be declined.

Issue: Interest carry forward rule

Submission

(PwC)

An interest carry forward rule should be introduced to offset and balance the volatility and uncertainty that arises as a result of the proposed legislation.

Taking non-debt liabilities into account in calculating an entity’s capacity for debt will introduce potentially material volatility and uncertainty to that entity’s thin capitalisation calculations. This uncertainty will be compounded due to the currently drafted thin capitalisation anti-avoidance legislation.

This volatility is a similar concern to that recognised in the Bill Commentary as a problem with an EBITDA-based test and that taxpayers should be afforded protection from such unpredictability. Given the volatility is similar to that under an EBITDA-based test, protection similar to that suggested by the OECD (that is, the ability to carry forward denied interest deductions) should be adopted. Such a mechanism is intended to smooth out the effect of such volatility over subsequent income years.

Comment

An EBITDA test compares earnings to interest. This test will be volatile, particularly for taxpayers near the upper limit, as very few businesses will be able to accurately forecast their earnings a year in advance.

These concerns are not as significant with New Zealand’s thin capitalisation test which compares debt to assets (proposed to be assets minus non-debt liabilities). While there will be some difficulty in being 100% accurate in forecasting assets and non-debt liabilities this should be much easier than forecasting earnings. To the extent there is difficulty in forecasting debt and assets, this has always been a feature of New Zealand’s thin capitalisation rules.

The changes to the thin capitalisation rules will not necessarily increase volatility as in some instances a non-debt liability will be matched by a corresponding asset so a change in one value will be matched by an equal change in the other so that excluding non-debt liabilities reduces volatility. The anti-avoidance rule will only apply where a taxpayer intentionally structures to avoid the application of the thin capitalisation rules and officials do not consider this can be treated as increasing volatility.

Recommendation

That the submission be declined.

Issue: Inbound de minimis

Submission

(Corporate Taxpayers Group)

In the proposed amended section FE 6(3)(ac)(i) the reference to “and not a party to a financial arrangement…” should not include the word “not”.

Comment

This amendment is intended to extend the existing outbound de minimis to persons subject to the inbound de minimis unless they have owner-linked debt. When section FE 6(3)(ac)(i) applies the de minimis does not; therefore the amendment to this subparagraph should remove inbound persons without owner-linked debt from its existing scope.

A person with owner-linked debt is a party to a financial arrangement that is removed under section FE 18(3B) therefore a person without owner linked debt would not be a party to such an arrangement which means they would satisfy that portion of the amended section and therefore be ineligible for the de minimis. Therefore, the submitter is correct that the amendment will not achieve the intended outcome.

However, current drafting, even with the word “not” removed, as the submitter suggests, would not achieve the intended policy outcome. The is because the amendments to section FE 6(3)(ac)(i) remove the reference to “not [being] an excess debt outbound company”. The consequence is the de minimis would be available to all excess debt entities except for natural persons or trustees described in section FE 2(1)(g with owner-linked debt.

Officials recommend section FE 6(3)(ac)(i) is amended so that the de minimis is available to any person unless they are subject to the inbound thin capitalisation rules and they have owner-linked debt.

Recommendation

That the submission be accepted, subject to officials' comments.

Issue: Definition of assets

Submission

(Corporate Taxpayers Group)

Asset, as used in proposed section FE 16(1BAA), is not a defined term. It could be made clearer if “asset” is only intended to assets which can be recognised for financial reporting purposes.

Comment

While “asset” is not a defined term for the purpose of the thin capitalisation rules it is already used in existing section FE 16(1) which proposed section FE 16(1BAA) provides how that asset may be valued. Furthermore, existing section FE 16(2) states that total group assets must be calculated under generally accepted accounting practice. This amendment is not intended to change the definition of “asset” that already applies to the existing section.

Recommendation

That the submission be declined.

Issue: Frequency of asset valuation

Submission

(Corporate Taxpayers Group)

It should be clarified that taxpayers do not need to have assets valued annually to apply the net current value in proposed section FE 16(1BAA). For example, a new valuation could be required where there is an impairment event for financial reporting purposes, or the taxpayer is seeking to increase the value from a previous valuation.

Comment

This section provides how a valuation is to be determined but does not provide any guidance on how frequently that valuation must be undertaken. Officials consider it would be impractical to require a valuation each time a thin capitalisation calculation is undertaken. Conducting a valuation each time there is an impairment event for financial reporting purposes or if the taxpayer is seeking to increase the value from a previous valuation would be consistent with the policy intent and officials consider this can be achieved under the current drafting in the Bill.

Recommendation

That the submission be noted.

Issue: Valuation of unique assets

Submission

(Corporate Taxpayers Group)

Guidance on section FE 16(1BAA) is required in the situation where a taxpayer has a very unique asset. It may be difficult to find an expert in the valuation of a unique asset (but it can be valued by a valuation expert)

Comment

This section requires the valuation to be done by an independent person who is an expert in the valuation of such assets. In the case of a particularly unique asset an experienced valuer may be best placed to conduct that valuation even though they have not valued an identical asset before. This would be available under the proposed requirement. If a valuer was an expert in their field but had no experience of such assets, for example getting a valuation on intellectual property from a person who specialised in valuing real estate this would not meet the proposed requirement.

Recommendation

That the submission be noted.

Issue: Independent expert valuations of assets

Submission

(Corporate Taxpayers Group)

The concept that a valuation must be done by “an independent person who is an expert” needs to be refined or otherwise clarified in guidance. “Independent” and “expert” are too vague.

Who is “independent” (this is only targeted at excluding employees / contractors of the taxpayer). A person can be either a business or an individual. A professional services firm may provide multiple types of services / have multiple supplier/customer relationships with a taxpayer, it should still be viewed as independent from the taxpayer for the purposes of this rule. To be an “expert”, this should be the valuation service provider holding themselves out as an expert. There are not professional bodies or relevant qualifications for all types of assets.

Comment

It would be difficult to define what is “independent” or an “expert” as this will be very fact specific to an individual valuation. Independence suggests the valuer has applied an objective approach to the valuation. This would not be met if people relying on that valuation, such as the taxpayer or the Commissioner, believe that the valuer may have been influenced by other parties. Being an expert will also need to be determined on a case-by-case basis taking relevant factors such as qualifications, experience in valuing similar (but not necessarily identical) assets and membership of a relevant professional body if applicable.

Recommendation

That the submission be noted.

Issue: Support for anti-avoidance rule

Submission

(Chartered Accountants Australia and New Zealand)

We support the proposal to introduce section GB 51B, an anti-avoidance rule where a taxpayer enters into a transaction near a measurement date with the purpose or effect of manipulating the thin capitalisation rules.

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: Anti-avoidance rule scope

Submission

(Corporate Taxpayers Group, KPMG, PwC)

The application of the anti-avoidance rule in section GB 51B(1)(a) is too broad, especially due to the inclusion of “or would have”. What is this intended to capture? Would taking a legitimate financial reporting position satisfy this? (Corporate Taxpayers Group)

Further clarity is needed on the intended application of proposed section GB 51B. Given the enhanced scope for changes in loan values to be disregarded for thin capitalisation purposes, clarity is needed on the circumstances in which the Commissioner will seek to apply new section GB 51B. (KPMG)

We appreciate the need for anti-avoidance rules to be appropriately general in nature so as to capture tax outcomes which contradict policy rationale. However, the proposed rule as currently drafted is far reaching and could be erroneously applied to challenge tax outcomes that are supportable by ordinary commercial arrangements and not motivated at all by thin capitalisation implications, including but not limited to revolving credit, working capital and other short-term funding facilities.

The current anti-avoidance provision in the thin capitalisation regime should be sufficient.

If the proposed provision is maintained, at a minimum, it should include a safe harbour time frame beyond which an increase or decrease in a value would not be considered to have the purpose or effect of defeating the intent or application of the thin capitalisation rules. Allowance should also be made for immaterial movements in the values.

The proposed provision, if enacted, will need to be accompanied by significant Inland Revenue guidance on its intended application. The proposed provision should not capture an increase or decrease in value that is attributable, co-incidentally or otherwise, to a seasonal factor, one-off transaction or standard end-of-year procedures. (PwC)

Comment

The proposed anti-avoidance rule will operate very similarly to a number of existing anti-avoidance rules in other parts of the Income Tax Act. Like other such rules, whether this anti-avoidance rule applies will need to be considered on a case-by-case basis. The types of transactions identified by the submitters such as revolving credit, working capital and other short-term facilities will not trigger the anti-avoidance rule merely because they occur near a balance date. What is crucial is whether the transaction was for the purpose or effect of avoiding the thin capitalisation rules.

Where particular safe harbours are inserted into an anti-avoidance rule this risks sanctioning an acceptable level of avoidance. For example if the rule stated that it did not apply to any transaction more than a month from a balance date, this would mean it would not apply if a taxpayer deliberately had less than the required level of equity for the measurement period but repaid debt one month and one day prior to the balance date. While this would mean the rule applied to the most blatant transactions such as those entered into then unwound the following day, it would allow many other undesirable transactions to go unchallenged.

Recommendation

That the submission be declined.

Issue: Anti-avoidance rule dominant purpose

Submission

(Corporate Taxpayers Group)

The anti-avoidance rule in section GB 51B applies if an arrangement has an effect of defeating the intent and application of subpart FE. This should be a dominant purpose test.

Comment

The anti-avoidance rule can apply when an arrangement, action or omission has a purpose or effect of defeating the intent and application of the thin capitalisation rules. As this is not a dominant purpose test it means it can still apply when a taxpayer has more than one purpose for their action with only one of these purposes being tax avoidance.

Taxpayers who have entered into avoidance arrangements will almost always argue that the arrangement was entered into for commercial purposes and that tax avoidance was not their intention or was merely incidental. Inserting a dominant purpose of tax avoidance is a much higher threshold, particularly on a provision that is aimed at a taxpayer’s balance sheet.

Of the existing anti-avoidance rules, many of which are in subpart GB, the most common approach is to be triggered by a purpose or effect of the arrangement. Officials see no reason why this particular anti-avoidance rule should have a higher threshold than that applying to other areas of the Income Tax Act.

Recommendation

That the submission be declined.

Issue: Owner-linked debt when the borrower is a trust

Submission

(Chartered Accountants Australia and New Zealand)

We support the proposal to amend section FE 18(3B) to ensure the rules operate clearly in relation to trusts.

Comment

Officials note the support.

Recommendation

That the submission be noted.

Issue: Worldwide group debt percentage and on-lending concession

Submission

(Bell Gully)

The thin capitalisation rules should provide express guidance as to the make-up of residual debt that remains after applying the on-lending concession in section FE 13 so that the worldwide group test in section FE 18 can be applied with certainty. The passage of the Bill provides an opportunity to correct this omission given that the Bill already contains corrective changes to section FE 18.

The issue, broadly, is:

The on-lending concession effectively allows a deduction for interest on notes issued by a securitisation trust (Trust) to the extent they fund financial arrangements with non-associated borrowers.
After applying the on-lending concession, the Trust’s assets and debt for thin capitalisation purposes consist (respectively) of non-financial arrangement assets and notes funding those assets.
When applying the worldwide group test to the residual debt and asset position, it is necessary to identify what proportion of the debt is related party debt (such debt being excluded from the calculation of the worldwide group debt percentage).
However, the statutory provisions provide no guidance as to the split of the residual debt amount (after applying the on-lending concession) between related party and third party components. This makes the worldwide group test difficult to apply.
A reasonable approach might be to apply the same percentage split of the total third party versus related party notes to the residual debt amount. However, the statutory provisions provide no express support for this, and other allocations could be adopted (e.g. all related party debt, or all third party debt).

Comment

Due to the fungibility of money, when a borrower has a mixture of third party and related party funding, it is not possible to determine which specific source of funding has been allocated between the on-lent funding and the residual application of funds. Officials agree that apportioning the residual between third party and related party debt in the same proportions as before the on-lending concession is applied would be appropriate and recommend changes to FE 18 to achieve this.

Recommendation

That the submission be accepted.

[4] As introduced this was the highest credit rating of the group, this report explains that officials now recommend this be replaced by the credit rating of the member of the group with the highest unsecured third party debt.