Skip to main content
Inland Revenue

Tax Policy

Chapter 4 – Problems with the existing rules measuring cross-border income


4.1 Introduction
4.2 Source rules - technical problems
4.3 Absence of adequate transfer-pricing rules and approved methodologies
4.4 Variability of tax rates
4.5 Conclusion


The ideal foundation for a set of rules for taxing cross-border income would be:

  • source rules that allow income from each tax base to be measured accurately;
  • uniform rates of tax applied to each tax base.

Current tax rules could be moved closer to this foundation by:

  • clarifying New Zealand’s source rules, including transfer pricing; and
  • reducing disparities in the tax rates applying to the different forms of foreign investment.

4.1 Introduction

The current tax regime, as outlined in Chapter 3, seeks to levy tax on all three income tax bases on which New Zealand could levy tax:

  • New Zealand-sourced income of residents;
  • foreign-sourced income of residents; and
  • New Zealand-sourced income of non-residents.

As discussed in Chapter 2, an important policy issue is the weighting that should be placed on each of these bases so that the tax impost is appropriate given economic efficiency, fairness, practicality and compliance cost concerns and given New Zealand’s international obligations. This is a matter requiring on-going dialogue.

A more immediate issue is to ensure that, as far as practical, New Zealand’s existing rules provide a sound foundation for levying whatever is considered to be an appropriate tax impost on each base.

The theoretical ideal foundation would be source rules that allow income from each base to be measured accurately and the application of relatively uniform rates of tax on each base.

Current tax rules do not measure up to this theoretical ideal. Significant problems are:

  • a lack of adequate statutory guidance for taxpayers to determine the source of income. Some of these deficiencies are technical, relating to the structure of the legislation. Others are more significant in policy terms; for example, the lack of adequate transfer-pricing rules;
  • in contrast to the theoretical ideal, current rules can impose highly variable rates of tax within each source base. For example, foreign investment is subject to different tax rates according to whether it is debt or equity financed.

This document does not include proposals to change New Zealand’s existing tax rules to bring them into line with the theoretical ideal. That would not be practical. However, improvements can be made. The remainder of this Chapter discusses various problems with existing rules. The next Chapter outlines a reform programme to deal with these problems and chapters 6 to 8 canvass details of that reform programme.

4.2 Source rules - technical problems

The main legislative source rule is found in s.243 of the Act. However, there are a number of other sections covering source issues, including ss. 245, 293 and 307. In addition, the general expenditure deduction provision (s 104) and the existing transfer-pricing provision (s 22), are important to the determination of net New Zealand-sourced income.

Technical problems with these rules tend to fall into one of four categories:

  • insufficient statutory detail on where income is sourced;
  • lack of explicit apportionment rules;
  • structural problems regarding whether statutory apportionment rules apply to gross or net income;
  • practical problems with the determination of source, leading to inconsistent rules and rules that are difficult to apply.

While some of these problems may seem to be relatively minor, at least when viewed in isolation, any weaknesses in the basic source rules are likely to flow through into the transfer-pricing rules. General income and expenditure source rules must be the basis upon which transfer-pricing rules are built, because transfer-pricing rules largely focus on the manipulation of prices to alter the source of net income.

4.2.1 Insufficient statutory detail as to where income is sourced.

The main statutory provision determining the source of income for New Zealand tax purposes, s.243(2), does not provide a general criterion for defining New Zealand-sourced income. Instead it merely lists some of the more common specific forms of such income. As a result, the section can provide insufficient guidance for determining the source of those forms of income which do not exactly match those listed and which therefore fall within the “catch-all” provision of paragraph (r). In those circumstances, it has been necessary to resort to common law rules to determine source. This can create uncertainty.

Problems in the definition of New Zealand-sourced income lead to uncertainties over whether income is foreign-sourced and, if foreign-sourced, from precisely where. The Act has no general definition of foreign-sourced income. It seems to work on the general presumption that income not sourced in New Zealand is foreign-sourced. The exception is dividend income. S 307 specifies that dividends paid by a company resident in a country which has a DTA with New Zealand are deemed to be derived from a source in that country. S 293(3) extends this source rule to dividend income from non-DTA countries.

As well as causing general uncertainty about the source of income, the above rules do not integrate well with New Zealand’s foreign tax credit rules. Those rules limit tax credits to the level of New Zealand tax payable on the income from a specific source on which overseas tax is paid. Thus the absence of general rules determining foreign-sourced income creates problems.

4.2.2 Apportionment issues

The general source provision (s.243) implies that some forms of income with more than one source must be apportioned to determine the extent to which they have a New Zealand source. In addition, s.245 explicitly requires business and contract income to be apportioned. S.104 then requires expenditure to be apportioned between that which is used to derive New Zealand-sourced income and that which is not.

Despite these apportionment requirements, the legislation does not provide any explicit guidance on either the process or the methodologies that taxpayers should use to apportion their income and expenditure between different countries. This can create problems, not only for non-residents calculating their net New Zealand-sourced income, but also for residents calculating the proportion of their net income that they have derived from each foreign country.

A specific apportionment problem arises with expenditure incurred jointly in deriving New Zealand and foreign-sourced income. Joint costs incurred in one country can give rise to assessable income in more than one country. For example, research and development costs incurred by a parent company in one country may increase the profitability of its branches and subsidiaries in other countries. Similarly, head office management expenses incurred in one country can produce assessable income in a number of other countries. One approach to joint cost allocation is to allocate on a marginal cost basis. However, this has some practical limitations. Overall, the joint costing issue in the international arena raises the same basic issues as in the domestic context. Joint costing in this context was considered by the Valabh Committee in their Tax Accounting Issues document.

4.2.3 Structural problems

In common with the rest of the income tax legislation, it can be unclear whether forms of income are intended to be expressed in gross or net terms. For example, it is not explicitly stated whether “business income” as used in s.243(2)(a) is intended to be net income (after deduction of expenses) or gross income (with expenses then deducted separately under provisions such as s.104). The Inland Revenue Department interprets s.243(2) as applying exclusively on a gross basis.

However, this can raise some other issues. For example, if non-New Zealand expenses are deductible under the normal provisions of the Act there seems to be no explicit requirement for those expenses to have a jurisdictional link with New Zealand. If a deduction is taken under s.104, that jurisdictional link is automatic. But that is not the case with some other deductibility provisions such as s.106(1)(h)(ii) where a jurisdictional links needs to be inferred.

Finally, while s.243(2) may best be interpreted as using concepts such as “income” in gross terms, it is schematically clear, though not explicitly stated, that the term “income” when used in reference to the foreign tax credit limit in s.293(2) should be interpreted as net income. Even in this case, there is no statutory guidance on the allocation of expenses in determining overseas net income; for example, whether an appropriate proportion of head office management expenses incurred in New Zealand should be deducted prior to calculating the net income on which foreign tax credits are allowed.

A clearer structuring of source rules would provide better guidance for taxpayers and better protection for the tax base.

4.2.4 Practical problems with implementing source rules

In principle, the economic source of income is the country in which the value-added activity leading to the generation of income takes place. In general, New Zealand’s source rules tend to follow this notion. However, the fungibility of money makes it difficult to apply the value-added principle to the determination of the source of income generated by debt or equity,

The interest source rule in s.243(2)(m) applying to money lent overseas seems in general to follow the value-added principle, but can be difficult to apply in practice. For example, New Zealand’s current rules deem interest on money lent overseas to a person living overseas (but resident in New Zealand for tax purposes) to have a New Zealand source (and thus be within the New Zealand tax net) if the funds are used for a non-business purpose. Clearly such a rule is difficult to administer.

New Zealand defines the source of dividend income as the residence of the paying company. This is an arbitrary rule. The global profits of a multi-national enterprise will be derived from the value-added activities in each country in which the company operates, not just the country in which it is tax resident. The dividend source rule is quite different from the interest source rule. This difference is difficult to defend on theoretical grounds, especially since New Zealand moved to an imputation system (as extended by FITC) which reduces the general differences in tax imposts on debt and equity.

New Zealand’s dividend source rule, when combined with the basic tax rule that a multi-national company resident in New Zealand is taxable on its worldwide profits irrespective of the ownership of that company, creates specific problems. It means that income which, in economic theory, is sourced overseas (where the value-added activity takes place) is subject to New Zealand company and dividend tax, even though the ultimate owners of the income are overseas resident shareholders. This imposes a tax penalty on multi-national companies operating out of New Zealand.

Other countries, such as the USA, have tried to mitigate this sort of problem by reducing tax levied on income derived offshore and then distributed to overseas owners. New Zealand should consider giving relief along similar lines if this can be done with reasonable compliance and administrative costs and without undermining the other desirable core features of New Zealand’s international tax regime.

4.3 Absence of adequate transfer-pricing rules and approved methodologies

Another factor that complicates the calculation of net New Zealand-sourced and foreign-sourced income is the absence of adequate transfer-pricing rules incorporating approved transfer-pricing methodologies which multinational enterprises can use to value their transactions between New Zealand and overseas arms of their operations. This creates uncertainty in the calculation of the net New Zealand income statistics and provides opportunities for multinational enterprises to manipulate prices on actual or deemed transactions in order to lower the New Zealand tax payable.

This area is now largely governed by s. 22 of the Act. This section empowers the Commissioner to readjust New Zealand income if a business carried on in New Zealand produces less income than the Commissioner might expect and when the business is:

  • controlled by non-residents; or
  • carried on by a non-resident company or a company controlled by non-residents or persons who have control of a non-resident company.

There are, however, a number of problems with this section, making it deficient as a key part of New Zealand’s source provisions. In particular:

  • the control tests which must be met before the provision comes into play are limited and therefore can be circumvented;
  • insufficient guidance is provided about what is, in terms of the section, the appropriate level of net income to be sourced in New Zealand; and
  • the relationship between s. 22 and other sections of the Act is not clear.

Clearer transfer-pricing rules, possibly buttressed by thin-capitalisation measures, would improve taxpayer certainty as well as base maintenance.

4.4 Variability of tax rates

As outlined in Chapter 3, statutory imposts of New Zealand tax on cross-border income flows are quite disparate. While differences in the level of tax imposed on outbound as opposed to inbound investment can have an economic justification, it is particularly difficult to justify:

  • direct inbound investment being subject to a statutory rate higher than the rate on domestic investment (33%). Under current rules, the tax rate on repatriated income from direct investment equity investment can be as high as 43% when the normal corporate tax rate plus NRWT on dividends (at the normal DTA tax rate) are combined;
  • the large differences that exist between the tax rates applying to foreign debt versus equity investment. As outlined in Chapter 3, foreign debt investment can be taxed at rates as low as about 1% whereas foreign equity investment is subject to a statutory rate of at least 33%.

The latter feature provides an incentive for foreign investment to be financed with debt rather than equity. The high tax on direct equity investment exacerbates this problem. To the extent that some investors cannot substitute debt for equity, differential taxation of debt and equity will distort the pattern of foreign investment in New Zealand.

Reforms to reduce the very high statutory tax on direct foreign investors therefore seem justified. Such a measure would in itself reduce some of the pressure on the debt/equity distinction but further measures to reduce pressure on this boundary should also be considered.

4.5 Conclusion

The current statutory rules applying to cross-border income can be improved. This would involve clearer and more detailed rules for determining New Zealand-sourced and foreign-sourced income and apportionment rules for income that has more than one source (where apportionment is allowed) and expenditures incurred in producing both New Zealand-sourced and foreign-sourced income. Such rules should include approved methodologies for determining transfer prices.

Reforms along these lines would help both resident and non-resident taxpayers in their attempts to fulfill their obligations to accurately determine their New Zealand tax liabilities and help protect the New Zealand tax base.

In addition, a reduction in the variability of tax rates applying to different types of cross-border investment would improve the economic efficiency of current rules.