Inland Revenue - Tax policy Tax Policy

News and information about the Government's tax policy work programme, including:
- proposed changes to the laws that Inland Revenue is responsible for
- updates on the progress of bills through Parliament
- policy announcements

Chapter 4 - The OECD's measures


4.1 This chapter sets out the measures being considered at the OECD, so the public can understand them and submit their feedback.

4.2 The OECD measures at this stage are high level proposals that have not been developed in detail. Accordingly, there is still uncertainty as to how they would apply in practice.

Summary of OECD measures

4.3 There are two broad measures currently being considered at the OECD:

  • A measure to allocate greater taxing rights over a multinational’s profits to market countries. The measure would not require the multinational to have a physical presence in the country. There are three proposals being considered for this purpose (only one of which would be adopted):
    • A limited proposal for digital services only, focussing on social media, digital advertising, multi-sided platforms and data.
    • A broader proposal, which would allow greater taxing rights to market countries (such as New Zealand) based on certain marketing intangibles created there by multinationals. This would apply beyond the digital economy.
    • A proposal which provides for apportionment of a digital multinational’s profit to market countries in which it has a significant economic presence. The apportionment would be based on an agreed formula and would depend on certain factors such as sales, assets and user participation.

It is possible that the OECD may adopt an option that incorporates elements of more than one of the three proposals, or an alternative proposal.

  • A minimum tax measure. This proposal would apply beyond the digital economy and would ensure that multinationals pay a minimum level of tax on profits earned in low tax countries. This proposal addresses some remaining base erosion and profit shifting (BEPS) issues and is not specifically directed at the digital economy (although it would also apply to digital companies).

4.4 Most of the proposals are not specifically targeted at the digital economy. With the exception of the first profit allocation proposal, they all apply more broadly, which reflects the views of many countries that the problems with the international framework are not limited to the digital economy. Accordingly, while these proposals arose out of the OECD’s work on the digital economy, they represent a wider reform of the international tax framework. For this reason, the OECD’s proposed measures have also been referred to as BEPS 2.0.

4.5 We describe these proposals in more detail in the next sections. The OECD’s February 2019 consultation paper, Addressing the tax challenges of the digitalisation of the economy,[29] contains further explanation and details of the measures. Consultation closed on this paper in March 2019 with over 200 submissions received.

Measure 1 – Reallocation of taxing rights

4.6 This measure seeks to address the ability of multinationals to be significantly involved in the economic life of a country without being subject to income tax there. It involves fundamental changes to the existing nexus and profit allocation rules, in order to expand the taxing rights of market countries over such multinationals.

4.7 There are three proposals being discussed at the OECD for this measure, each supported by a different group of countries. The proposals have significant differences, including in their scope and rationale. However, they all have the same objective – to allow the taxation of value created in a market country that is not recognised under the current framework. Some of the proposals also share similarities, such as the use of a simplified residual profit split method to allocate profits in the user participation and marketing intangibles methods. Therefore, the proposals could be combined or more closely aligned by the OECD going forward.

Proposal 1 – User participation approach

4.8 This is the narrowest of the proposals. It considers that the main problem with the international tax framework is that it does not recognise the value generated in a country by the active participation of certain digital companies’ users. This active user participation contributes to the value of the company’s brand, the generation of valuable data and the establishment of a critical mass of users which increases the value of the company for other users. However, this value creation is not recognised under the current profit allocation rules, as those rules only consider the activities of the taxpayer (and not those of its customers) in determining where value is created.

4.9 Examples of valuable active user contribution include user generated data and user generated content – such as videos posted on YouTube and pictures posted on Instagram. In addition, some platforms derive much of their value from the size of the user base. That is, the more people that use the platform, the more useful and valuable it becomes. For example, a social media platform with a single user would be worth very little, while another social media platform, with identical functionality but with one billion users, would be very useful and valuable. This increase in value of the digital platform in proportion to the size of its user base is referred to as a network effect.

4.10 The proposal would only apply to those types of business activities for whom user participation is a significant source of value. These are:

  • social media platforms;
  • search engines; and
  • online market places.

4.11 Accordingly, the proposal would only apply to a narrow subset of digital companies.

4.12 The proposal would work by:

  • deeming some of the profits from these in-scope business activities to be attributable to their users in a particular country; and
  • giving that country the right to tax those profits, regardless of whether the company had a permanent establishment or other physical presence there.

4.13 The user participation proposal considers that the value contributed to a company by its users is generated in the market countries where those users reside. Therefore, the market countries should be able to tax the company on the value generated by those users.

4.14 Mechanically, the proposal would involve these simplified steps:

  1. The residual profit of the group from its in-scope business activities needs to be calculated. The residual profit is a transfer pricing term, and it refers to the profit of a company that remains after all of the routine activities of the company (or group) have been compensated on an arm’s length basis. It can be thought of as the entrepreneurial profits of the company. The proposal acknowledges the existing challenges with determining residual profit under the current transfer pricing principles. Accordingly, the calculation of the residual profit under this proposal may require the use of simplified formulas.
  2. A portion of this residual profit would need to be attributed to active user participation. This allocation could be determined by looking closely at the relevant business and considering the value contributed by users in that particular case. Alternatively, it could be determined under simple pre-determined percentages, which could vary by the particular type of business.
  3. The residual profit attributable to users’ participation would be allocated between the various countries where the users were located. This would be determined according to an agreed metric or metrics, such as the numbers of users in each country and the revenue derived by the multinational from the country.
  4. The countries would be given the right to tax those profits, regardless of the degree to which the multinational was physically present in that country.

4.15 The residual profit attributable to the market countries would be determined by treating a multinational as a single entity – that is, it would look through the separate legal identity of the various subsidiaries and try to allocate the total profit of the entire group, rather than the profit of each separate company. However, mechanically the proposal would need to reallocate some of the residual profit currently earned by specific companies in the group to market countries (as the current income tax framework does not apply to groups on a consolidated basis). A tax credit would be given by the home country of the company earning those residual profits.

4.16 This proposal would not affect the allocation or taxation of the routine profit earned by members of the group. It would only affect the companies earning residual profits under the current transfer pricing rules.

4.17 The proposal could incorporate a range of additional restrictions based on the size of the business to further reduce the administrative burden for tax administrations and taxpayers.

4.18 Further information on the proposal and its background are set out in the OECD’s public consultation paper, and in a United Kingdom position paper on corporate tax and the digital economy.[30]

Example 2:  User participation proposal

In this example a digital group (Sales Co, Digital Co and Marketing Co) are all outside New Zealand. They collectively earn $5 billion of global profits. Principal Co operates the platform and retains the group’s residual profit. Sales Co and Marketing Co carry out routine activities, the arm’s length remuneration for which is $500 million each. New Zealand has one percent of Principal Co’s global users.

The process for determining the amount of income to be allocated to New Zealand in these circumstances would involve these mechanical steps:[31]

Step 1: Determine the global profit of the group on a consolidated basis. In this case the global profit of Sales Co, Principal Co and Marketing Co is $5 billion.

Step 2: Determine the profit allocable to the routine activities of the group. In this case Sales Co and Marketing Co are allocated $500 million of profit each to reward them for their routine activities.

Step 3: Determine the residual profit by subtracting the routine profit from the total profit. In this case $5 billion total profit less $1 billion routine profit equals $4 billion residual profit.

Step 4: Determine the share of residual profits attributable to global user participation. In this case, the share is twenty five percent, meaning $1 billion of residual profits need to be allocated to global user participation (25% × $4 billion = $1 billion).

Step 5: Determine New Zealand’s share of the residual profits allocated to global user participation. This is determined in accordance with the proportion of the group’s global users that are in New Zealand. In this case one percent of the group’s global users are in New Zealand, so $10 million of profit is allocated to New Zealand (1% of $1 billion).

Step 6: New Zealand taxes its share of the profit. New Zealand is given the right to tax its $10 million share of the group’s profits at New Zealand’s standard corporate tax rate, regardless of whether Principal Co has a PE or other physical presence in New Zealand. Country B gives Principal Co a tax credit for the New Zealand tax paid.

Comments

4.19 The user participation proposal has some benefits. The notion that users create value for particular businesses is plausible. In allowing countries to tax this value, the proposal is consistent with a fundamental principle of the current international framework – aligning taxing rights with value creation.

4.20 The narrow scope of the proposal has advantages and disadvantages. On the one hand the proposal is narrowly targeted at those business types which create the most difficulties for the current tax framework. This narrow focus means most businesses would not be affected by the changes. In addition, we would expect the proposal to benefit New Zealand overall, given we import more highly digitalised services than we export.

4.21 However, the narrow scope of the proposal means that it does not fully address the wider tax challenges of digitalisation. It does not apply to the sale of goods or ordinary services over the internet for example, meaning non-residents could continue to supply goods to New Zealanders online without income tax being payable, while our own retailers are fully taxable on their profits.

4.22 The proposal also involves ring-fencing the digital economy, which the OECD recommended against. This is on the basis that, given increasing digitalisation, the digital economy is rapidly becoming the economy.[32] Arguably therefore any solution to the tax challenges of digitalisation needs to apply more broadly. In addition, a tax measure targeted at a small part of the digital economy may be less fair or efficient than a broader proposal. This is because a narrow measure may disincentivise involvement in that part of the economy compared to the other parts, which may equally be able to benefit from the current problems with the international tax framework.

4.23 There is also the question of how other advancing technologies will affect the international tax framework in the future, such as artificial intelligence, robotics and 3D printing. A solution narrowly targeted at user participation does not seem capable of addressing the potential future challenges posed by these technologies.

4.24 There are also technical issues with the proposal that would need to be addressed (as with all the proposals). Even determining the profit of a multinational is not straightforward, given the different countries in which it operates. Determining both the residual profit and the allocation of that residual profit to user participation is also inherently complex and subjective. If transfer pricing type principles were used, then the proposal would be difficult to administer (particularly for developing countries) and subject to dispute. For this reason, we consider that the proposal may need to use fairly simple formulas. However, these formulas would necessarily be inaccurate to some degree, as they would not reflect the particular circumstances of individual businesses. 

Questions for submitters

  • What do you think of this proposal?
  • Is it too narrowly targeted?
  • Is user value creation an appropriate basis on which to allocate profit to a country?
  • How do you think it would affect New Zealand businesses and consumers?
  • What kind of de minimis thresholds and restrictions do you think should apply to the proposal?
  • What do you think of the method for determining and allocating income?
  • Would simpler formulas to calculate this be better?
  • What kind of technical issues do you see arising if the proposal was implemented?
  • How would the proposal affect the development of the digital economy in New Zealand and globally?
  • What kinds of additional details would you like to see included in further development?

Proposal 2 – Marketing intangibles

4.25 This proposal has some similarities to the user participation approach, but it would not be limited to a subset of highly digitalised businesses.

4.26 The supporters of this proposal view the main challenges to the international tax framework as being scale without mass, and the importance of mobile intangibles. While these features are exacerbated by digitalisation, they are not limited to the digital economy.

4.27 These features mean that a multinational can essentially reach into a country, either remotely or through a limited presence, and develop a customer base and other marketing intangibles. These marketing intangibles represent a significant source of value creation in the market country, but this value is not currently taxable in the market country.

4.28 For example, a well-known online retailer with no physical presence in a country can develop a large customer base in that country and know more about those customers’ shopping preferences than the local bookshop. The same is true for many branded consumer goods companies that can directly and digitally engage with their customers. Whereas previously a company would need a local presence to develop a market, high value sales and marketing activities can now be carried out online, from another country.

4.29 To address this issue, the proposal would change the international tax framework to require marketing intangibles and their corresponding risks to be allocated to the market country. This allocation would occur independently of the current transfer pricing rules, and so it would not depend on the current transfer pricing factors used to allocate income from intangibles. The market country would also be given the right to tax the allocated income, regardless of whether the multinational had a physical presence there. The market country would also be able to tax the profit allocated under the proposal if the multinational had a subsidiary in the country carrying out only low value activities (for example, a limited risk distributor).

4.30 The proposal would directly not tax marketing intangibles themselves (or their transfer). Instead it would require a portion of the multinationals profit to be apportioned to the market country by reference to the value of the marketing intangibles located there.

4.31 The marketing intangibles covered by this proposal would include intangibles that relate to marketing activities in the country, or which aid in the commercial exploitation of a product or service or have an important promotional value (such as brands and trade names used in a country, customer data, customer relationships and customer lists).

4.32 The proposal would treat these kinds of intangibles as linked to the market country. This is because the positive attitudes in the mind of its customers are created by, and the customer information and data are acquired through, the active intervention of the multinational in the market. Accordingly, the proposal is presented as being consistent with the existing principle of aligning taxing rights with value creation.

4.33 In this regard, the proposal differentiates marketing intangibles from favourable demand conditions in the market country that exist independently of the actions of the multinational.[33] Marketing intangibles are also differentiated from trade intangibles, such as patents and other technology related intangibles generated by research and development. This is because trade intangibles are not seen as having an intrinsic link with market countries.

4.34 This proposal is expected to apply to highly digitalised businesses, given their reliance on marketing intangibles. In this context, marketing intangibles could include those generated by free search services, free email, free data storage and the like. Accordingly, the marketing intangible proposal could produce a similar result in practice for highly digitalised firms as the user participation proposal.

4.35 However, the proposal would also apply to non-digital companies with significant marketing intangibles in a country, unlike the user participation proposal. For example, it could apply to a fast-moving consumer goods company whose ability to set prices and shift products is significantly dependant on its brand value and consumer regard.

Mechanics

4.36 The mechanics of the proposal have not been developed in any detail. In general however, this proposal would function similarly to the user participation proposal, in that it would ignore existing transfer pricing rules and involve a simplified residual profit spilt. The proposal would likely involve these simplified steps:

  1. The residual profit of the multinational group would be determined by deducting the profit attributable to the group’s routine activities from its total profits (as with the user participation proposal).
  2. Part of that profit would then be allocated to marketing intangibles. This allocation would not include any profit attributable to trade intangibles. This allocation could be done under traditional transfer pricing principles, or it could be based on a simplified residual profit split method, which would use mechanical approximations.
  3. The profit allocated to marketing intangibles would be allocated between the various markets countries in which the multinational operates. This would be based on agreed metrics, such as sales or revenues. This would occur regardless of which entity in the multinational group held legal title to the marketing intangibles, regardless of which entity carried out the functions related to those intangibles, and regardless of how the current transfer pricing rules would ordinarily allocate the income from the intangibles.
  4. The market countries would be given the right to tax their share of those profits, regardless of the degree to which the multinational was physically present in that country.

4.37 Double tax is intended to be prevented (or at least minimised) by means of a tax credit granted by the countries where the residual profit was returned under standard transfer pricing principles.

4.38 The OECD consultation paper acknowledges that the scope of this proposal would need to be subject to some restrictions and limitations. The idea is that the proposal should only apply to businesses where the contribution of marketing intangibles to profits is substantial. These limits could include materiality thresholds (for example, cost ratios, size of customer base), de minimis exclusions, exclusion of certain industry sectors, exclusion of commodities, and so on. However, there is no detail on any of these yet.

Example 3: Marketing intangibles proposal

This example indicates how the proposal could work if it used simple formulas. The particular formulas are indicative only.

SM Group is a multinational which operates a social media platform with users around the world. The group provides free access to the platform and raises revenue by selling advertising on it. The head office in Country A developed and owns all of the platform’s technology.

According to its financial statements, SM Group has $27 billion of consolidated group revenue for the year, $15 billion in expenses and $12 billion in total profits. The Group’s expenses comprise $4 billion in cost of goods sold, $4 billion marketing and sales costs, $6 billion research and development costs, and $1 billion in overheads and administration.

The SM Group generates ten percent of its revenue from Country B. It has a subsidiary in Country B which provides marketing services and sales support. All of SM Group’s residual profit is returned in Country A under the current transfer pricing rules.

The application of the marketing intangibles proposal would involve these mechanical steps:[34]

Step 1: Determine the total amount of SM Group’s profit. The SM Group’s financial statements would be used for this purpose. These show a total profit of $12 billion.

Step 2: Determine the profit attributable to routine activities (that is, the routine return): The routine return is assumed to be ten percent of all expenses, other than cost of goods sold. This means the routine return is calculated as $11 billion expenses (excluding cost of goods sold) × 10% = $1.1 billion.

Step 3: Determine the residual profit, by subtracting the routine return from total profits. The total profits are $12 billion and the routine return is $1.1 billion. This means SM Group’s residual profit is $10.9 billion.

Step 4: Determine the proportion of residual profit attributable to marketing intangibles. A formula is used to do this. The formula divides the residual profit between the marketing intangibles and the trade intangibles. It does this by comparing the costs incurred by SM Group for each type of intangible – that is, it compares SM Group’s marketing and sales (M&S) costs with its research and development (R&D) costs. In addition, the formula assumes that expenditure on R&D contributes twice as much value as expenditure on M&S for social media groups like SM Group. Based on this, the residual profit allocated to marketing intangibles is calculated as follows:

$4 billion M&S cost                                × $10.9 billion residual profit

$4 billion M&S cost + (2 × $6 billion R&D cost)

= 0.25 × $10.9 billion

= $2.725 billion residual profit allocated to marketing intangibles

Step 5: Allocate the marketing intangibles profit to market countries. SM Group’s marketing intangibles profit is allocated between its market countries in proportion to SM Group’s revenue from each country. Country B contributes ten percent of SM Group’s total revenue, so Country B will be allocated ten percent of SM Group’s marketing intangibles profit. This means that country B will be allocated 10% × $2.725 billion = $272.5 million profit.

Step 6: Market countries tax the profit allocated to them. Country B will tax SM Group on $272.5 million at its ordinary corporate tax rate, regardless of whether SM Group has a physical presence in Country B. Country A will give SM Group a credit for this tax.

Comments

4.39 The proposal has the advantage of addressing the problems posed by scale without mass and reliance on intangibles, without trying to ring-fence the digital economy. Its broader scope means that it could be a solution to the wider tax problems proposed by digitalisation.

4.40 On the other hand, the proposal’s expanded scope means that it would apply to a much wider group of companies, and therefore the impact of any compliance costs or technical problems would also be magnified. For this reason, the proposal should arguably include strong limitations and exclusions to ensure it is targeted at large companies that derive significant value from their intangibles. However, these limitations and exclusions have not been developed yet.

4.41 One issue is that is still unclear how the proposal would apply to highly digitalised (and non-digitalised) companies in practice. These would also need to be appropriately taxed by the proposal in practice for it to be an effective solution.

4.42 This proposal would also affect more New Zealand companies than the user participation model. The Government would be concerned if the proposal had a serious adverse effect on our export sector. This possibility of this adverse impact is a real downside of the proposal (in its current form) from a New Zealand welfare perspective. Another concern is the potential increase in compliance and administration from the proposal (although this is something the OECD will try to mitigate in its design of the proposal). The Government is currently doing some work to ascertain the likely economic impact of the proposal on New Zealand, including our export sector. However, we will need to wait until the proposal is more detailed before we can come to any firm conclusions about this.

Questions for submitters

  • What do you think of this proposal?
  • Is it too broadly targeted?
  • Are marketing intangibles an appropriate basis on which to allocate profit to a country?
  • Do you think marketing intangibles are sufficiently linked to a country to permit local profit allocation?
  • How do you think it would affect New Zealand businesses and consumers?
  • How would it affect New Zealand’s export sector?
  • What kind of de minimis thresholds and restrictions do you think should apply to the proposal?
  • What do you think of the method for determining and allocating profits?
  • Would simpler formulas to calculate this be better?
  • What kind of technical issues do you see arising if the proposal was implemented?
  • How would the proposal affect the development of the digital economy in New Zealand and globally?
  • What kinds of additional details would you like to see included in further development?
  • Should the proposal be combined with the user participation proposal, and if so, how?

Proposal 3 – Significant economic presence

4.43 The final proposal was submitted more recently by the OECD. For this reason, it has less detail.

4.44 The rationale for this proposal is that the digitalisation of the economy and other technological advances have enabled business enterprises to be heavily involved in the economic life of a country without needing a significant physical presence. This has rendered the existing nexus and profit allocation rules ineffective.

4.45 The proposal would apply to any multinational that developed a significant economic presence in a country through technology or other automated means. This means the proposal would apply to the digital economy in the broadest sense of the word – but it would not apply beyond the digital economy. Accordingly, the proposal is broader than the user participation proposal, but narrower (in principle) than the marketing intangibles proposal.

4.46 Whether a multinational had a significant economic presence would depend on a variety of factors. At a minimum, the multinational would need to generate revenue from a country on a sustained basis. One or more digital factors would then also need to be present. The other factors could include things like the existence of a user base, a minimum volume of digital content, billing in the local currency, a website in the local language, the presence of delivery or support services in the country, marketing activity, and so on.

4.47 Once a significant economic presence was established, profit would be allocated to that presence using a fractional apportionment method. This method involves four simplified steps:

  1. The definition of the tax base to be divided. For example, the tax base could be determined by using the total group profits for accounting purposes.
  2. The determination of the allocation factors for dividing that tax base between countries. These factors could include sales, assets, employees and users (where relevant).
  3. The weighting of these allocation factors – that is, the factors could all be treated as equally important, or some factors may be given a higher weighting than others.
  4. The allocation of the tax base between the market countries, using the weighted allocation factors. 

Example 4: Significant economic presence proposal

This example illustrates general fractional apportionment methods. The actual method used under this proposal will likely differ in some details.

Ship Group operates an online platform selling various goods around the world. Customers order goods on Ship Group’s website or app. The goods are then shipped by Ship Group from the nearest regional warehouse to the customer. Ship Group is headquartered in Country A, which is also where all of its R&D is carried out. Ship Group’s residual profit under existing transfer pricing rules is returned in Country A (50%) and Country Z (50%).

Ship Group’s consolidated financial statements show that it has total worldwide assets of $30 billion, total worldwide employees of 15,000, total revenue of $20 billion and total profits of $12 billion.

Ship Group has no assets or employees in Country B. Ship Group derives $500 million of revenue from Country B.

The amount of Ship Group’s profit allocated to Country B is determined through these mechanical steps:

Step 1: Determine Ship Group’s tax base. In this case Ship Group’s tax base is its consolidated group profits for accounting purposes, which is $12 billion.

Step 2: Determine the allocation factors. In this case, the allocation factors, which will be used to divide Ship Group’s profit between its market countries, are assets, employees and sales.

Step 3: Weigh the allocation factors. Each factor has an equal weighting.

Step 4: Allocate Country A a share of Ship Group’s tax base using the allocation factors. The OECD consultation paper does not discuss this step. However, it could involve dividing Ship Group’s tax base between the allocation factors, and then further dividing its tax base between countries by reference to each allocation factor. This could be calculated as follows:

  • Divide Ship Group’s tax base between the allocation factors. Ship Group’s tax base is its $12 billion profit, and the three allocation factors have equal weighting, meaning that each factor is allocated $4 billion of Ship Group’s profit.
  • Allocate the profit attributable to each factor to Country B. This would be done by reference to Country B’s share of each allocation factor, that is, its share of Ship Group’s total assets, employees and revenues. This would be calculated as follows:
    • Ship Group does not have any assets or employees in Country B, so Country B is not allocated any of the $8 billion (in total) of Ship Group’s tax base that is allocated to these factors.
    • For the $4 billion allocated to revenue, Country B’s share would be:

$0.5 billion revenue from Country B × $4 billion

$20 billion total revenue

= 0.025 × $4 billion

= $100 million

  • Add Country B’s share of the tax base allocated to each of the three factors together – so $0 million (assets) + $0 million (employees) + $100 million (revenue) = $100 million.

Country B would tax Ship Group on $100 million at Country B’s ordinary corporate rate, regardless of whether Ship Group had a physical presence in Country B. Country A would give Ship Group a foreign tax credit for fifty percent of the tax payable in Country B and Country Z would give Ship Group a foreign tax credit for the remaining fifty percent.

4.48 This proposal may also involve consideration of other simplified methods for allocating profit, such as the deemed profits measures. Deemed profit measures deem certain types of business to have a predetermined profit margin. The taxpayer’s gross turnover is then multiplied by this profit margin, to give its net taxable income. For example, if a profit margin of ten percent was deemed for an industry and a non-resident made $10 million of revenue from that industry type in Country A, then the non-resident would have net taxable income in Country A of $10 million × 10% = $1 million.[35]

4.49 Finally, withholding taxes could be used under the proposal as a collection mechanism and enforcement tool. One possibility would be to have a gross withholding tax at a low rate, with the multinational able to elect to file a tax return and claim a refund for any excess tax withheld.

Comments

4.50 This proposal would apply to the digital economy in the broadest sense of the word, so it would be a comprehensive solution to the problems posed by digitalisation (unlike the user participation approach). On the other hand, it also involves ring-fencing the digital economy, which would be contrary to the views of the OECD and the New Zealand and Australian Productivity Commissions.

4.51 The proposal also involves the most radical departure from the current profit allocation rules. A fractional apportionment method would not allocate taxing rights strictly on the basis of value creation. Instead it would move closer to a destination-based tax, where countries are entitled to tax the profits of a company based on the amount of goods sold there.

4.52 In this regard, consumption in a country is already taxed by goods and services taxes. Income tax on the other hand has traditionally taxed the factors of production (that is, the assets and activities of the taxpayer which produce, sell and deliver the goods or services). This implies that income taxing rights should be allocated by reference to where the factors of production are located, rather than where consumption occurs.

4.53 On the other hand, the inclusion of other allocation factors, such as employees or assets would offset this destination bias to some degree (although these other factors also give rise to avoidance opportunities, which suggests they may need to be given a lower weighting).[36]

4.54 Further, it could be argued that the country providing the market is making a real contribution to the value of the multinationals which sell their goods there. This is on the basis that both supply and demand are required for a company to ultimately create value, and therefore both supply side and demand side locations should be allocated part of a multinational’s profits.

4.55 In addition, in selling into a market, a multinational is benefiting from the public infrastructure and legal system which effectively creates and supports that market. Supporters of the proposal argue that the Government that provides this infrastructure should therefore have the right to tax some of the profit of the multinationals that benefit from it.

4.56 In addition, this method would be much simpler to administer than the current rules. It would also largely remove the current strategies multinationals can use to avoid tax (although care would need to be taken that new avoidance opportunities were not created).

4.57 The proposal would in principle affect most New Zealand businesses that export goods or services online (although we would expect the proposal to provide an exemption for smaller businesses as it is developed). It would also allow New Zealand to tax most multinationals that sell goods or services to New Zealanders online.

4.58 One concern is that the more radical nature of the proposal may prevent it from gaining sufficient support from other countries. However, the proposal would have a better chance of acceptance if it was supported by businesses.

Questions for submitters

  • What do you think of the proposal?
  • Is it too wide?
  • Are the proposed allocation factors of sales, assets, users and employees sufficient?
  • How would it affect New Zealand consumers and businesses?
  • Would it be beneficial or harmful to New Zealand?
  • What technical issues can you see arising from it?
  • What further details (such as de minimis thresholds) would you like to see if the proposal was developed?

Other proposals

4.59 The OECD may also consider other proposals for reallocating profit. For example, the pharmaceutical and healthcare company Johnson & Johnson submitted an alternative proposal at the recent OECD public consultation on 13 and 14 March 2019. This provided for a fixed benchmark return to market countries (for example, 3% of sales), which could be increased or decreased depending on certain factors such as overall group profitability and marketing expense.

4.60 Participants were impressed by this alternative method, including businesses and the OECD Secretariat. Given the enthusiasm expressed for this approach there is a chance this method will also be considered by the OECD, alongside the existing three proposals. Johnson & Johnson’s proposal is available online along with all of the other OECD public submissions.[37]

Measure 2 – Minimum tax measure

4.61 The other measure being discussed at the OECD would ensure that multinationals pay a minimum amount of tax on their profits from low tax countries. This measure could be adopted in addition to one of the three proposals for measure 1 discussed previously. This measure could apply much wider than digitalised multinationals.

4.62 The rationale for the measure is that, while the OECD’s BEPS measures closed many of the gaps in the international tax framework, they did not provide a comprehensive solution to the problem of shifting profits into low tax countries. In particular:

  • BEPS Actions 8–10 (Aligning Transfer Pricing Outcomes with Value Creation) sought to align taxing rights with value creation, and so prevent the use of artificial structures to shift profit. While this prevented the use of aggressive tax avoidance structures, it is still possible to shift profits to low tax countries, provided a minimum level of economic activity is carried out there.
  • BEPS Action 2 (Neutralising the Effects of Hybrid Mismatch Arrangements) sought to prevent multinationals from exploiting differences in countries’ domestic tax regimes to generate double deductions or avoid paying tax. While these rules are effective against hybrid arrangements between two normal tax countries, it is still possible to generate the same overall result by inserting a low tax jurisdiction between the two countries.[38]

4.63 These issues are particularly acute for intangibles, which are prevalent in the digital economy, but they also apply in a broader context. For example, a group finance company can be located in a low tax country. This finance company could then generate un-taxed profits from providing internal loans to other group members. Those loans would in turn reduce the tax payable by the other group members in their home countries (though a deduction for the interest payments).

4.64 Furthermore, the current system facilitates a race to the bottom, where countries lower their corporate tax rates in order to be more attractive to foreign investors than other countries. This risks shifting the taxes needed to fund public goods onto less mobile sources, such as labour and consumption.

4.65 For these reasons, while the minimum tax measure would be effective in addressing the profit shifting strategies which digital companies are ideally placed to use, the measure would not be limited to the digital economy.

4.66 The measure comprises two rules:

  • An income inclusion rule. This would allow a country to tax its residents on income earned by their foreign branches or subsidiaries if insufficient foreign tax was paid on that income.
  • A base eroding payments rule. This would deny a deduction and/or treaty benefits for certain payments to foreign companies if the payment was not subject to a minimum effective rate of foreign tax.

Income inclusion rule

4.67 This would apply to a resident taxpayer with a direct or indirect interest in a foreign company over a specified percentage (for example, twenty five percent). It would require the taxpayer to return its share of the net income of the foreign company or branch if that net income was not subject to a minimum rate of tax. For example, if the taxpayer held fifty percent of a foreign company with net income taxed at below the minimum rate, the taxpayer would return fifty percent of the foreign company’s net income.

4.68 The minimum tax would a be a single, fixed rate that would apply across all countries (and would be lower than most countries’ domestic tax rates). If a multinational’s effective tax rate was below the minimum rate, then it would pay top up tax to that minimum rate (and not its domestic rate). For example, suppose a company was attributed with $100,000 of income under the rule, on which $5,000 (5%) of foreign tax was paid. The company’s domestic tax rate was 20% and the minimum effective tax rate under the rule was 10%. In this case, as the $100,000 of income was taxed at 5%, the company would need to pay additional top up tax of $5,000 under the rule, which would bring its total tax payable up to the 10% minimum rate. The company would not need to pay a further $10,000 to bring its tax up to its 20% domestic rate.

4.69 The amount of net income used to calculate a group’s effective tax rate would in principle be calculated under the parent company’s controlled foreign company rules or their domestic tax rules and would include a credit for any tax paid in the foreign country. However, to simplify the rules the OECD will also consider using a group’s financial accounts to calculate its net income.

4.70 For taxpayers with a foreign branch, the rule would achieve the same effect as for a taxpayer with a wholly owned foreign subsidiary. The rule would make the branch net income taxable in the taxpayer’s home country, with a credit for the foreign tax paid in the country where the branch is located.

4.71 There is little detail available yet about the rule, and a number of design features still need to be considered, including the minimum effective tax rate that would apply. Further important details are:

  • How the effective tax rate paid on the income in the foreign country would be calculated – for example, how would loss offsets (if any) be taken into account?
  • Whether the minimum tax rate should be calculated on a global basis (which would allow the blending of tax rates paid in low-tax and high-tax countries in calculating the multinational’s global effective tax rate), a country basis (so it would allow the tax rates of different group members in a country to be blended to calculate the effective tax rate paid by the multinational in that country) or a single entity basis?
  • The scope of the rule – the rule could be limited to only certain kinds of entities and payments.
  • The role of economic substance in applying the test. While the OECD paper states it would apply regardless of economic substance, this proposition has not been agreed yet and several countries have already signalled their opposition to it.
  • Safe harbours and de minimis thresholds.

4.72 The minimum tax measure is similar to the current United States global intangible low taxed income (GILTI) rule, and the measure will likely draw on elements of the GILTI rule as it is developed further.

4.73 The GILTI was enacted by the United States in its recent Tax Cuts and Jobs Act (December 2017). The GILTI taxes United States shareholders on the income of their foreign companies to the extent that income exceeds a certain threshold (ten percent of the value of the company’s tangible assets). The United States shareholder pays tax on this excess income at the rate of 10.5%, which is half the United States 21% corporate tax rate. Tax credits are allowed for eighty percent of the foreign tax paid, meaning the GILTI applies where a United States multinational pays tax of less than 13.125% on its overseas profits.

Example 5: The income inclusion rule

This example is indicative of how the income inclusion rule could work. It assumes some important details, such as a minimum effective tax rate of 12.5%, and that any income attributed under the rule will be taxed at the shareholder’s ordinary corporate tax rate.

Shareholder is resident in New Zealand. Shareholder owns fifty percent of ForeignCo, which is resident in Country B. ForeignCo earns $200 million of income, which it only pays a 1% effective tax rate on in Country B.

This 1% tax rate is below the income inclusion rule’s 12.5% minimum effective tax rate. Therefore, Shareholder is attributed with its share of ForeignCo’s income. Since Shareholder owns fifty percent of ForeignCo, Shareholder is attributed with fifty percent of ForeignCo’s $200 million income, or $100 million. Shareholder then returns this $100 million of income in New Zealand and pays tax on it at 28% (New Zealand’s ordinary corporate rate). Shareholder can claim a foreign tax credit in New Zealand for the 1% tax paid by Foreign Co in country B.

Base eroding payments rule

4.74 The rule would apply to a broad range of payments. This rule would deny a deduction for payments to a related party (for example, twenty five percent common ownership) unless the payment was subject to a minimum level of tax. Tax for this purpose would include both tax paid in the related party’s country and also any withholding tax paid in the resident’s home country.

4.75 The rule could alternately deny treaty benefits to payments unless the payment was subject to a minimum tax rate. For example, the rule could deny the benefit of Article 7 of a DTA, which prevents a country from taxing the business profits of a non-resident except to the extent they are attributable to a permanent establishment in the country. This would mean that, if a payment from one country was not subject to a minimum effective rate of tax in the other country, the first country would still be able to tax the payment even if it was not attributable to a permanent establishment.

4.76 Again, there is little detail available yet about this rule, and similar design issues arise as for the income inclusion rule. In addition, the rule would need to be modified in its application to dividends, to ensure that these were not double taxed. There would also need to be ordering rules to stop the base eroding payments rule and the income inclusion rule from both applying to the same income.

Example 6:  Base eroding payments rule

This example indicates how the base eroding payments rule could work. It assumes some important details, such as a minimum effective tax rate of 12.5%.

SubCo is resident in New Zealand. SubCo makes a $100 million interest payment to its parent, ParentCo. ParentCo is resident in Country B, and pays no tax on the interest payment in Country B. New Zealand charges withholding tax of 15% under its domestic law, but this is limited to withholding tax of 5% under the DTA between New Zealand and Country B.

Total tax is payable at the rate of 5% on SubCo’s interest payment to ParentCo (5% withholding tax charged in New Zealand + 0% income tax charged in Country B). This is lower than the rule’s 12.5% minimum effective tax rate, so New Zealand denies SubCo a deduction for its interest payment to ParentCo.[39]

Comments

4.77 A minimum tax measure seems to be a promising solution to the problem of tax avoidance by multinationals. It would significantly reduce the ability and incentive for multinationals to engage in global tax avoidance. A multinational will have less incentive to avoid local taxes if it must pay a minimum level of tax on those profits elsewhere. It would also remove the competitive benefits of countries lowering their tax rates (or providing preferential tax regimes), as tax forgone by a multinational in one country would still be payable in another.

4.78 Global minimum taxes have also been supported by some economists, such as Joseph Stiglitz,[40] as a way of solving the current problems with global tax avoidance. The United States’ adoption of a minimum tax with its GILTI is also a significant endorsement.

4.79 The effectiveness of the OECD’s measure would depend on several factors, most importantly the rate at which it was set. There are also several technical issues that would also need to be addressed, not least of which is how to determine the effective foreign tax rate.

4.80 There is also a question of compliance and administrative costs. The OECD’s minimum tax would need to be significantly easier to administer than the United States GILTI for example, particularly if it is to be applied by developing countries.

4.81 Therefore, while the OECD’s proposed minimum tax has potential, it needs to be developed in more detail.

Questions for submitters

  • What do you think of the proposed minimum tax rate?
  • What should be the minimum effective tax rate?
  • Should application of the rules be subject to some kind of substance or avoidance test?
  • Should the rules be limited to certain kinds of payments (for example, interest) or entities?
  • How do you think the rules would affect New Zealand businesses and consumers?
  • What details and design features would you like to see included in the rules going forward?

Likelihood and timing of an OECD solution

4.82 The Government is hopeful that there will be an OECD solution in a reasonable timeframe, but this will be challenging. While countries have agreed to reach a consensus solution by 2020, they still disagree on what should be done. The OECD is making progress on developing a potential solution, but this is currently being done on a without prejudice basis. This means that countries are not committing to ultimately support the technical solutions being developed. On the other hand, the increasing adoption of unilateral measures is making a strong case for a multilateral solution, while incentivising countries that benefit from the current framework to agree to changes.

4.83 The OECD is aiming to get G20 approval of its preferred measures in June 2019. If approval is given, then an OECD solution is more likely. On the other hand, failure to receive G20 approval would be a serious, perhaps fatal blow to an OECD solution (in a reasonable timeframe at least). Therefore, we should have a better idea of the chances of an OECD solution after the June G20 meeting.

4.84 If an OECD solution was achieved in 2020, it may be at a fairly high level and require further development. In addition, any agreed solution would still need to be implemented. This would require changes to DTAs, with a multilateral instrument needing to be agreed and drafted for countries to sign.

4.85 If we look at what happened previously in the BEPS project, the OECD made its final recommendations in October 2015, and the DTA related BEPS changes have just started coming into effect from 2019 (with the changes for many DTAs not applying until 2020).

4.86 Based on this, we consider that any OECD solution for the digital economy would not take effect for another three to five years after that solution was agreed. This means that if a solution is reached at the OECD in 2020, it may not be effective until 2025.

Revenue impacts of the OECD measures

4.87 It is not possible to estimate the revenue impact of the OECD proposals. This is because they lack critical details, such as their precise scope and de minimis thresholds.

4.88 The revenue impact is only one factor that the Government will consider it evaluating the proposals. Other important factors include the effect of the proposals on compliance costs, and their impact on New Zealand’s economic efficiency and wellbeing.

 

[29] Available at http://www.oecd.org/tax/beps/public-consultation-document-addressing-the...

[30] HM Treasury, Corporate tax and the digital economy: position paper update, March 2018 https://www.gov.uk/government/uploads/system/uploads/attachment_data/fil...

[31] We have broken down some of the steps from paragraph 4.14 into multiple steps to make the example easier to follow.

[32] This was also the conclusion of the Australian and New Zealand Productivity Commissions, as discussed in chapter 2

[33] There are some countries that think such demand conditions should entitle the market country to taxing rights. An important subtext of the marketing intangibles approach is that it seeks to expand a market country’s taxing rights, without supporting the right of countries to tax a multinational simply by virtue of providing the market for its products

[34] We have broken down some of the steps in paragraph 4.36 into multiple steps to make the example easier to follow.

[35] New Zealand currently uses the deemed profit method to tax non-resident general insurers, who are deemed to derive income equal to ten percent of their New Zealand sourced premiums (section CR 3 of the Income Tax Act 2007).

[36] For a discussion of the issues with these apportionment factors, and with formulary apportionment methods in general, see J Andrus and P Oosterhuis, Transfer Pricing After BEPS: Where Are We and Where Should We Be Going, Taxes: The Tax Magazine, March 2017.

[37] Available at https://www.dropbox.com/s/hou6dvuckmahoft/OECD-Comments-Received-Digital...

[38] Although the anti-avoidance rules of some countries may prevent this in some cases.

[39] It is not clear at this stage whether the base eroding payment rule would deny both a deduction and DTA benefits to the same payment – however it seems unlikely they would do both where the result of denying DTA benefits was to tax the payment at above the minimum effective tax rate.

[40] For example, see https://www.project-syndicate.org/commentary/corporate-tax-avoidance-end...