1. Top
  2. Policy Proposals
  3. Business Law
  4. Comments regarding the Public Consultation Document on Addressing the Tax Challenges of the Digitalization of the Economy

Policy Proposals  Business Law Comments regarding the Public Consultation Document on Addressing the Tax Challenges of the Digitalization of the Economy

March 6, 2019

Tax Policy and Statistics Division
Centre for Tax Policy and Administration
Organisation for Economic Co-operation and Development

Comments regarding the Public Consultation Document on Addressing the Tax Challenges of the Digitalization of the Economy

1. Introduction

Keidanren welcomes the Inclusive Framework's intensive work addressing the tax challenges of the digitalization of the economy. The Base Erosion and Profit Shifting (BEPS) project has played an important role in securing tax certainty for Multinational Enterprises (MNE groups) by promoting consistent action among countries and helping prevent and resolve disputes. We hope similar results will be achieved when the Inclusive Framework concludes its work on digitalization of the economy.

Digital technology is the driving force behind innovation and future growth in the world economy. Digital technology is transformative and is expected to help resolve important social issues. A simple and consistent long-term solution that promotes fair competition and avoids double taxation should be identified, whilst avoiding obstruction of development of digital technology and industries that use the digital technology. It is inappropriate to require MNE groups to shoulder additional tax burdens that go beyond addressing the digital economy. Prevention and elimination of disputes and double taxation is essential, and we expect the G20 and OECD to lead on these issues. Countries should refrain from introducing unilateral taxation measures until a long-term solution is agreed upon.

The proposals in the Public Consultation Document involve considerable changes to the principles of international taxation. While it is impossible to "ring-fence" the digital economy, some of the proposed taxation measures appear to be excessive. Discussions should be balanced and the principle of proportionality should be applied when considering actions. If the discussion will primarily focus on the marketing intangibles proposal, we believe that the identification of issues which really need to be addressed and revisions to the existing rules to the extent necessary to solve these issues may lead to a proportional long-term solution. Accordingly, there should be many cases where the existing transfer pricing methods work well.

Based on the above, we hereby respond to the questions for public comments:

2. Revised profit allocation and nexus rules (First Pillar)

(1) What is your general view on those proposals?

  1. (i) The "user participation" proposal
    The relationship between user participation and value creation requires further review. Even if we accept that user participation enables MNE groups to accumulate data, it is more natural to attribute most of the value creation to an entity of the MNE groups that analyses the data to improve their products and services. In addition to the examples given (social media platforms, search engines and online market places), the same is true of other businesses engaged in activities that could potentially involve user participation, such as the accumulation of data from the internet of things (IoT) or car journeys. In addition, where data is gathered to enhance after-sales services, profits are not generated directly from the information itself. Even if value of user participation is recognized to some extent, weighting is expected to differ considerably between sectors.

    We believe there may also be practical issues to address that affect the tax analysis. For example, MNE groups may find it difficult to gather detailed user data, or count users with dummy or multiple registrations.

    Foreign tax credits are unlikely to be available for tax imposed on sales proposed by some countries and a region, which may therefore result in double taxation. A Long-term solution should be based on income taxation.

  2. (ii) "The marketing intangibles" proposal
    We recognise that this proposal may be worth considering when contemplating long-term solutions. However, depending on the design of the mechanism, it may result in excessive allocations of taxation rights to market jurisdictions, which is not consistent with the basic principles of "alignment of taxation with the location of economic activity" and "alignment of transfer pricing outcomes with value creation." confirmed in the BEPS Final Reports. This proposal also raises concerns for increased double taxation and administrative burdens. If we are to consider a resolution based on this proposal, the review should be conducted in accordance with the principle of proportionality.

    In addition to highly digitalized business models, the scope of this proposal includes consumer product businesses using local limited risk distributors (paragraph 42). However, we question whether it is appropriate for the scope to be drawn so widely. Firstly, establishment of a local limited risk distributor in a marketing jurisdiction in and of itself should not suggest insufficient allocation of profit to the market jurisdiction. Secondly, it cannot be assumed that marketing uniformly plays a significant role in value creation for consumer product businesses. Thirdly, as pointed out in paragraph 61 of the Public Consultation Document, there are cases where marketing activities outside the market jurisdiction do not contribute to intangibles in the market jurisdiction.

    Adopting this proposal in its current form may require a complex calculation process where the total profit of the MNE group from a specific business is aggregated, from which the amount of routine profit is subtracted, based on which the portion of residual profit attributable to marketing intangibles is extracted for prorating to the market jurisdiction following a certain formula (paragraphs 47 and 48). This could perhaps be termed a global residual profit split method and deviates considerably from the arm's length principle. There are significant issues to be addressed. For example, which accounting/tax accounting standard should we use to identify total profit? How should we differentiate the specific business from other businesses? Who would provide assurance that the method is properly applied? How many cases can there be which cannot be resolved without resorting to such a method?

    For so-called traditional consumer product businesses, one-sided transfer pricing methods such as the transactional net margin method (TNMM) already work well in pricing related party transactions involving local limited risk distributors and we do not see compelling reasons to change the existing approach. In particular, MNE groups that do not engage in arrangements under which intangibles are transferred to low-tax jurisdictions, but which have established limited risk distributors in local market jurisdictions (paragraph 13) would be such an example. In theory, disputes may not arise if countries can agree on the definition of various terminologies such as marketing intangibles and calculation methods. However, such assumptions may be unrealistic. In our view this proposal as it stands will inevitably result in increased disputes, double taxation and administrative complexity, rather than tax certainty.

  3. (iii) The "significant economic presence" proposal
    This proposal is very problematic since it would require fundamental changes even in areas where the existing international taxation principles work well. The scope of this proposal is drawn too widely and accordingly we cannot support this proposal.

    The proposal contemplates calculations based on the fractional apportionment method, which applies a global profit rate to local sales (paragraph 52). First of all, we are not sure which accounting/tax accounting standard we should adopt when making calculations, and who would make a fair audit. It may be somewhat simpler compared to the marketing intangibles proposal, but the trade-off is fairness of the profit allocation. Under this proposal, the profit allocation that should be based on transfer pricing, taking into account functions performed and risks assumed by foreign affiliates, is replaced with a mechanical formula based on the overall profit rate of MNE groups. The result is a loss of fairness in taxation.

    Taxes based on sales may result in taxation where there is no income. If a permanent establishment (PE) already exists in the market jurisdiction and its profit rate is lower than the global average, taxation in the jurisdiction based on significant economic presence will cause significant tax uncertainty.

    Withholding tax is proposed as a collection mechanism (paragraph 55), but there have been cases of inappropriate refunds of withholding tax in some countries. In practice, it usually takes several years before a refund is received.

    For entities collecting the withholding tax, administrative burdens will increase as these entities will be responsible for confirming whether a payee is an MNE group member with a significant economic presence.

    We cannot adopt this proposal without an appropriate income allocation rule, a fair audit system, and certain and timely refunds.

(2) To what extent do you think that businesses are able, as a result of the digitalization of the economy, to have an active presence or participation in that jurisdiction that is not recognised by the current profit allocation and nexus rules?

We are aware that businesses with features described as "cross jurisdictional scale without mass" have been identified as examples of industries not captured by the existing profit allocation/nexus rules.

(3) What would be the most important design considerations in developing new profit allocation and nexus rules consistent with the proposals described above, including with respect to scope, thresholds, the treatment of losses, and the factors to be used in connection with profit allocation methods?

Our comment will focus on the "marketing intangibles" proposal. The scope of application and the taxation method are the most significant factors in designing the tax mechanism. We suggest that the scope of application be limited to MNE groups that depend considerably on marketing intangibles for value creation, and MNE groups that shift intangible-related income to low tax jurisdictions, rather than uniformly applying the proposal to traditional consumer product businesses. In addition, expanding the scope to all business-to-business transactions requires very careful consideration, as stated in paragraph 61. Thresholds should be established to reduce administrative burdens for small and medium-sized enterprises and start-up companies.

Flexibility should be permitted for enterprises falling within the scope of this proposal to use traditional transfer pricing methods, rather than uniformly applying the global residual profit split method. When allocating the residual profit, marketing intangibles should not be overvalued. For example, in the manufacturing industry, the result of research and development as well as capital expenditure (trade intangibles) are significant factors that promote purchasing by customers. In addition, the suggestion that purchasing power in the market jurisdiction is not relevant for the profit allocation of businesses (paragraph 33) warrants attention. As a result, we assume that additional profit allocation to market jurisdictions, if any, should be modest. Moreover, if residual profits are to be split, residual losses must also be split.

In practice, the treatment of offices etc. that perform big data analysis in the process of transfer pricing may become an issue.

Countries should ensure that the new residual profit split method does not lead to additional taxation including exit tax, which may be levied on a transfer of intangibles to another jurisdiction from a jurisdiction where the intangibles have accumulated under the current mechanism.

Furthermore, an appropriate method for compensating start-up losses corresponding to the business life cycle should be considered. In particular, initial losses from business development (for example, concerning significant amounts of risk-involved investment for data collection) will be incurred in the country where the headquarters is located, but once the business becomes profitable, the profits will be allocated to multiple market jurisdictions. To resolve this mismatch, we would suggest appropriate amount of residual profit be reported at the country where the business was developed, for example, by managing profit and loss on a cumulative basis and allocating profits to other jurisdictions only after certain period of time has passed since the business turned profitable. Having stated that, it should be noted that even in this case, care should be taken to strike the right balance between fairness and administrative complexity.

Suggested use of additional data in country-by-country reporting (CbCR) (paragraph 85) is inconsistent with the stated purpose of CbCR, which is high-level risk assessment. Business-line data can hardly be deemed to be country-by-country data. Some may think it will reduce administrative complexity compared to individual data reporting, but, based on our view that the scope of a long-term solution should be limited, an expansion of CbCR, which requires submission by all MNEs above certain threshold, does not meet the principle of proportionality. We therefore cannot support this suggestion.

(4) What could be the best approaches to reduce complexity, ensure early tax certainty and to avoid or resolve multi-jurisdictional disputes?

If a long-term solution is to be implemented based on Pillar 1, a profit calculation method should be clearly agreed upon. In addition, a mandatory binding arbitration must be introduced as a minimum standard and its implementation must be monitored to ensure tax certainty. If a global residual profit split method is to be introduced, a multilateral dispute prevention and resolution mechanism will be required, as disputes may involve more than two jurisdictions. In addition to the International Compliance Assurance Programme (ICAP) etc., we may need to consider establishing a permanent auditing institution and a dispute resolution body.

3. Comments on Global anti-base erosion proposal (Second Pillar)

(1) What is your general view on those proposals?

We understand that the global anti-base erosion proposal for the purpose of reducing tax competition among nations is being considered in light of the introduction in the US of the Global Intangible Low-Taxed Income ("GILTI") and the Base Erosion and Anti-Abuse Tax ("BEAT"). However, in terms of preventing tax avoidance, there are many areas where implementation of existing BEPS recommendations such as transfer pricing (Actions 8-10); Controlled Foreign Company (CFC) rules (Action 3); interest limitation rules (Action 4); and anti-treaty abuse rules (Action 6) should be sufficient. This proposal is therefore redundant. In recent years, corporate administrative burdens have increased sharply to respond to the BEPS Recommendations, and adding to those burdens is a matter of concern. Introducing a mechanism that only focuses on no or low taxation status without considering actual economic substance would have a negative impact on cross-border trade and investment.

An agreement on the long-term solution based on the First Pillar should be settled first. We do not feel a need to focus on the Second Pillar at this time.

(2) What would be the most important design considerations in developing an inclusion rule and a tax on base eroding payments? In your response please comment separately on the undertaxed payments and subject to tax proposals and also cover practical, administrative and compliance issues.

  1. (i) Income inclusion rule
    There is not much need for jurisdictions with robust CFC rules that reflect BEPS Action 3 recommendations to adopt the income inclusion rule. Even if it is to be adopted, the income of foreign subsidiaries with economic substance should be excluded. To determine whether minimum tax rates are being applied, adverse results should be avoided for entities that are reporting losses or are receiving legitimate preferential tax treatment, such as R&D credits, tax cuts on capital investments and participation exemptions.

    Assuming there is a parent company (country A), its subsidiary (country B) and its sub-subsidiary (low tax jurisdiction C), if countries A and B introduce the income inclusion rule, income of the sub-subsidiary in jurisdiction C may become subject to multiple taxation. Rules of application should be clarified, for example, by prioritising the tax mechanism of the jurisdiction of the parent company.

  2. (ii) Undertaxed payments rule
    If payments are included in taxable income in a residence jurisdiction and deduction of the payments is denied in a source jurisdiction, the payment will be subject to taxation in both jurisdictions. If this rule is to be adopted, its application should be limited to payments not taxed at all at the residence jurisdiction, or countries should be allowed to establish a sufficiently low minimum tax rate.

    The design of the mechanism should take into consideration the economic substance of the concerned parties and the business needs for the payments. Related party transactions occur because optimal and efficient supply chains have been built for the business. Some overseas payments, such as those comprising cost of goods sold, should be excluded. In addition, royalties paid by subsidiaries to compensate R&D costs borne by a parent company should not be denied. Interest payments are already subject to earnings stripping rules, thin capitalization rules and transfer pricing rules, and we seriously doubt that further regulation is necessary or desirable.

    Under hybrid mismatch rules in a certain jurisdiction, if dividends are partially tax deductible by the payer, the corresponding portion shall be taxable by the payee. Therefore, as stated in paragraph 105, a mechanism under which deduction of undertaxed payment is not denied in full, but on graduated basis reflecting the level of taxation in a payee jurisdiction of the recipient, may be worth considering.

    Dividends from foreign subsidiaries are exempted in many jurisdictions, and should be excluded from the scope of application (the same applies to (iii)).

  3. (iii) Subject to tax rule
    Based on the recommendations regarding prevention of treaty abuse (Action 6), the Multilateral Instrument (MLI) and bilateral treaty revisions are set to incorporate the principal purpose test (PPT) and limitation on benefits (LOB) clauses into individual tax conventions. We should first review the status of implementation. Regarding Article 7 (Business profits), replacing the attributable income principle with the entire income principle (force of attraction rule) would present significant problems.

(3) What, if any, scope limitations should be considered?

The income inclusion rule and the undertaxed payments rule define related parties as those with direct or indirect 25% common ownership (paragraphs 96 and 103). However, timely obtaining of data from parties with small ownership interests can be difficult. In addition, the subject to tax rule seems to imply a broader scope for Articles 11 to 13 (Interest, Royalties and Capital Gains) (paragraph 107), but we doubt the availability of data for this. In any case, the risk of base erosion is low in a 25% ownership situation. Related parties should be defined as those with more than 50% common ownership.

The income inclusion rule is to be applied on a per jurisdiction basis rather than on an entity basis, including foreign tax credits (paragraph 96). However, there is concern that this could result in a complex mechanism depending on the design.

Taxation of the included income at the full domestic rate (paragraph 100) is excessive. If the rule is to be implemented, taxation should be at the lower rate.

(4) How would you suggest that the rules should best be co-ordinated?

In theory, the income inclusion rule and the subject to tax rule may be preferable to the undertaxed payments rule in that double taxation can be eliminated with foreign tax credits. However, there are separate issues including administrative complexity and enforceability.

The undertaxed payments rule and the subject to tax rule should not be applied to the same transaction.

(5) What could be the best approaches to reduce complexity, ensure early tax certainty and to avoid or resolve multi-jurisdictional disputes?

If the mechanism is to be introduced in each jurisdiction, enhancement of foreign tax credit systems is critical to ensure elimination of double taxation. Some jurisdictions may face issues such as increases in tax credit amounts and extension of the carry-forward period.


Subcommittee on Taxation

Business Law