Working Party on Tax Questions (Direct Taxation CCTB) Challenges of the digital economy for direct taxation - State of play and the way forward

Transcription

1 Brussels, 11 October 2017 WK 11067/2017 INIT LIMITE FISC ECOFIN MEETING DOCUMENT From: To: Subject: Presidency Working Party on Tax Questions (Direct Taxation CCTB) Challenges of the digital economy for direct taxation - State of play and the way forward Delegations will find attached a document from the Presidency in view of the Working Party on Tax Questions (Direct Taxation - CCTB) on 16 October WK 11067/2017 INIT LIMITE EN

2 Introduction 1. The EE PRES launched the discussion on the challenges of the taxation of the digitalising economy at the very beginning of its Presidency at the HLWP meeting of 4 July At that time, the OECD was already preparing for its interim report on the state-of-play of taxation of the digital economy for Spring 2018, but there was no active debate at the level of the EU. Some MSs had however already been taking unilateral actions (e.g. Hungarian advertisement tax) or these were under way (e.g. French court case against Google). 2. A number of developments have occurred since then, in particular: a. Presidency conference on 'Current Issues in European Tax Law', with a session on the concept of the virtual permanent establishment, in Tallinn on 7 September; b. Political statement ('Joint initiative on the taxation of companies operating in the digital economy') co-signed by 10 EU Finance Ministers; c. Session on the 'Corporate Taxation Challenges of the Digital Economy' at the informal ECOFIN Council meeting of 16 September 2017 in Tallinn; d. Communication by the Commission on a 'Fair and Efficient Tax System in the European Union for the Digital Single Market' (doc /17) issued on 21 September 2017; e. Discussions at the Digital Summit in Tallinn on 29 September 2017; f. Completion of the first round of technical discussions at WPTQ level on the concept of virtual PE (12 July), the sharing economy (21 September) and the valuation of big data (3 October 2017); g. Follow-up presentations at the ECOFIN Council meeting of 10 October Against this background, a common understanding was reached that, in the absence of physical presence, the current international rules do not allow to tax the profits of digitalising economy where the value is created. 4. Such an undesired outcome is a structural shortcoming caused by the outdated nature of international tax rules. It is not a matter of tax avoidance or something related only to certain large MNEs or companies entering the EU s internal market from certain 3 rd countries. 1

3 5. MSs have expressed a clear preference for an update of international rules on a global level in close co-operation with the OECD. This would ensure that the EU s and global businesses are treated equally and that there is no competitive advantage for third country companies due to the remaining outdated international tax rules outside the EU. 6. However, a growing number of MSs are also of the opinion that this co-operation with the OECD cannot result in only waiting for the OECD to reach a global agreement. Every day that we spend analysing and discussing the ultimate best solution to the challenges of the taxation of the digital economy, businesses are suffering from unequal competition, countries are losing valuable tax revenues, and unilateral measures are undermining the Internal Market. 7. It should also be highlighted that there are two other workflows taking place in parallel to the discussions in the Council: a. Firstly, the OECD is trying to choose a solution for the Spring 2018 interim report that would suit best for updating the international rules to allow to tax the value created even without physical presence. b. Secondly, the COM has announced in its recent Communication on the digital economy that it will analyse the policy options ahead of a possible proposal in Spring 2018 to tackle this issue, although the envisaged solution to be included in the proposal is yet undecided. 8. Deriving from the above and taking into account the members of the OECD and the procedure to discuss COM proposals, the EE PRES sees a unique opportunity for MSs to have a major influence on the choice of solution that will be proposed by the OECD and COM if MSs manage to be ambitious and prescriptive in the Council Conclusions in December The present room document aims at a better understanding of the pros and cons of each option in order to identify the preferred one. It also outlines the main messages envisaged for the draft Council Conclusions. 2

4 EU or global solution 10. As a starting point, there is no doubt that response to the challenges of taxation of digital economy should be global. However, if necessary, any agreed solution could be enacted at the EU level without the predating global agreement in the OECD. The latter situation will occur if the OECD has not reached an agreement before the time MSs see an urgency to act even without global agreement. 11. This possibility needs to be borne in mind while examining the different options for a solution described below as different solutions might have different implications if they are enforced only at EU level. 12. Further, it needs to be acknowledged that despite the EU solution being clearly less favorable than a global one, it is preferable over the possibility of having 28 different new tax regimes targeting digital services or specific companies in the digitalising economy. The harm of that kind of development in the Internal Market would be greater than the possible negative impact in the case of EU action without global agreement. There is no mechanism currently to prevent MSs from taking unilateral actions. Pros and cons of EU action without global agreement 13. Pros: a) Avoids unilateral measures by MSs and therefore protects the Internal Market against fragmentations; b) Potentially creates momentum to have a global agreement afterwards; c) Re-enforces a level playing field at least between EU businesses; d) Would capture most of the profits earned in the EU at least until a major restructuring of 3 rd country MNEs doing business in the EU. 14. Cons: a) Could lead to counteraction by 3rd countries, which would impact EU businesses and create further double taxation disputes; 3

5 b) Raises the question of whether MSs should unilaterally take measures to avoid double taxation in situations where its business has been taxed in 3 rd countries under similar measures; c) Mismatches between an EU approach and the rest of the world would create new opportunities for aggressive tax planning (e.g. new types of mismatches, deliberately setting up loss-making permanent establishments without physical presence). Pros and cons of globally coherent EU action following global agreement 15. Pros: a) Consistent application globally creating a level playing field for all comparable businesses; b) Reduces the risk of double taxation and the additional burden created by unilateral measures; c) Minimises impact on EU competiveness; d) Reduces the risk of creating new opportunities for aggressive tax planning. 16. Cons: a) Agreement at a global level could take a longer time to reach than at EU level, which would force the less digitalised businesses to keep standing the competitive disadvantage and MSs to keep losing revenue; b) All the additional time spent on waiting for a global agreement will guide MSs and 3 rd countries to enforce unilateral measures making reaching an agreement even more difficult. 4

6 Possible EU solution 17. Building on the discussions held in different forums (e.g. HLWP, WPTQ, Tax Conference in Tallinn, Informal Ecofin) and emphasising the need to create a level playing field for the whole digitalising economy while trying to avoid unwanted spill-over effects, the EE PRES considers that the following two main options should be further explored as possible solutions: an equalisation levy; a new nexus. 18. The equalisation levy option is quite straightforward as it works only outside the DTCs and therefore is by default an additional tax with its own tax base, tax rate and taxpayers. All of these components are still to be discussed and agreed. 19. The new nexus option works within the current corporate income tax system essentially relying on the already known rules of tax base and taxpayers and using the current tax rate. The OECD calls this 'tax nexus' or 'significant economic presence'. Others call this 'digital presence'. Essentially, everyone is talking about the same concept amendments to the definition of permanent establishment and changes to the transfer pricing and profit attribution rules to tax profits where the value is digitally created without physical presence. 20. The new nexus option can be enacted in different forms of which three are most promising: Amendment to bilateral tax treaties between Member States through a coordinated protocol (hereinafter Protocol); Amendment to ATAD accompanied by amendments to transfer pricing and profit attribution guidelines; Various adjustments to the 2016 Commission's CCCTB proposals, including a new apportionment rule. 5

7 Equalisation levy 21. The idea of introducing an equalisation levy was first presented by France and supported by Germany, Italy and Spain before the Informal Ecofin. The equalisation levy is perceived as a quick solution to an imminent problem that is the non-taxation of revenues earned by the companies who earn considerable income but have no physical presence in the market country. These countries have presented the equalisation levy as an interim solution that should remain in place only for the time that is required for reaching a comprehensive solution at the EU or global level. 22. The equalisation levy could be designed to have a narrower (fees from advertising only) or more comprehensive (digital services provided in the market country) scope. In order for it to apply to revenues received by the residents of third countries as well, it should remain outside the scope of existing tax treaties which means that the levy should not have elements that could be associated with taxes on income. 23. In addition to cross-border transactions, the equalisation levy should also apply to transactions taking place within one jurisdiction in order to avoid possible infringements of fundamental freedoms. When choosing to apply the levy to the revenues earned by the residents of third countries, the compatibility of such a levy with other international agreements (e.g. WTO agreements) should be analysed as well. 24. The design of the equalisation levy requires a complete formulation of all the main elements of a tax (e.g. base, payer, rate, exemptions) giving MSs full flexibility compared to amending corporate income tax rules. This of course is a challenge at the same time and none of these elements has been discussed nor elaborated in detail yet. In order to give an idea of how such a levy could work there are two examples provided in the Annexes: a. Annex 1: short summary of the equalisation levy enacted in India; b. Annex II: The French vision of an EU equalisation levy. 25. It needs to be highlighted that, independently from the design of the equalisation levy, it could only be considered as a temporary solution because by nature it does not fulfill the purpose of the EE PRES initiative: it will not level the playing field between companies, because it is difficult to effectively apply it to B2C business models, including the sharing 6

8 economy, and it will not re-establish MSs' right to tax value where it is created, because its tax base cannot be profits. 26. Pros: a) Consistent application across MSs from a certain date; b) Equally effective towards 3 rd countries; c) Tax on turnover/revenues would not be covered by the tax treaties and would therefore not violate them; d) Rules on exchange of information and assistance in collection of the OECD Multilateral Convention would apply; e) Could be enacted relatively quickly because it does not require amendments to transfer pricing and profit attribution rules or agreeing on a formula apportionment; f) Requires neither a renegotiation of tax treaties nor changing their interpretation. 27. Cons: a) All the main elements of a tax (e.g. base, payer, rate, exemptions) need to be agreed which might take a lot of effort; b) As the tax base cannot be profits, the equalisation levy does not follow the principle of taxing profits where the value is created; c) It is almost impossible to effectively enforce it on B2C business models (e.g. the sharing economy or subscription based models); d) It cannot take into account the value of data, because the data is flowing within MNEs without transactions; e) A levy that applies only to digital services could negatively affect the freedom to provide services if the digital services become taxed more heavily than other services not subject to that levy; f) Existing international dispute resolution mechanisms would not apply; g) Similar levies to digital services provided by EU companies could be introduced by 3 rd countries; h) Accumulation of taxes and levies on digital services might not be avoided. 7

9 The new nexus in the form of a protocol to DTCs 28. Since it is evident from the discussions held at the OECD that the global consensus with respect to changes in the international tax rules has been and will be difficult to achieve, an alternative presented at the Tallinn Tax Conference 1 would be to intervene at the level of the interpretation of existing tax rules, which could be implemented rather quickly and would not require any renegotiation of existing tax treaties. For the purposes of this document, this suggestion is hereinafter referred to as Protocol. 29. The idea of a Protocol essentially consists of interpreting the concept of a fixed place of business in Article 5 of the OECD Model as not only including cases of physical presence, but also those of virtual presence of an enterprise on its territory. The term fixed does not indeed necessarily imply physically fixed assets, employees, or any other attributes of a company. 30. The implementation of this solution within the EU could take place in three steps: a. Firstly, MSs would issue an authentic interpretation of the concept of a PE that considers a physical presence as unnecessary to have a fixed place of business. b. Secondly, the MSs would sign an agreement in which they formally declare their intention to interpret and apply their tax treaties in line with such a concept. c. Thirdly, the MSs would make an observation to the relevant paragraphs of the Commentary on Article 5 of the OECD Model (2017), which are affected by this interpretation. 31. The effects of the Protocol within the EU could operate immediately after MSs sign, ratify and implement it. Consequently, MSs would have to interpret all their tax treaties on the basis of the new standard to comply with the content of the Protocol. As a result, all of the rules applicable to physical PEs would automatically apply to the broader concept of a PE, which would also include the virtual PE, following the signature, ratification and implementation of the EU Protocol. 1 Y. Brauner & P. Pistone, Adapting Current International Taxation to New Business Models: Two Proposals for the European Union, 71 Bull. Intl. Taxn. 12 (2017), Bulletin for International Taxation IBFD (published online 5 October 2017) 8

10 32. Such interpretative protocols to the tax treaties are nothing new in international tax law (see Annex III for an example). In this case, however, a greater number of tax treaties would be affected than usual. 33. Pros: a) Does not require a renegotiation of tax treaties; b) Almost consistent application across EU MSs with the possibility for 3 rd countries to join individually without any action from the OECD; c) Does not require the adoption of secondary EU law as the matter is currently dealt in tax treaties, therefore MSs would retain their competence to legislate in this particular aspect of taxation; d) The division of taxing rights between Member States with respect to activities requiring physical presence would remain largely unchanged, i.e. the existing PE thresholds (construction PE, agency PE) in bilateral tax treaties would not be modified; e) Once an agreement on the content of the interpretative provisions has been reached, the implementation would be relatively quick and straightforward; f) Changes in the profit attribution and transfer pricing rules could be addressed in the same instrument (interpretation of terms? assets, risks, functions of Article 7 could be elaborated); g) Existing international dispute resolution mechanisms would continue to apply; h) Should a different global consensus be reached on that issue, a protocol could easily be repealed. 34. Cons: a) Would not apply in case there is no tax treaty between the Member States, in that case the domestic PE rules would need to be amended to achieve the desired effect; b) Would not apply to tax treaties between MSs and 3rd countries unless the 3 rd countries join the Protocol; c) It would not work without adapting transfer pricing and attribution of profit rules (e.g. what is the role of risks, functions and assets in the absence of physical presence?); 9

11 d) The date of application would depend on the completion of domestic procedures, which means it would vary among MSs. The new nexus in the form of an amendment to ATAD 35. The Virtual Permanent Establishment (hereinafter VPE) concept would constitute an amendment to the criteria determining whether there is a permanent establishment or not. It would be similar to the concept of construction site permanent establishment, worded as an exception to the general rule and therefore only modifying and not replacing the current rules (see Annex IV for examples). 36. In practical terms, it could be an amendment to the ATAD (constituting ATAD3). As the directive supersedes DTCs between MSs, it would create a level-playing field irrespective of the current wording of the DTCs and even in the case of absence of DTC between MSs. 37. The conditions for determining when a VPE is considered to be created would be the same as in the Protocol option. The relationship with transfer pricing and profit attribution rules is also similar the VPE concept would not work without adapting them. If the VPE concept is agreed at OECD level, one can expect that the OECD would simultaneously amend its appropriate guidelines. If this concept is enforced only at the EU level in the absence of agreement at OECD level, then clear rules would be needed to guide MSs on how to override the OECD guidelines in a co-ordinated way. However, it is questionable whether such rules would fit in an EU directive. 38. Pros a) Could apply to situations between MSs irrespective of the treaty network within the EU; b) Consistent application across the EU from a certain date; c) Could be extended to cover transparent entities as well as corporates (ATAD2 structure); d) Building on the experience of ATAD2, it could be designed to have an effect on 3rd countries; 10

12 e) The scope would be defined by targeting digital services, therefore no need to decide whether whole corporate entity or group is affected (contrary to Digital- CCCTB) f) Would fit all business models, including the sharing economy, irrespective of the source of revenue (B2B or B2C); g) Would allow at least partially taking into account the value of data; h) Could have whatever kind of specific threshold most suitable to target who we want to target (number of customers, size of turnover, level of service localisation, etc.); i) Staying within the limits of corporate income tax allows to solve all the double taxation disputes under current EU and international mechanisms. 39. Cons a) The approach might need to override or underride treaties with 3rd countries this will lead to disputes with 3rd countries; b) It will not work without adapting transfer pricing and attribution of profit rules (e.g. what is the role of risks, functions and assets in the absence of physical presence) which means there is a lot of work to be carried out beyond agreeing on the mere wording of the VPE; c) In the absence of an agreement at OECD level, special EU rules are needed to guide MSs on how to override the OECD guidelines in a coordinated way. The new nexus in the form of adjustments to the CCCTB proposals 40. The Digital-CCCTB option would be the CCCTB, as currently proposed by COM, with at least the following amendments: a. Changes to the scope (limiting it to corporate entities and PEs mostly engaged in providing digital services and not to corporate entities and the PEs engaged in traditional economy, unless member states decide otherwise, while abolishing the threshold); b. Incorporating VPE to the definition of permanent establishment in Article 5; c. Changes to the formula apportionment to take into account value created without physical presence. 11

13 41. One can say that the Digital-CCCTB option is the VPE option with a harmonised tax base, using a formula apportionment instead of the transfer pricing and attribution of profit guidelines. 42. As a harmonised tax base and formula apportionment are not feasible at OECD level, these need to be developed wholly at EU level irrespective of whether the OECD can agree on a VPE concept or not. 43. Without agreement on a VPE concept at OECD level, it is nearly impossible to design the Digital-CCCTB option to have an effect on 3 rd countries. In case the OECD reaches an agreement on a VPE concept, adapted transfer pricing and profit attribution rules would still be needed in relation to 3 rd countries in addition to a formula apportionment. 44. Pros: a) Could apply to situations between MSs irrespective of the treaty network within the EU; b) Consistent application across the EU from a certain date; c) Potentially could enhance EU competitiveness instead of damaging it even in case of absence of agreement on a VPE concept at OECD level (CCCTB itself has been argued to be business friendly and pro-growth); d) Would fit all business models, including the sharing economy, irrespective of the source of revenue (B2B or B2C); e) Would allow to take into account the value of data; f) Staying within the limits of corporate income tax while harmonising the tax base creates good opportunities to avoid double taxation disputes within the EU; g) Staying within the limits of corporate income tax allows to solve all the double taxation disputes under current EU and international mechanisms; h) In the absence of an agreement at OECD level, there would be no substitution of existing rules but rather the creation of new ones, because at the moment the digital economy lacks applicable corporate income tax rules in most MSs; i) In the absence of an agreement at OECD level and due to the formula apportionment, there is no need to amend profit attribution and transfer pricing rules. 12

14 45. Cons: a) The structure of the CCCTB would not allow to extend the scope to transparent entities; b) The structure of the CCCTB would allow to define the scope only on the basis of corporate entities/permanent establishments/groups and not a separate ring-fenced approach specifically for 'digital services - which would create problems in situations of equal share of activity in the digital and traditional economy; c) It is almost impossible to design the Digital-CCCTB option to have an influence on 3 rd countries without a prior agreement on a VPE concept at OECD level; d) It does not work without agreeing on a VPE concept and apportionment formula; e) Council discussions on the 2011 CCCTB and 2016 CCTB have proved that it is very challenging to reach consensus on the common tax base and formula apportionment in a short time frame. Questions to delegations: 1. Do you agree with the analyses presented above? 2. Which course of action would you prefer for the EU and for the global level? Way forward 46. Based on the feedback of delegations, the Presidency will present a draft of Council conclusions for the first round of comments at the HLWP meeting of 27 October. 47. A possible structure for these conclusions could be: a. Introduction: reminder of relevant European Council conclusions, reference to recent events, and reiteration of some general principles; b. Opportunities and challenges: outcome of the discussions at WPTQ/HLWP on the challenges of taxation of digital economy including sharing economy and valuation of data; c. Considerations of options for action at the EU and global level; d. Outlining the preferred option of the MSs to be communicated to the OECD and the COM; 13

15 e. Call to reform international tax rules according to this preferred option. 48. The objective of the Presidency is to have these conclusions adopted at the ECOFIN Council meeting of 5 December 2017, for input into parallel OECD discussions on their interim report to be delivered at the G20 Finance Ministers meeting of April 2018 and for the Commission in preparing the possible legislative proposal in Spring Questions to delegations: 3. Do you agree with the proposed way forward? 4. Which other important aspects should be addressed in the Council Conclusions? 14

16 India s equalisation levy 2 Annex I Equalisation levy has been said to be presumptive tax whose aim is to eliminate double nontaxation and ensure tax neutrality between different businesses and business models. For creating a level playing field between service providers with and without a permanent establishment in India, equalisation levy on some specified digital services provided by non-residents was implemented in India from 1 June The aforementioned specified digital services include online advertisement and any provision for digital advertising space or any other facility or service for the purpose of online advertisement between business-tobusiness (B2B) transactions. Payments by natural persons are outside the scope of equalisation levy. The levy is applied to services which amount to more than Rs (approximately 1297,98 euros) per year supplied by non-residents who do not have a permanent establishment (PE) in India and are not therefore liable for income tax in India. The rate applied is 6 percent and has to be deducted from the amount paid or payable to a non-resident. The sum deducted from the paid or payable amount has to be paid to the Central Government by the seventh day of the month immediately following the calendar month in which the sum has been deducted. For the non-resident service provider no additional compliance burden is created. The Indian payer has the obligation to submit an annual statement that contains information about payments for specified services made to non-residents and the amount of equalisation levy that has been withheld. The statement needs to be filed by 30 June immediately following the financial year. According to the authors of the article, one of the possible negative effects of the equalisation levy is that it might not be creditable against income tax and may therefore result in double taxation for the non-resident company. Furthermore, equalisation levy can affect the competitiveness of a company because the impact of the levy depends on whether the levy is passed through to (business) customers or the non-resident without PE being willing to absorb it and reduce its margin on the services provided in India. 2 Equlization levy, Ashok K. Lahiri, Gautam Ray, D. P. Sengupta, Brookings India Working Paper - 02, January

17 Annex II FACT SHEET Re: EU equalisation levy on digital services. Purpose of the EU equalisation levy on digital services Internet and digital technologies bring a major contribution to value creation and growth worldwide. It has proved difficult, however, to tax the profits generated by these new business models, based in particular on data, in the countries where profits are generated. The digital era thus represents a challenge for the current international tax principles, which rely on brick-and-mortar installations and people functions. Every company should pay its faire share of tax to public policies and competitive distortion should be curtailed. Therefore, in order to ensure neutrality, it is vital to develop clear and adequate rules. Therefore, rules governing territorial nexus of companies' profits and how they are attributed between countries must be modified at an international level. In particular, it is for the EU, which is promoting the Single Digital Market, to commit to developing innovative solutions. An EU equalisation levy on digital services can be a directly operational response, as of course the long-run solution will be an adaptation of international tax rules to the digitisation of the economy. Scope The scope of the digital economy taxation covers potentially any sale or service (business-tobusiness and business-to-consumer) provided through Internet or other similar automated networks, which generally entails the processing of data. A definition of electronically supplied services already exists in the EU, for the purpose of value added tax (VAT). Though it is limited to digital services, it could offer a starting point for technical work 3. Liability Firstly, consistent with its purpose, the EU equalisation levy should be designed to target large taxpayers (groups of companies above a certain size). Indeed, those mutinationals have massive impacts on market competition and public revenue. In addition, they have greater ability to use arrangements allowing them to engage in aggressive tax planning strategies. It is also necessary to avoid excessive additional administrative burden for SMEs and start ups. A criterion for defining a size-related threshold could refer to the total consolidated digital turnover of a group. For instance, in the EU context, the cap of a medium-sized enterprise 3 Services which are delivered over the Internet or an electronic network and the nature of which renders their supply essentially automated and involving minimal human intervention, and impossible to ensure in the absence of information technology. (Council implementing regulation n 282/2011 of 15 March 2011). 16

18 is turnover. For the CbC reporting directive or the CCCTB project, the threshold is Subject to thresholds, the EU equalisation levy could apply to any company carrying on digital activity in the internal market whether established or incorporated in an EU Member State or not. Another option would be to ring-fence the tax to companies that have no taxable presence for their digital activity in the EU according to current corporate tax rules (neither a subsidiary nor a permanent establishment). Such an approach, however, may raise legal questions (e.g.: WTO rules) and allow circumventing schemes. Further examination is required. Finally, for each MNE, a second threshold (eg: turnover, number of commercial transactions/web users) will be necessary to test the level of digital tax presence in each MS. The company will become liable to tax there only if it is fulfilled. Economic neutrality of the EU equalisation levy The aim of the EU equalisation levy is to ensure neutrality for companies. Consequently, the avoidance of double taxation is a key requirement. If a company already pays corporate income tax on a specific digital activity within the EU, it should be taken into account in calculating the EU equalisation levy For instance, its amount could be reduced by the CIT paid in whichever MS for the same activity. This offset mechanism will lead to reduction or suppression of the «EU equalisation levy» in such a case. For the same purpose, it would alternatively be possible to provide for a consolidation mechanism. The EU equalisation levy would be assessed at EU level for every digital activity conducted in the internal market by an MNE and reduced by the overall amount of CIT paid to Member States. Under this approach, the EU equalisation levy would be paid only to a single MS (where a company is incorporated or established, as the case may be). It would then be repaid to each MS according to an apportionment key taking into account the digital activity conducted on its terrritory. Base The tax base of the EU equalisation levy should correspond to the digital turnover made on (but not necessary paid from) the territory of an EU Member State. Pursuant to that principle, the definition of the nexus to match a digital activity and a territory will be crucial. As the EU equalisation levy aims at capturing the value creation generated by digital activities, it should rely for instance on data collection or contact with web users. When necessary, as it works for VAT, the MS of the final beneficiary of a service (possibly different from the direct client) could be confirmed through a set of criteria: place of consumption, public and private regulation, market specific features, language In particular, it would prevent possible schemes to transfer artificially transactions off-shore. Rate The rate of the EU equalisation levy should be defined in accordance with its neutrality purpose. It has to be a proxy of a CIT contribution for the same digital activity. 17

19 Therefore, it should be decided at the EU level on the basis of a profitability analysis. Of course, cost-efficiency ratios are diverse depending for instance on the business model, size and/or market positioning of the company. An average margin (without excessive standard deviation) could be defined for the companies covered so as to produce a bottom-line impact equivalent to CIT. Further examination will be necessary in this area. A targeted scope for the tax would make this technical work easier. It would also be possible to contemplate different rates according to types of business model or size. Finally, there should be a discussion whether a single rate at EU level is preferable (such convergence would be simpler, economically neutral and avoid competition) or if it should depend on the CIT level in each MS. Administration The EU equalisation levy has to be implemented in a simple way. In this respect, the experience of the VAT Mini-One Stop Shop (MOSS) is helpful as it allows the companies to file a single and dematerialised return for their overall EU activity. They have not to comply with diverse national procedures. The application of the EU equalisation levy will firstly rely on the companies declaration, thus on accounting information that they hold on their own activities. However, the existing framework for exchange of information and assistance in tax collection (EU directives, OECD multilateral convention - MAC, bilateral tax treaties, CBC reporting...) will help the MS assess and enforce the EU equalisation levy. Further examination will be necessary on the impact of those provisions, and possible additional measures. Finally, other obligations that may be created beyond tax matters for digital companies operating in the EU ( European Passport, obligation to have at least one legal entity or a representant in a MS) would also allow to provide for additional safeguards for the purpose of the EU equalisation levy. 18

20 PROTOCOL Annex III At the moment of the signing of the Convention between the Grand Duchy of Luxembourg and the Republic of Estonia for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income and on capital, both sides have agreed upon the following provisions, which shall form an integral part of the Convention: 1. With reference to Article 4: A collective investment vehicle which is established in a Contracting State shall be considered as a resident of the Contracting State in which it is established and as the beneficial owner of the income it receives. For purposes of paragraph 1 of Article 4, the term "collective investment vehicle means: a) in the case of Estonia, any pool of assets (common fund) established or company founded for collective investment on the basis of Investment Fund Act, and b) in the case of Luxembourg, (i) an investment company with variable capital (société d' investissement à capital variable); (ii) an investment company with fixed capital (société d investissement à capital fixe); (iii) an investment company in risk capital (société d'investissement en capital risque); (iv) a collective investment fund (fonds commun de placement), as well as any other collective investment vehicle established in either Contracting State which the competent authorities of the Contracting States agree to regard as a collective investment vehicle for purposes of this paragraph. II. With reference to Article 25: The competent authority of the requesting State shall provide the following information to the competent authority of the requested State when making a request for information under the Convention to demonstrate the foreseeable relevance of the information to the request: a) the identity of the person under examination or investigation; b) a statement of the information sought including its nature and the form in which the requesting State wishes to receive the information from the requested State; c) the tax purpose for which the information is sought; d) grounds for believing that the information requested is held in the requested State or is in the possession or control of a person within the jurisdiction of the requested State; e) to the extent known the name and address of any person believed to be in possession of the requested information; f) a statement that the requesting State has pursued all means available in its own territory to obtain the information, except those that would give rise to disproportionate difficulties. IN WITNESS WHEREOF the undersigned, being duly authorised thereto by their respective Governments, have signed this Protocol. DONE in duplicate at Brussels on this 7th day of July 2014, in the English language. 19

21 Annex IV Examples of Virtual Permanent Establishment proposed in the past 1. Academic Luc Hinnekens has proposed the creation of the virtual permanent establishment already in The taxing nexus for electronic commerce under his proposal should be the continuous commercially significant conduit of business activity. The source State would have the taxing right even in the absence of a permanent establishment in that State. 2. The OECD has discussed the permanent establishment definition for virtual businesses in the past. 5 The OECD has in this report proposed the following options to be considered for the extension of the PE definition: 1) Virtual Fixed Place of Business PE - The place of business would be a virtual web site for the maintenance of which the enterprise uses a server located in the same jurisdiction. The OECD has concluded that this approach would remove the need for intangible property within the jurisdiction. However, the requirement for a place would still apply to ensure the degree of permanence; 2) Virtual Agency PE implies the extension of existing dependant agent PE concept to electronically equivalent dependent agents, i.e. when contracts on behalf of the enterprise are habitually concluded through technological means rather than in person; 3) On-site Business Presence PE taxation on source, with a focus on the economic presence of an enterprise within a jurisdiction. 3. BEPS Action 1 Discussion Draft analysed also the challenges of the digital economy in the context of permanent establishments. 6 A number of potential factors were suggested in the Discussion Draft to determine when a fully dematerialised digital activity was conducted. Discussion Draft suggests that when an enterprise engaged in 4 Luc Hinnekens, Looking for an Appropriate Jurisdictional Framework for Source-State Taxation of International Electronic Commerce in the Twenty-first Century (1998). Intertax OECD, CTPA. Are the Current Treaty Rules for Taxing Business Profits Appropriate for E-Commerce? Final Report 6 Public Discussion Draft BEPS ACTION 1: ADDRESS THE TAX CHALLENGES OF THE DIGITAL ECONOMY 20

22 certain fully dematerialised digital activities it would have a permanent establishment if it maintained a significant digital presence in the economy of another country. For an enterprise engaged in a fully dematerialised business, a significant digital presence could be deemed to exist in a country when, for example 7 : 1) A significant number of contracts for the provision of fully dematerialised digital goods or services are remotely signed between the enterprise and a customer that is resident for tax purposes in the country; 2) Digital goods or services of the enterprise are widely used or consumed in the country; 3) Substantial payments are made from clients in the country to the enterprise in connection with contractual obligations arising from the provision of digital goods or services as part of the enterprise s core business; or 4) An existing branch of the enterprise in the country offers secondary functions such as marketing and consulting functions targeted at clients resident in the country that are strongly related to the core business of the enterprise. 4. BEPS Monitoring Group has suggested to replace permanent establishment with a significant presence test. 8 The criteria for applying the test should include: (a) relationships with customers or users extending over six months, combined with some physical presence in the country, directly or via a dependent agent; (b) sale of goods or services by means involving a close relationship with customers in the country, including (i) through a website in the local language, (ii) offering delivery from suppliers in the jurisdiction, (iii) using banking and other facilities from suppliers in the country, or (iv) offering goods or services sourced from suppliers in the country; (c) supplying goods or services to customers in the country resulting from or involving systematic data-gathering or contributions of content from persons in the country. 5. European Parliament has also discussed the shortcomings of permanent establishments. 9 In their opinion the scope of the CCTB directive should also include a digital business establishment. 7 Ibid., para p.8 21

23 European Parliament has proposed to add point 33a to the Directive Article 4 paragraph 1: (33a) digital business establishment means an establishment which is specifically directed towards consumers or businesses in a Member State and for that purpose regard shall be had to whether the establishment is conducting its business under the top level domain of the Member State or of the Union or, in relation to mobile application based businesses, is distributing its application via the Member State specific part of a mobile application distribution centre. 6. Academics Hongler and Pistone have also voiced for the support of updating the permanent establishment rules. 10 Their proposal supports the introduction of a new article 5(8) of the OECD Model with the following wording (see section 4.2.): If an enterprise resident in one Contracting State provides access to (or offers) an electronic application, database, online market place or storage room or offers advertising services on a website or in an electronic application used by more than 1,000 individual users per month domiciled in the other Contracting State, such enterprise shall be deemed to have a permanent establishment in the other Contracting State if the total amount of revenue of the enterprise due to the aforementioned services in the other Contracting State exceeds XXX (EUR, USD, GBP, CNY, CHF, etc.) per annum. 9 European Parliament Committee on Legal Affairs, Draft opinion of the Committee on Legal Affairs for he Committee on Economic and Monetary Affairs on the proposal for a Council directive on a common Corporate Tax Base (COM(2016)0685 C8-0472/ /0337(CNS)) 10 Peter Hongler, Pasquale Pistone, Blueprints for a New PE Nexus to Tax Business Income in the Era of the Digital Economy, IBFD Working paper 20 January