The Platform for Collaboration  on Tax      COMMENTS  RECEIVED  ON  PUBLIC  DISCUSSION  DRAFT:     The Taxation of Offshore Indirect Transfers—  A Toolkit        December 2017              International Monetary Fund (IMF) Organisation for Economic Co-operation and Development (OECD) United Nations (UN) World Bank Group (WBG)             Table of Contents    Aneri Dani & Associates ................................................................................................................................ 1  BIAC ............................................................................................................................................................... 4  BEPS Monitoring Group .............................................................................................................................. 11  CBI ............................................................................................................................................................... 21  China State Administration of Taxation ...................................................................................................... 30  Deloitte ....................................................................................................................................................... 33  International Chamber of Commerce (ICC) ................................................................................................ 41  Government of India ................................................................................................................................... 45  International Tax and Investment Center (ITIC) ......................................................................................... 51  Jubilee USA Network ................................................................................................................................... 69  KPMG .......................................................................................................................................................... 70  PwC ............................................................................................................................................................. 78  .......................................................................................................................................................... 84  Repsol  Sergio Guida ................................................................................................................................................ 92  Silicon Valley Tax Directors Group (SVTDG) ................................................................................................ 94  Tax Executives Institute (TEI) .................................................................................................................... 112  .............................................................................. 120  Transfer Pricing Economists for Development (TPED)  United States Council for International Business (USCIB) ........................................................................ 133      ii    1 Aneri Dani Associates Comments on discussion draft titled “The Taxation of Offshore Indirect Transfers – A Toolkit” I. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers (‘OIT’) reasonable? • Despite the broad title of Toolkit, it only emphasizes on taxation of OIT of shares1 principally deriving value from immovable properties. While no explicit reason has been pointed out for distinction between immovable and movable properties, the economic consideration for taxing OIT has been primarily linked with Location Specific Rents - a concept more closely associated to immovable property than movable property. However, other economic considerations could be evaluated to justify taxation of OIT of shares deriving value from assets other than immovable properties. • On a conceptual level, taxation of OIT closely resembles the concept of exit taxes. Many developed countries including US, Canada, etc. levy exit taxes on individuals permanently exiting their home country. Thus, where multinational groups extinguish their ties with source country by initiating an OIT, isn’t it fair to assume that they should also pay their share of taxes based on the same economic considerations that back levy of exit taxes in case of individuals? • By limiting the right of source country to tax only those cases where immovable properties (though with extended meaning) are involved, the Toolkit does not comprehensively address this burning issue which is the need of the hour, given the tax and political importance it has gained in various countries around the world. II. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? • As acknowledged in the Toolkit, Location Specific Rents could also arise, for instance, from access to domestic markets. The fact that there may be difficulties to gauge and distinguish from rents associated with brand names or intellectual property, is not an apt reason for not developing rules to tax them. Enabling provisions to tax OIT involving all kinds of assets (and not restricted to immovable property alone) would help tax values linked to these attributes as well. III. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? Model 1, i.e. deemed disposal, is favoured over Model 2 in Toolkit for reasons of simplicity, tax cost adjustment logic and ease in enforcement. However, the following points need consideration: • As acknowledged in the Toolkit, Model 1 may result into economic double taxation where the seller’s country of residence also imposes tax on such transfer. Since taxpayer in residence country (seller) and source country (local entity) would be different, taxes paid 1 In this document, transfer of “shares” is intended to include transfer of “shares or comparable interests” 2 in source country may not be available as a credit. Effectively, while the rationale behind Model 1 is to plug the loophole of double non-taxation arising in typical OIT structures, this being a uniform rule and not GAAR provision, may result into economic double taxation for genuine OITs, defeating the primary objective of avoiding double taxation. • Also, since Model 1 anyways runs the risk of double taxation, source countries may possibly be more inclined towards formulating alternative solutions complementing their own tax framework better rather than adopting Model 1. • Under Model 1 a tax cost step-up is granted in the hands of local entity which becomes relevant for any future direct or indirect transfer of control. Thus, this model is likely to mitigate risk of double taxation of unrealised gain upon subsequent change of control at same intermediate holding entity level or otherwise. However, as acknowledged in the Toolkit this is a typical feature that arises when multi-tiered holding structures are involved – whether domestic or offshore. For instance, assume a case of domestic indirect transfer of shares of a company, instead of direct sale of the underlying immovable property. The source country may levy tax on the seller for transfer of shares. Similarly, the source country may also levy tax on the company on subsequent transfer of such immovable property. No tax cost step-up is granted on account of taxes paid on previous sale of shares. Accordingly, step-up in tax cost as envisaged under this model may ultimately result into lower tax share for source country. • Also, it could very well be a case that tax cost of “local assets in hands of the local entity” may be substantially different to tax cost of “shares of local entity in the hands of non- resident seller”– resulting into higher/lower (typically lower) tax share for source country, depending on facts of each case. For instance, the tax cost of shares of local entity in the hands of non-resident seller may be lower where the acquisition of local assets has been financed from funds other than those infused by the non-resident seller by way of share capital (like borrowings, retained earnings, etc.) • Finally, it may be worth highlighting that, since the underlying principle of Model 1, propagates shifting of source based tax to residence based tax, countries may cite this as a precedent to carry out unilateral amendments in their domestic legislation for establishing taxing rights over varied transactions which currently fall outside their ambit due to tax treaty restrictions. In view of this fundamental drawback, Model 1 entirely loses its relevance. It is rather inexplicable to see Model 1 being favoured by collaborative efforts of International Organisations whose principle objective is to develop solutions for avoiding double taxation. • Despite the complexities involved, Model 2 appears to be more logical and appropriate for taxation of OIT. III. Are the complexities in the taxation of these international transactions adequately represented? 3 • Guidance on the methodology to determine Market Value of assets may be provided in the Toolkit to lay down a common principle that countries may use. This will reduce the risk of double taxation and disputes on account of difference in valuation norms used by countries involved. 4 William Morris Chair, BIAC Tax Committee 13/15, Chaussée de la Muette, 75016 Paris France The Platform for Collaboration on Tax Submitted by email: GlobalTaxPlatform@worldbank.org October 20, 2017 Ref: DISCUSSION DRAFT: THE TAXATION OF OFFSHORE INDIRECT TRANSFERS – A TOOLKIT Dear Members of the Platform for Collaboration on Tax, BIAC is pleased to have the opportunity to comment on the Discussion Draft: The Taxation of Offshore Indirect Transfers – A Toolkit (the “Discussion Draft”) issued on 1 August 2017. However, in our view, the Discussion Draft (which “continues to be commented on and reviewed by the Platform partners”) still requires substantial work and should be further refined and then released again for additional public comment. We are concerned that the focus of the Discussion Draft appears to be more on encouraging countries to take the decision to tax Offshore Indirect Transfers (OITs), than it is on encouraging and facilitating a uniform approach for consistent and equitable implementation once countries have made the complex policy decision to tax OITs. We believe whether OITs should be taxed or not (and whether that should be within or outside of a treaty framework) is a matter for sovereign nations to decide based upon their own policy objectives and should thus not be considered in the context of a “toolkit”. Nevertheless, the Discussion Draft proposes potentially significant shifts in taxing rights for “source” and “residence” countries. Decisions on significant shifts in taxing rights ought to be debated among countries in the appropriate multilateral forum and not resolved by guidance of this type issued without debate among the countries. Our understanding of what the toolkits were meant to be was that they should be a box of tools – or a “how to” manual – to ensure that developing countries (in particular) could design tax rules to were clear, administrable, and as simple as possible in order to enhance certainty and (we assume) economic growth in those countries. Put slightly differently, they were to be the practical guide to implementing some of the complex BEPS decisions that could otherwise overburden some tax authorities. They were not, however, intended to articulate new tax policy considerations and recommendations as this document appears to do. ******************************************* Just to highlight a number of specific concerns in the document, which we spell out in more detail in the attachment, we would note:  When advising on the options to tax, sufficient consideration should be given to ensure maximum neutrality and symmetry in dealing with offshore indirect transfers. Furthermore, there should be a focus on ensuring that economic decisions are not distorted by tax rules. Various issues are very relevant to consider in relation to indirect transfers, including (for 5 example), how to determine the potential capital gain, how to ensure a step up in asset value, whether deferral is possible, how to limit the scope to ensure most effective taxation whilst avoiding unintended taxation and how to deal with offshore indirect capital losses. In our view, these important issues have not been adequately addressed in the Discussion Draft.  In particular, too little attention is given to the difficult area of valuation. We recommend that further thought be given to the situations where the inherent value of assets will be realised through locally generated (taxable) profits rather than ever increasing capital gains. We have outlined high-level responses to some of the questions raised by the Discussion Draft to illustrate our specific concerns, but hope to have an opportunity for further, more detailed responses to a revised discussion draft.  The Toolkit must also avoid the inference that all companies seek to aggressively avoid taxes. This perception would encourage countries to introduce far broader measures than may be necessary in order to protect their tax bases (e.g. imposing taxation on indirect transfers only in abusive cases, thus allowing them to focus their attention on tax abuse while limiting unintended consequences to investments). More targeted anti-abuse measures could be achieved (or complemented) by advance rulings. Under such regimes, the appropriate considerations would differ from those put forward in the Discussion Draft (for example, burden of proof, motive vs main benefits, etc).  Finally, we note that demands for tax that are not clearly supported by underlying tax law will be problematic for companies to comply with, so it imperative that such laws are clear. Payment of undue taxes could result in such taxes being non-creditable at best, and expose businesses to anti-bribery and corruption concerns, at worst. So, it is imperative that all such issues are considered further. Notwithstanding our general concerns, we have answered the questions posed in the Discussion Draft in Appendix I. Again, we thank you for the opportunity to comment on this Discussion Draft, but encourage you to issue another draft for discussion and look forward to the opportunity to provide additional comments when the Toolkit has been fully commented on and reviewed by the Platform partners. Sincerely, Will Morris, Chair BIAC Tax Committee 6 Appendix I: Responses to Questions 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? As a general matter, we question why the Toolkit should address the rationale for taxing OITs, except to the extent that the policy rationale behind each country’s sovereign decision to tax is relevant to implementation. Notwithstanding this, BIAC does not believe that the Toolkit does effectively address the rationale(s) for taxing OITs of assets. The Discussion Draft assumes that the “source” country has the primary right to tax the gain on the underlying property and does not discuss the rationale for residence based taxation of that property. It misstates the current treaty rule – where tax treaties exist, the country of residence generally has the right to tax all income and gains other than where explicitly provided for in the treaty. We believe that the political economy argument focuses on a few high-profile cases that are not representative of the vast majority of asset transfers, whether direct or indirect. Such might be more appropriately dealt with narrower targeted rules. Additionally, we do not believe that indirect transfer taxation should be promoted as a method to finance public spending and stimulate growth. In countries where capital gains taxation is present, introduced or expanded, it is likely less transfers of ownership will occur regarding prospects, potentially leaving them under-developed (either in efficiency or time). For example, transfers to ensure the most appropriate party is involved in the prospect may not occur where the tax cost of transfer is too high. The capital gains tax will likely also be grossed up in the calculations for the transfer, making the prospects relatively more expensive and equally reducing their ability to be transferred. Overall, investors look to after-tax cash flows, and therefore the tax burden, whilst not in itself conclusive, will always be a factor that impacts investment decisions. Tax systems are often specifically designed to encourage investment and increase employment. As capital is constrained, investors will compare alternatives, typically considering total retained cash. Especially for extractive industries, corporate taxation and capital gains taxation are only part of the overall burden that affects retained cash (e.g. royalties and signing bonuses bring forward the timing of taxation, potentially before any prospect becomes profitable). Having a capital gains tax and/or an indirect capital gains tax can further frontload the timing of taxation, pushing the point of return of investment further back. If a local system is already frontloaded, policy makers need to consider whether to introduce or expand their capital gains taxation in that context. In addition, introducing capital gains taxation ignores the economic double taxation of taxing the future income stream upfront on a disposal, and then again as the new owner realises income and pays tax (and other economic contributions to society) in the country where the asset is located. Indirect capital gains taxes further increase that risk of double taxation. It would therefore seem to be more important for policy makers in developing countries to be fully informed (in a balanced fashion) and made aware of pros and cons of OITs in order to end up with predictable and clear tax systems. Having a tax regime that is based on the consistent and predictable application of principles-based tax rules can help to promote and attract investment. Principles and rules should be transparent, proportionate, administrable, fair, reasonably certain, conducive to timely determination of results and avoid double taxation of profits or non-deduction of costs. Trying to tax transactions ad hoc, in contravention to agreed taxing rights should be discouraged. Clearly 7 presenting pros and cons of capital gains taxation seems like a better approach to allow countries to make an informed decision on how to approach this in a manner that achieves their aims. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? In our view, the two proposals are clear in their general outlines, but as noted below in our response to Question 9 many difficult issues are ignored or treated cursorily. For example, the focus of taxation for OITs should remain centred on immovable property (e.g. mining rights and land) which should capture the assets capable of generating significant location-specific rents in that country, but should not extend to the value created attributable to historic intellectual property and knowhow developed by a group outside of the local country. The Discussion Draft notes on page 21 that “[t]he increased value of the entity sold may reflect in part managerial and other expertise contributed by the seller, beyond what has been recovered in managerial fees, royalties and other explicit payments.” However, the guidance addressing this counterargument is considerably limited. This notion of increased value is particularly relevant for any group that may provide substantial technical and engineering expertise into a project, which is often not available in the local country and this topic (along with the other topics notes in our response to Question 9), appear to be largely glossed over. 3. Is the definition of an offshore indirect transfer of assets satisfactory? We do not believe that the definition of an OIT of assets is satisfactory. The example deals with the simplest of cases and the rules would need to adopt and define many thresholds if these rules are to be practical in application. For example, the definition of an OIT of assets in the Discussion Draft fails to include or even discuss the possibility of exemptions for reorganisations where no economic disposal takes place. The importance of appropriate exemptions in such circumstances cannot be overstated. Special consideration must be given to excluding internal reorganisations where there is no indirect change in control as it is quite common for assets to be indirectly transferred to other parts of a commonly-controlled group in order to meet other corporate objectives and improve coordination and efficiency in the identification and development of local assets. The Draft Discussion touches on the exemptions considered by the OECD MTC such as: “…it could exclude from taxation alienators holding below a certain minimum level of participation in the entity; or the sale of shares of companies listed in an approved stock market, or gains from transfers of shares in a corporate reorganization.” We believe the Toolkit should go further and make clear recommendations on how a country could introduce sensible exemptions into domestic legislation to address such scenarios. For example, we understand that such reorganisation provisions are missing from recently enacted laws in Namibia designed to target OITs where there is an underlying interest in a mineral license or right to mine minerals. By making the need for such provisions explicit in the Toolkit, it is more likely that there will be consistent implementation across taxing jurisdictions. The Toolkit simply fails to address a number of critical issues including how to address listed companies and how to deal with Joint Venture partners of a company transferring its ownership. With regards to listed companies, when countries consider expanding their capital gains taxation to indirect transfers, 8 all share transfers could create concerns. Shares in some companies may change hands frequently so when assessing whether >50% of ownership has changed during any time period, it would create uncertainty on how to consider normal day to day trading of a company’s shares. In the event of a Joint Venture partner transferring its ownership, at what point is a capital gains tax triggered when only one partner transfers, either all or part of its ownership, whether or not in a phased sale? Additionally, who is liable and how would a step up in basis work in a JV scenario? Finally, how do the proposed approaches envisage funding to be made available to a JV? These outstanding questions should clearly support the need for this Discussion Draft to be further refined and released for additional public comment. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? We do not believe the discussion regarding source and residence taxation in this context is balanced or robustly argued, and, as noted above, we are concerned that this question is raised in a Toolkit aimed at assisting with implementation rather than as a discussion between the related sovereign nations. In addition to the points raised in our response to Question 1, this discussion is lacking in providing developing countries with the proper guidance and tools to address double taxation where the possibility of both source and residence taxation exists. A discussion of this topic is of little value without practical recommendations for arriving at a solution. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? BIAC does not believe that the suggested possible expansion of the definition of immovable property is reasonable. The Discussion Draft abandons the treaty definition of immovable property and advocates an expansive definition of immovable property, which the draft itself acknowledges would be difficult to capture in legislative language. This is a prescription for uncertainty and double taxation. The definition of immovable property is a critical issue and the policy objective for a country wishing to ensure a taxing right over gains realised on finite mineral resources sourced in that country is understandable (notwithstanding the view that this value is better taxed through levies on output as it is extracted, than through a capital gains tax). However, we believe strongly that the proportion of the gain realised by a non-resident upon which a given country is permitted to tax, is restricted only to that portion related to immovable property in that given country. We would not consider telecommunications and broadcast spectrum / networks assets to be immovable property in this sense. We also note that the telecommunications and extractives industries are already subject to a wide range of taxes, including industry specific taxes in many cases designed to meet local policy objectives. Taxing the immovable property of an extractive should capture most, if not all, of the assets within a mining operation that generate significant location-specific rents in that country. An expansion of immovable property to target other assets could risk overstepping the reach of the source country by incorrectly permitting the taxation of value that was generated in the non-resident’s country, or a third country where the group may have investments. Through attaching the non-resident capital gain to the value of appropriately defined immovable property, this ensures that the non-resident capital gain is limited to the amount which is clearly local 9 country sourced. It is often the case that reliance on double taxation treaties is required to arrive at this position, as often the source country will already try to assert a taxing right over a whole gain made by a non-resident on an OIT where only a percentage of that gain is attributable to underlying immovable property in the source country. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? We do not find the concept of location-specific rents helpful in addressing these issues. As the Discussion Draft acknowledges, access to a local market could be considered to generate location specific rents. A concept that is intended to be interpreted expansively and that is poorly defined will be interpreted in ways that will reduce certainty and deter investment. 7. Are there other implementation approaches that should be considered? We have no specific comments at this time related to alternative implementation approaches as we do not believe the Discussion Draft has adequately addressed the underlying principles that should be first considered by countries in deciding whether to tax OITs (if, indeed, that is deemed an appropriate starting point for such a Toolkit). We would like to take this opportunity, however, to stress that a revised Discussion Draft should make it clear that proposed legislation be applied only on a forward- looking basis. The commercial projects at issue are often very long-term in nature and are entered into a view of the tax treatment over the entire life of a project. Therefore, we strongly believe that new domestic provisions should not have retroactive or retrospective effect, and this could be achieved through the inclusion of grandfathering provisions to support certainty and stability on tax treatment of existing long-term projects (including those critical for the economy such as natural resource exploitation and infrastructure). 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? Before discussing the two proposed Models, the Discussion Draft should deal with considerations of symmetry and neutrality in a more broad sense. Such aspects help to address double taxation concerns in many capital gains tax systems and should be made available in the Toolkit. Whilst the Toolkit recognises that transfers of assets – either directly or indirectly – can generate capital gains as well as capital losses, none of the options deal with what to do when indirect transfers/offshore indirect transfers result in a loss. The summary several times emphasises the need for neutrality between direct and indirect transfers of capital gains. Various capital gains tax systems actually exist that provide tax neutrality for share transactions, provided the profit generating assets remain in the country. Although the Toolkit supports tax neutrality in principle, it does not provide enough detail on such or other approaches that actually induce tax neutrality. Symmetry in tax treatment is also not developed in the Toolkit. Generally, tax policy that taxes capital gains allows deductibility of losses. If the Platform intends to make the case that where capital gains are taxed, indirect capital gains should also be taxed, then it must at least detail how to deduct indirect capital losses to limit the taxation to double taxation rather than multiple taxation. We do not believe the Toolkit, as currently drafted, provides the detail and analysis necessary to support a preference for either Model 1 or Model 2. We look forward to an additional opportunity for 10 public comment to address this question following refinement of the Discussion Draft. At a high-level, however, we note several specific issues that would need to be addressed regarding Model 1 (taxation of a deemed direct sale by a resident) and Model 2 (taxation of the non-resident seller):  Under Model 1, the tax charge is levied upon the underlying in-country investment being purchased. This runs counter to the principal that a capital gain tax should be levied upon the person making that capital gain (i.e., the investor making the disposal). It seems logical that the tax liability on a capital gain is directly suffered by the entity making a disposal and receiving proceeds, which it can use to settle that liability.  Model 2 appears to better address (albeit does not eliminate) the risk of double taxation by attaching the capital gains tax on OITs to the non-resident person making the disposal. Indeed, the Discussion Draft appears to acknowledge this deficiency with Option 1, but provides no reasonable solution.  Under Model 1, it is not clear that an investor would receive any protection under a double tax treaty. Investors should be able to rely upon treaties to ensure that they are not taxed in both states on the same gain. We do not see any economic difference between double taxation of the same legal entity, and double taxation of the same gain in two separate legal entities.  The calculation under the ‘deemed disposal’ method of Model 1 is fairly untargeted, and is calculated upon all assets of the local company. This may allow tax authorities to inappropriately tax value changes on company assets which fall far outside the traditional definition of immovable property. For example, for an extractive asset this could include stockpiles of extracted resources included within inventory, mining equipment, or even non- mining assets.  Model 1 and Model 2 both present considerable issues related to determining the value of the actual gain. The Toolkit seems to correctly imply that the gain should be considered (rather than the proceeds). However, further clarification is required on what assets the gains taxation should apply to, and how they are to be allocated and valued. It is important here to have generally accepted source material. For example, the Discussion Draft references a study by Beer and Loeprick but the approach to that study is problematic, leading to results that are not consistent with the experience of multinational oil and gas companies and valuators. 9. Are the complexities in the taxation of these international transactions adequately represented? BIAC does not believe the Discussion Draft adequately represents the complexities in the taxation of these international transactions. The simplified example that forms the basis of the analysis contained in the Discussion Draft ignores the complexities involved in determining whether the transaction should be subject to tax. The Discussion Draft also ignores or provides limited commentary on critical topics such as the difficulties dealing with minority shareholders, valuation issues, foreign exchange, the tax base, the treatment of losses, and how economic multiple-taxation should be avoided. 11 The BEPS Monitoring Group Comments on the Public Discussion Draft on THE TAXATION OF OFFSHORE INDIRECT TRANSFERS – A TOOLKIT These comments have been prepared by the BEPS Monitoring Group (BMG). The BMG is a network of experts on various aspects of international tax, set up by a number of civil society organizations which research and campaign for tax justice including the Global Alliance for Tax Justice, Red de Justicia Fiscal de America Latina y el Caribe, Tax Justice Network, Christian Aid, Action Aid, Oxfam, and Tax Research UK. These comments have not been approved in advance by these organizations, which do not necessarily accept every detail or specific point made here, but they support the work of the BMG and endorse its general perspectives. They have been drafted by Jeffery Kadet and Sol Picciotto, with contributions and comments from Tommaso Faccio and Pooja Rangaprasad. We appreciate the opportunity to provide these comments, and are happy for them to be published. October 2017 SUMMARY We welcome this discussion draft, which deals with an important issue of particular interest to developing countries, and was only partly dealt with in the G20/OECD project on base erosion and profit shifting (BEPS). We agree with the argument it makes that principles of inter-nation equity clearly support the right of the country where an asset is located to tax the gains on its transfer, even if the seller and/or acquirer are not resident in that country. The country is of course free to decide whether and at what rate to tax such gains, taking account of the effects of such taxation on investment in the development of such assets. This right should therefore not be restricted by tax treaties, and we support the proposals in the BEPS project for inclusion in all treaties of a provision equivalent to article 13(4) of the model treaties. This can most effectively be done if all countries sign the Multilateral Convention on BEPS and adopt its article 9(4). This Toolkit should be amended to clearly and unambiguously urge all countries to do so. In our view, the proposals should extend to indirect transfers of all kinds of assets, without limitation to immoveable assets. This is in accordance with the global consensus that profits and gains should be taxed in the jurisdiction where the economic activities giving rise to them are 12 located. The reference to article 13(5) of the UN model in the DD is therefore misleading, and should be amended, to provide countries that choose to tax a wider range of gains the necessary guidance to address movable assets such as shares. We make a number of other comments which we hope would help improve the DD. A. GENERAL REMARKS 1. Background and Principles We applaud the effort and thought that went into this Discussion Draft (DD) of a toolkit for the taxation of offshore indirect transfers (OITs). This is an important area not covered by the G20/OECD Base Erosion and Profit Shifting (BEPS) project that has simply cried out for the attention that this DD is now giving it. In particular, we agree with the DD that offshore indirect transfers (OITs) are a significant issue for many developing countries. OITs are also a significant issue for many developed countries as well. Some specific country actions taken over a number of years are evidence of this, including the 1980 enactment of the U.S. Foreign Investment in Real Property Tax Act. Hence, this toolkit once finalized will be an important aid and resource for all countries. We agree with the analysis in section B of the DD (p.18) that principles of inter-nation equity clearly support the right of the country where an asset is located to tax the gains on its transfer, even if the seller and/or acquirer are not resident in that country. The country is of course free to decide whether and at what rate to tax such gains, taking account of the effects of such taxation on investment in the development of such assets. It is therefore inappropriate that this right should be constrained by tax treaties, especially as they were generally executed when parties to their negotiation had little appreciation of what practical taxing rights they were giving up. It should therefore be a high priority to remove tax treaty provisions that constrain this right. A major step in this direction would be adoption of a provision equivalent to article 13(4) of the model conventions in all treaties. This was agreed as part of the G20/OECD project on Base Erosion and Profit Shifting, and is proposed in article 9(4) of the Multilateral Convention on BEPS. We therefore urge all countries to sign the MC- BEPS and accept this article. We are disappointed that of the 71 countries which have so far signed the MC-BEPS, 37 have made reservations against article 9(4). Unless these reservations are withdrawn, this important anti-BEPS measure, which is particularly important for developing countries, would be implemented only partially. The Toolkit acknowledges that some countries seek to tax some moveable asset transfers. In our view this is justified, and should be achieved by inclusion in all treaties of a provision based on article 13(5) of the UN model convention. This inclusion in all treaties would allow countries that currently do not tax moveable transfers to be free to do so in the future. We strongly suggest that the toolkit include in discussion and within the Conclusions section three clear recommendations. • That all countries signing the MC-BEPS should accept its article 9(4) for all their covered treaties; 13 • That all countries should renegotiate their non-covered treaties to include article 13(4) of the UN model; and • That all countries should renegotiate all their existing treaties to include article 13(5) of the UN model. Within this comment letter, we provide our thoughts on how this toolkit may be made even better. 2. Overall Conclusion on Model We agree with the conclusion that Model 1 is the better of the two approaches discussed, treating the transfer as a deemed disposal by the local entity that directly owns the asset in question, even though it takes the legal form of an offshore transfer of shares or other direct or indirect ownership interest by a non-resident. The fact that existing domestic rules relevant to tax residents apply and its practicality of enforcement make it a usable tool for any country, whether developed or developing. We applaud this clearly stated and sensible conclusion. 3. Reaction to Taxpayer Concerns and Complaints We can imagine that there may be some number of comments on this DD that represent taxpayer complaints that the DD’s recommendations and/or the variety of approaches applied by different countries will complicate their lives and increase uncertainty. Such comments should be summarily ignored as disingenuous. Most typically, taxpayer efforts to achieve non-taxable offshore indirect transfers involve careful planning and structuring specifically meant to overcome appropriate and legitimate local taxation on realized economic gains. Such structuring seldom would be conducted in the absence of the anticipated tax reductions. With this in mind, we believe that no concern should be given to the risk, described on page 23, of “amplifying the uncertainty that taxpayers face in arranging their affairs.” When taxpayers do not intentionally try to sidestep legitimate local taxation on realized gains, their outcome will normally be very certain. B. SPECIFIC COMMENTS 1. Need for Expanded Coverage beyond only Immoveable Property All of the language and discussion in the DD in relation to both Model 1 and 2 assume that a country will only want to cover OITs that involve an indirect interest in immoveable property, although the discussion of the definition of ‘immoveable property’ includes both a minimal and an extended definition. In our view, the same considerations apply to indirect transfers of assets which may be considered moveable property. A number of countries do tax transfers of interests in resident companies or partnerships and would want to include language to include OITs that are indirect transfers of such assets. This is recognized in the DD, including in the Conclusions on page 58 where it is acknowledged that some countries tax ‘intangibles such as corporate stock issued in regard to a domestic company but held by a non-tax resident’. We strongly suggest that additions be made to Model 1 and 2 so that countries desiring such broader coverage will be properly served by this new Toolkit. In our view, it should clearly recommend that countries desiring such broader coverage should renegotiate their treaties to 14 include a provision based on Article 13(5) of the UN model, which allows the source country to tax indirect transfers of a substantial shareholding in any company resident in that state. 2. Avoid References to a ‘Uniform Approach’ The Executive Summary states: There is a need for a more uniform approach to the taxation of OITs. Countries’ unilateral responses have differed widely, in terms of both which assets are covered and the legal approach taken. Greater coherence could help secure tax revenue and enhance tax certainty. The Conclusions on page 58 also expresses concern about ‘uncoordinated measures that jeopardize the smooth and consensual functioning of the international tax system’. We agree that a more uniform approach that discourages new loopholes that taxpayers can exploit could well be helpful. However, considering both the sovereignty of countries and their varying conditions and concerns, we believe that expressing in the Executive Summary and Conclusions this ‘more uniform approach’ goal, which in the end may not be achieved, is misleading to readers on what they will find within the DD. In our view, it is the final paragraph in the Executive Summary that does briefly describe the real achievement of the DD, which is several workable best-practices options that countries may consider in light of their particular circumstances and needs. We believe that there is no need to refer to any unattainable ‘uniform approach”. In this regard, in addition to the final paragraph in the Executive Summary, we note that the DD comments on page 10: … [The toolkit] does not set out a single, definitive approach suitable in all circumstances. The aim rather is to identify practicable options, with a particular view to the circumstances of developing countries. It does, however, make some tentative recommendations. The DD does set out two concrete Models and a definition of immoveable property. It provides a number of options that countries may consider. This is laudable and should be described as such in the Executive Summary and Conclusions, which unfortunately are probably as much as the majority of readers will read in this understandably long and excellent document. Hence, we believe that the Executive Summary and Conclusions should not make mention of a ‘uniform approach’ goal that may not be reached. 3. Addition to Purchaser Tax Consequences On page 14, various purchaser tax consequences are noted. We suggest that the following be added at the end of footnote 12: Indeed, it will often happen in the case of indirect transfers of appreciated depreciable property that the purchase price will reflect an economic discount related to the lost future depreciation deductions since the tax basis of the indirectly acquired asset will not be stepped-up to the purchase price actually being paid by the indirect acquirer. 4. Misleading Revenue Effects 15 The discussion at the top of page 16 on the revenue impact on capital gains from basis adjustments seems to us very misleading. We believe that it would cause a typical reader to ask, why do we bother to impose any tax on capital gains? The discussion states, in part: … the total nominal (undiscounted) revenue raised from the capital gains tax over time will be zero: that is, the same as if there had been no sale, or no capital gains tax. … The reason for this “zero” result is the basis adjustment in the asset that is equal to the price paid for the asset. This basis adjustment will then mean an offset in determining gain on a future sale. We of course agree with the basic accounting and tax computations that allow a basis increase for any purchase price paid. However, this ignores the actual tax that is collected from gains on assets that appreciate. Take as a simple example a piece of raw land. (We use raw land to avoid complicating the discussion with depreciation.) Owner A, who acquired it for 100, sells it to Owner B for 150, who in turn sells it to Owner C for 300. There is cumulatively 200 of gain (300 minus 100). If this is an asset that by its nature will not likely decrease in value (raw land being in short supply) and is not depreciable, then there is little chance that there will be any reversal of this 200 gain. The basis adjustment will of course prevent this 200 gain from being taxed a second time, but it will not reverse the tax collected on that 200 gain. With the above in mind, we suggest that the discussion of the revenue impact should be amended to indicate these real revenue effects. 5. Gains as Reflecting Accumulated Undistributed and Future After-Tax Earnings On page 18, the DD notes the counterargument that gains may reflect earnings that the location country has either chosen not to tax or that are, as yet, unrealised gains and will not be taxed until some realisation event occurs. The discussion includes the sentence: The gain, that is, reflects earnings that the location country has in a sense simply chosen not to tax. While we understand what this sentence is meant to convey, we believe that it will be too simplified for many readers. We suggest that it be changed to: The gain, that is, reflects economic earnings that the location country has either chosen not to tax or the events that would result in a taxable event have not yet occurred. 6. Need for Additional Balance in Discussion Throughout the DD, there are many references to natural resources and sometimes telecommunication license rights as examples in discussions, but few references to more traditional real property assets. While real property is of course included within the discussion starting on page 55 where defining immovable property is covered, it seems to us that more references to real estate generally within the DD’s discussion would provide readers a more complete and balanced understanding. 7. Matters Concerning Article 13.5 of the U.N. Model Tax Convention 16 On page 25 in the Assessment section, the following is stated: … It would seem that Article 13(5) [of the U.N. Model Tax Convention] is generally not needed as long as the definition of “immovable property” in both any applicable treaty under Article 13(4), and especially in domestic law, is sufficiently broad. This sentence appears incorrect and misleading to us. Article 13(5) allows location country taxation on gains from any alienation of shares, other than those covered by Article 13(4), where the seller holds directly or indirectly some minimum percentage ownership in the capital of the company. The focus of Article 13(5) is on the level of ownership regardless of what assets the company might own and the activities it might conduct. Such a company, of course, may own little or no immovable property and may conduct significant and varied businesses and investments both within and outside the location country. Even with the broadest of definitions of immoveable property, there will often be no taxation under Article 13(4) for situations where Article 13(5) would apply. Therefore, to suggest that Article 13(5) is not needed if there is a sufficiently broad definition of immovable property is simply incorrect. We therefore strongly suggest that this sentence be deleted from the DD. 8. Concern with Taxpayer Burden On page 52 is the following in regard to a withholding mechanism: … As noted, the withholding tax can only be collected as an estimate of the seller’s final income tax liability (as the actual quantum of the seller’s gain is unlikely to be known by the purchaser) and so withholding necessarily increases the compliance burden for the purchaser (who is subject to the withholding obligation) and the seller (who needs to file a tax return and determine any outstanding balancing amount or refund after claiming a credit for the amount of the tax withheld)—although this burden could be manageable. … [Emphasis added.] The tenor of this should definitely be changed. OITs are a real problem for many countries. Taxpayers who enter into transactions to sell or buy properties in an indirect manner often do so with an intention to lower their tax obligations (and perhaps other costs as well, e.g. real estate transfer taxes and other costs), and do so with full knowledge of their intention and their specific structuring. To include this kind of language, “although this burden could be manageable”, is making an excuse for something that absolutely needs no excuse. We suggest the following language as an alternative. … As noted, the withholding tax can only be collected as an estimate of the seller’s final income tax liability (as the actual quantum of the seller’s gain is unlikely to be known by the purchaser). While any imposition of a taxation approach that includes a withholding tax will create a compliance obligation for the purchaser (who is subject to the withholding obligation) and the seller (who needs to file a tax return and determine any outstanding balancing amount or refund after claiming a credit for the amount of the tax withheld), this is not an added burden that merits concern. This reflects the fact that the seller and buyer are the parties that intentionally structured their indirect OIT and typically have done so with the primary intention of avoiding income tax obligations 17 and/or other transaction costs, e.g. real estate transfer taxes, need to apply for new licenses, etc. … We agree with the content of the remainder of this paragraph on page 52 regarding a ‘prudent third party purchaser’. However, it will be recognized that some reasonable percentage of third party purchasers are either less than fully prudent or are actively complicit in attempting to structure a transaction that will avoid tax and other costs on the seller, thereby sharing in the seller’s savings through a lower purchase price. Again, for this reason, the tenor of this paragraph must reflect the reasonableness of withholding and return filing obligations and not make excuses for them. 9. Concern with Potential Double Taxation In various places within the DD there is mention of how an advantage is that one Model or the other will avoid double taxation in particular situations. See for example the bullet points on this on pages 47 and 55. While we do not dispute that avoiding double taxation is desirable, we believe that the DD should be less concerned regarding any potential double taxation risks. This is because OIT structures have normally been specifically created to avoid any taxation. Often, such structures have been set up specifically as an exit strategy when an investment was first made with a goal of avoiding tax on a contemplated future disposition. One has only to look at some of the structures used in publicly disclosed transactions involving countries such as India and China to see the truth and reality of this. Taxpayers, especially those seeking double-non- taxation, should not receive sympathy when their structures backfire and they end up with some amount of unrelieved double taxation. 10. Concern with Certain Wording Used in the Illustrative Cases It seems inappropriate and belittling to developing countries a) to state on page 28 that in all three cases, the country in which the underlying asset was located lost in court, especially given that in the Uganda – Zain Case, this is not actually the case; and b) to suggest on page 29 that countries responding to defeat in court by quite sweeping policy changes result in more incoherence and uncertainty in international taxation than already exists, for no apparent gain (emphasis added). We have already commented earlier herein on the issue of uncertainty. This reference to ‘incoherence and uncertainty’ should be deleted. We recognise that the phrase “no apparent gain” is simply referring to the fact that the three governments concerned might not yet have received any additional taxes from their efforts. Irrespective of this, these cases represent material issues to the countries involved (as clearly outlined in the previous paragraphs: 5% of total government revenue in the Zain case and 2% of central government revenue in the Vodafone case). More importantly though, as future events unfold, both Uganda and India through their future actions may realize significant sums in regard to these cases. In any case, this paragraph on page 29 should be rewritten to avoid the implication it now provides to readers that these efforts are damaging, and in the end, not worthwhile. At a minimum, the last sentence on page 29 should be deleted. We may add that if criticism is to be made, it should also be directed at the companies that resort to protracted legal and political campaigns, including resorting to external private arbitration, to 18 prevent a state from exercising its legitimate right to tax gains from assets deriving their value from within its territory. 11. Concern with the Separate Legal Entity Distinction From page 48: This approach undermines the separate legal entity distinction between the local asset holding entity and its relevant tiers of parent entities. [See also first bullet point on page 55.] That this concern is mentioned is understandable since much of international taxation is, regrettably in our view, based on the separate entity principle. However, as this principle is also the basis for many BEPS structures and at the heart of each structure specifically created to avoid tax on OITs, this concern should be summarily dismissed. 12. Need to Add Consideration of Non-Income Taxes, Fees, and Other Costs The focus here is on immoveable property. Some countries will have national or local transfer costs (e.g. transfer taxes, registration fees, etc.) that will apply to actual transfers of some immoveable property. As some such costs may be based on the actual transfer price, the amounts can be significant and may encourage OITs even where the OIT is covered by an income tax charge. The DD should make clear that the Model 1 ‘Change in Control’ provision described in Box 4 on page 44 should be enacted in a manner that will make it effective not only for a country’s income tax but also for all transfer taxes and other costs and fees that accompany any transfer of immoveable property. Where Model 2 is applied, then it seems clear that there would be no ability for a national or local government to collect these transfer taxes and other costs and fees. As this may be a very important loss to national and local revenues, this should be included as a disadvantage of Model 2 on page 55. 13. Addition Needed to Reflect Statute of Limitation Concerns For both Model 1 and Model 2, there should be discussion in the DD noting that amendments should be made to appropriate statute of limitation rules providing that any normally applicable statute of limitations will not start to run until after notification and filing of applicable information and tax forms required by any change in control. 14. Additional Mechanisms to Highlight OITs A suggestion that could be more strongly made in the DD is that registers of real property, natural resources, and other assets should be expanded to include not only the title owner of the applicable property, but also all higher tier owners that have indirect interests in the property. The required updating for any changes in indirect ownership would be an additional mechanism to alert applicable tax authorities of possible OIT taxable events. We suggest expanding the second paragraph in the ‘Enforcement/collection rules’ section on page 51 of the DD. Presently, through the parenthetical, the focus is on extractive licenses rather than on real property more generally. 19 15. Guidance on Domestic Law Anti-Avoidance and Treaty Override Tailored to OITs Taxpayers that invest in some business or asset typically consider their exit strategy at the time of the initial investment. Such taxpayers structure their affairs in advance, often using multiple legal entities not for commercial or legal concerns, but rather with some tax motivation in mind. Importantly, they have the advantage of structuring whatever will arguably avoid tax obligations and reporting under whatever objective local country rules are in place. As a result of this, there is a very “unlevel playing field”. This “unlevel” situation means that the only truly usable tools will often be the principal purpose test where a treaty applies and domestic law anti-avoidance rules that grant reasonable discretion to tax authorities. On page 55 is the following: Even with appropriate domestic legislation, under this model the taxing right of the location Country L could (unless there was a treaty override) still be limited by an applicable tax treaty, if the relevant treaty does not include an article similar to Article 13(4) of the OECD or UN Model MTC. [Emphasis added.] What is truly needed for developing countries is specific guidance for amending their domestic tax rules so that OIT transactions are directly addressed, whether through appropriate treaty overrides or through other anti-avoidance rules. Such rules should make taxable any OIT that is not caught by the domestic objective rules (whether Model 1 or Model 2) and is not legitimately covered by a tax treaty provision. 16. Amendment of Appendix B On page 61 is the following: For example, the payment for the sale of an asset could be timed to occur after the entity engaged in the U.S. trade or business had been liquidated, so that the capital gain would be realized when the foreign resident had no U.S. business to be connected to. We suggest that this sentence be deleted. See §§864(c)(6) and (c)(7) of the Internal Revenue Code. 17. A Further Minor Item Within the Executive Summary, the first sentence in the last paragraph should read: “The report outlines two main approaches for enforcing the taxation of OITs by the country in which the asset is located—provisions for which careful drafting is required.” C. RESPONSES TO SPECIFIC QUESTIONS RAISED The announcement of the DD set out the following nine questions. Our above comments cover many of these questions in some detail. We have only added additional responses below where we have something additional to say. 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? Response: Yes. 20 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? Response: Yes. 3. Is the definition of an offshore indirect transfer of assets satisfactory? Response: Yes. The broad definition is excellent. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? Response: Yes. We approve particularly of the clear statement that neutrality between direct and indirect transfers is important. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? Response: Yes. It is both reasonable and appropriate. See also section B.1 above stating that the toolkit should also provide for countries that tax transfers of moveable assets such as interests in resident companies or partnerships. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? Response: Yes. 7. Are there other implementation approaches that should be considered? Response: Yes. See section B above. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? Response: Yes. See section A.1. above. 9. Are the complexities in the taxation of these international transactions adequately represented? Response: For the most part, yes. See various suggestions in section B above. 21 20 October 2017 CBI RESPONSE TO PLATFORM FOR COLLABORATION ON TAX: A DISCUSSION DRAFT - THE TAXATION OF OFFSHORE INDIRECT TRANSFERS – A TOOLKIT Background As the UK’s leading business organisation, the CBI speaks for some 190,000 businesses that together employ around a third of the private sector workforce, covering the full spectrum of business interests both by sector and by size. Thank you for the opportunity to respond on the toolkit for the taxation of offshore indirect transfers (the “Toolkit”) that has been developed by the Platform for Collaboration on Tax (the “Platform”), comprised of the IMF, OECD, UN and World Bank. Our response below builds on the initial comments made on 13 September and we appreciate the additional time that has been given to allow us to provide you with a fuller response. We will be happy to provide additional information on any points covered in either of our responses, if that would be helpful. In our view, relatively substantial further work is required before the Toolkit can be finalised. We think it would be helpful to bring together taxpayers, tax authorities and representatives from the Platform to discuss businesses’ concerns and potential solutions before the Toolkit is finalised. We also strongly recommend that the Toolkit is released again as a revised draft for public comment once it has had further work. Key messages Our key comments on the Toolkit are as follows: • We are strongly in support of the overall goal set by the G20 to improve tax certainty. We also recognise that increased clarity on the treatment of Offshore Indirect Transfers (“OITs”) is one way in which tax certainty can be improved for tax payers. This could ultimately protect and enhance investment into developing countries. • Accordingly, we welcome the intention of the Platform to coordinate efforts to bring more clarity to the treatment of OITs at both the domestic law and treaty levels. • This said, we are concerned that the Toolkit appears to make a recommendation for taxation of OITs rather than providing countries with a detailed guide on whether to tax such transfers and, if so, how. Such information could help countries to make an informed decision on whether taxation of OITs would be a sustainable contribution to their broader tax policy. Any revision to the split of taxing rights should be debated and agreed between countries. Inclusion of a recommendation to tax OITs in guidance without such debate and agreement between countries may lead to inconsistent use of the guidance alongside existing law and treaty obligations, thus increasing uncertainty for tax payers. • We strongly recommend that the Toolkit is re-focussed as a more balanced summary of the pros and cons of taxation of OITs, with guidance on detailed considerations for appropriate tax rules if countries wish to pursue that route. • We recommend that more commentary is included on the fact that most OIT’s are normal commercial transactions undertaken at the level they are for good commercial reasons and are not motivated by aggressive tax avoidance. The commentary should also consider further how to minimise complexity on the tax treatment of these bona fide commercial transactions. For instance, the Toolkit may wish to align with other current developments in the international tax sphere by focussing on aggressive tax 22 planning (in this case in the form of structuring of OITs) that purposefully avoids legitimate tax take. • We strongly recommend that new domestic provisions should not have retroactive effect and we would welcome the Toolkit highlighting that domestic provisions could include grandfathering provisions to support certainty and stability on tax treatment of long term existing projects. • In our view, it is too early to expand the definition of “immoveable property” to a much broader range of assets, e.g. “other rights issued by governments”. The Toolkit itself acknowledges that further work would be required before a clear definition of these broader assets can be reflected in statute. In view of the focus on increasing uniformity of treatment of OITs, we urge the Platform to recommend that the definition of immoveable property is not considerably broadened within domestic provisions at this time. • We are concerned that a number of the most complex areas of taxation of OITs are not covered in much detail in the current draft Toolkit. We recommend that more work is undertaken on mitigating double (or multiple) taxation, developing guidance to assist with the subjective exercise of valuations and value attribution, and minimising the effect on ordinary commercial transactions (e.g. through developing suitable exemptions and de minimis thresholds, and providing exemptions for transactions not motivated by aggressive tax avoidance). We strongly recommend that this work is undertaken and consensus is built on the preferred option before the final Toolkit is released. • When further elaborating on the principles above, we would prefer to see a focus on Model 2 as approaches for domestic legislation to tax OITs. Model 2 is better aligned with established international tax practices, is aligned with the legal form of the transaction and better manages some of the risks of double taxation. We recommend that further work is undertaken to set out how governments might develop robust tax collection and enforcement procedures. • Anti-avoidance legislation is an alternative approach which may be worth pursuing further. This may offer asset-owning countries the opportunity to introduce legislation to tax OITs but only where they consider tax avoidance practices are being undertaken. Further work should be undertaken to support developing countries in drafting suitable domestic provisions which are not burdensome to administer. These points are covered in more detail in the remainder of this paper. General remarks In this section, we have outlined a number of general remarks on the Toolkit as a whole. In the following section, we respond to the specific questions on which you have asked for feedback. Clarity on objectives There is a general concern on the objectives the Platform aims to address with the Toolkit. As the various organisations comprising the Platform have different objectives and deliverables, further clarification of aims and objectives would be helpful. This would especially help to mitigate concerns we have regarding the apparent recommendations within the current draft for countries to tax OITs. This would seem to go beyond the approach that some of the participating organisations generally take. As described, we think that the potential subjectivity on the legal standing of such recommendations within guidance can create confusion and we would want to ensure that any new allocation of taxing rights has been appropriately debated and agreed between countries and subsequently included within domestic legislation and international tax treaties. Commercial considerations of OITs As stated in OECD’s press release of 1 August, this Toolkit has been designed “to help developing countries to tackle the complexities of taxing offshore indirect transfers of assets, a practice by which some multinational corporations try to minimise their tax liability.” The CBI supports efforts to tackle aggressive tax avoidance and evasion: for instance, we have actively participated throughout the OECD’s project to address Base Erosion and Profit Shifting. 23 It is important to acknowledge that there are many businesses that do not undertake aggressive tax avoidance practices but will still need to consider tax measures being introduced in relation to OITs. In our experience, it is common for anti-avoidance measures to impose disproportionate compliance burdens, and sometimes unfair tax burdens, on businesses and transaction which are not within the intended scope of the measure. We believe it would be helpful for the normal commercial practices undertaken by the majority of businesses to be given more attention throughout the Toolkit and recognition given to the fact that most OITs are undertaken for sound commercial reasons rather than tax-avoidance motives. With more of the commercial realities captured in the Toolkit it should ensure that the rules are clear enough to allow businesses to be compliant with any new legislation put in place. On page 21 and 22 it is stated that “A general principle of good tax design is that the tax system should, so far as is practicable, not distort investors’ decisions: unless there is good reason to do so, taxation should not lead businesses to change their commercial decisions.” This is helpful wording, but is perhaps the only reference in the Toolkit to the fact that commercial drivers will outweigh tax drivers in the majority of cases. We recommend that more commentary on the commercial considerations that affect businesses undertaking OITs is included in the Toolkit. For instance, the second paragraph on page 15 could acknowledge commercial reasons why businesses chose to structure a transaction as a sale of shares rather than an asset sale. For example, in the context of mining, a sale of shares preserves the corporate owner of mining rights, licences, land leases, output contracts, etc. This leads to a much more straight-forward sales process than an asset sale which would require multiple assignment or novation of contractual obligations. It is, therefore, often the most commercially viable route to follow. In other cases, a transaction may be undertaken to allow a joint venture partner to hold an interest in the business/asset concerned by acquiring shares in an upstream company. Shared interest would not be possible or practical via a direct ownership of the asset itself. We are concerned by language in the Toolkit which suggests businesses can simply structure around tax complexities or less desirable tax outcomes, e.g. on page 31 where it is stated that “double taxation may occur, though tax payers would presumably avoid structuring transactions in ways subject to such treatment. Such structures would make use of countries that, while having the right to tax, grant an exemption to the transaction in question – which itself could result in non-taxation.” This is not always possible in many cases, for instance, where joint venture partners are involved. We consider that an important constraint on any proposed measure should be that it should not lead to double, or even multiple, taxation. We have provided further comments below regarding helping businesses that are undertaking ordinary commercial transactions to manage the impact of new tax measures aimed at tackling avoidance. By targeting any new measures on aggressive avoidance motivations, this not only allows the target transactions to be better defined but also the controlling tax payers to be identified. Uniformity but above all clarity of treatment The Toolkit acknowledges that there is currently a wide range of approaches adopted by countries unilaterally and that a more uniform approach to the taxation of OITs (amongst countries seeking to tax such transactions) should be sought. We welcome these efforts and agree that if achieved it should help countries to secure tax revenue, in cases of aggressive tax avoidance, and deliver enhanced tax certainty for both tax administrations and tax payers. The unification of the UN and OECD models as part of this effort and coordination between international organisations regarding efforts in this respect are welcomed. Our members who currently deal with OITs find that existing treaties are often ambiguous and there is wide variation in how they are applied. Updating these treaties would increase the likelihood that they are respected by tax administrations and will help to increase certainty for tax payers. As a broader point, we wonder whether it would be beneficial to use a multilateral instrument ( “MLI”) to update these treaties, building on the experience with the existing OECD MLI. We think this topic may be worth further debate. 24 Timing of application of new rules Some of the examples that are cited of countries bringing in new legislation to protect taxing rights in respect of OITs, include the desire to make these new rules retroactive in their application. To help manage tax certainty for tax payers, we strongly recommend that the Toolkit advocates that new legislation should only apply on a forward-looking basis. Businesses that operate in extractive industries, for instance, have also highlighted that the projects they are involved in are often very long term in nature. These projects will have been entered into with a view of the tax treatment over the full life of the project. In light of this, efforts to establish as much long-term certainty and stability as possible on taxation of OITs is appreciated. With respect to existing projects, it would also be welcomed if the Toolkit could refer to the possibility of some form of grandfathering rule within domestic legislation for individual countries to consider. Building consensus Our members are concerned that some of the most complex aspects of taxation of OITs are not sufficiently addressed in the Toolkit as currently drafted. We have expanded on these points below and strongly recommend that Toolkit is updated to include further discussion on these aspects. Before there is uniformity in treatment, domestic treatment actually has to be clear to improve tax certainty. Many basic issues relevant to providing such clarity are not sufficiently elaborated on, e.g. what the capital gain consists of, how to deal with minority shareholders or other joint venture partners, etc. Without greater consensus on the types of transfers and assets to be taxed by the territory in which the asset is located, we are concerned that the Toolkit may not be successful in achieving the objective of greater uniformity of taxation of OITs by countries (where they have chosen to tax such transactions). In fact, there is a risk that some of the topics left open-ended (e.g. how broad the definition of “immoveable property” should be) could greatly increase the variation in approach within domestic provisions and accordingly the amount of uncertainty for tax payers. We think the Platform has an important role to play in building more consensus on the details of a uniform approach for those countries seeking to tax OITs. Accordingly, the aim to release this Toolkit as a final document by the end of 2017 seems ambitious and we recommend further work being undertaken on the areas we have highlighted below before the Toolkit is finalised. Responses to specific questions posed 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? The Toolkit fails to examine why many developed countries (particularly those with a participation exemption) do not tax either direct or indirect transfers. More discussion is needed about the economic inefficiency which could be caused by taxing OITs: for example, this could inhibit sales by an inefficient investor (who prefers to continue to realise a lower rate of return) rather than sell the asset to a new owner (who could generate higher returns) at the price of an immediate tax charge. This could lead to some assets becoming “zombie” assets, with a reduction in potential economic growth for the country of source. As a specific example from the Oil and Gas industry, Production Sharing Agreements ( “PSAs”) are entered into with the government’s share value agreed at the outset when the investor undertakes the significant expenditure and risk of developing a project. If a tax on OITs were brought in, this could adversely impact the economics of current activities through uncertain future costs thus potentially deferring investment in existing and / or new projects. Furthermore, the Toolkit notes on page 36 that only about 35% of double tax treaties include Article 13.456. Given that these double tax treaties are bilaterally negotiated, it may be helpful to explore some of the reasons why these articles are not agreed upon, e.g. in recent treaty negotiations. 25 We also note that the frustration felt by developing countries over their ability to tax OITs is exacerbated where there are limited taxing rights over the other sources of income being derived from the immoveable property. Developing countries should be supported in protecting those other taxing rights, so that we don’t see undue focus being put on the taxation of OITs which is an undoubtedly complex type of transaction to tax. We would prefer to see countries have meaningful taxing rights over the rents realised from a location specific asset, which should typically be higher value whilst the asset remains in operation. For countries that have taxing rights over the subsequent rents, it is important to note that taxing an OIT is effectively economic double taxation. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? Further to the comments above, the principle for taxing OITs, as well as the discussion of source versus residence taxation (question 4), may not be sufficient to help tax policy makers answer the question on whether they should (or want to) consider such taxation. Unlike the quoted UN guidance note on direct and indirect capital gains taxation, the Toolkit does not elaborate sufficiently on the pros and cons on choosing to tax capital gains and / or OITs. We hope the Toolkit can provide a more balanced view on such policy arguments that would allow countries to make an informed decision on whether or not the taxation of OITs would be a sustainable contribution to their tax policy. The elaborated principle does make a start into providing a view on how to tax such gains. However, even for the two models, it does not provide sufficient detail on a number of issues to really help a country to provide the necessary certainty to investors and tax authorities on when to expect taxation. The Toolkit could consider other potential mechanisms for taxing economic rents. For example, where the asset relates to resource exploitation and the gain is driven by a change in price of the underlying commodity, an excise tax may be a simpler and more appropriate mechanism. 3. Is the definition of an offshore indirect transfer of assets satisfactory? The simple example in Figure 1, is assumed to be relevant for the whole of the discussion on OITs. In fact, group structures can be complex and an underlying asset may be owned through a chain of investing companies. Where that is the case, it is necessary to explore questions such as: which entity would be taxed on an OIT, can tax apply to multiple entities, etc. More consideration to these aspects is required in the Toolkit. Our recommendation would be that some limitations are put on the taxation of OITs where there is tiered ownership. For example, 1) only taxing a transfer of the immediate parent, 2) only including an upper-tied parent where its principal asset is its investment in the subsidiary owning the subject asset. Limitations on the taxation of OITs between related parties would also be welcomed and this is discussed further below. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? The Toolkit assumes that the source country has the primary taxing right to tax the gain on the underlying property and does not discuss the rationale for residence based taxation of that property. We think that this topic should be further explored in the Toolkit, as part of providing source countries with information on whether or not to seek taxation of OITs. Discussions then must be had between countries if a new allocation of taxing rights is to be agreed. As mentioned throughout this paper, the implications on double taxation must be addressed as part of Toolkit and within these country discussions. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? The potential to expand the definition of immoveable property to “cover rights to receive variable or fixed payments in relation to extractive industry rights or government issued rights with an exclusive quality” , which is left open for countries to consider on page 57, is in our view too broad. It is acknowledged in the executive summary, on page 20 and on page 57 that “the concept of location specific rents is much easier to conceive in economic terms than it is to convey in legal language. This is an area in 26 which further thought is needed.” Accordingly, we would recommend that the Toolkit is updated to clearly state that domestic provisions for the definition of immoveable property should not be drafted so widely at this stage. In our view, one of the strongest cases against proceeding with a wide definition at this stage is that it will increase the instances of varying interpretation of the types of assets to be covered by domestic OIT legislation. Accordingly, there is a risk that uniformity in approach lessens and tax uncertainty increases. This is entirely contrary to the objectives of the Toolkit. We outline below some of the difficulties that our members experience in the taxation of OITs today. For instance, as stated on page 41, “In order to determine whether the value of the interest is principally (more than 50 percent) derived from that immovable property, a comparison is ordinarily required to be made of the value that the immovable property (relevant asset) bears to the value of all the property owned by the entity (all assets) without taking into account debts or other liabilities.” This is not a straight forward exercise for assets currently within the definition of “immoveable property” and, as acknowledged in the Toolkit, will become even more complicated task if a broad range of intangible assets are brought within the definition of “immoveable property”. Without detailed legislation and guidance, there is a risk that tax administrations will seek to find a large number of OITs under the expanded definition of “immoveable property”. We would expect this to lead to many instances of tax controversy. This may put businesses off investing in developing countries that take an aggressive stance on taxation of OITs in light of vague legislation. The Toolkit itself acknowledges at page 33, that the current definition of “immoveable property” in both the OECD and UN MTCs “clearly, […] leaves much scope for more precise definition in domestic law, which [currently] varies quite widely.” We urge the Platform to remain strongly committed to this objective in the overall conclusions drawn in the Toolkit. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? Some simple practical examples would add clarity to the concept. The Tooklit appears to conclude that it would be more efficient to seek to tax the location-specific rents, but is weak on its analysis of the potential mechanisms available. The option of taxing OITs is acknowledged to be an imprecise way of capturing location-specific rents, but it is recommended almost by default. At the very least, a summary of the pros and cons of the different options would be helpful. 7. Are there other implementation approaches that should be considered? As stated above, we understand that this Toolkit has been designed “to help developing countries to tackle the complexities of taxing offshore indirect transfers of assets, a practice by which some multinational corporations try to minimise their tax liability.”. That is, the primary objective appears to be tackling tax avoidance practices. Accordingly, we would encourage the Platform to further explore the possibility of carefully drafted anti- avoidance provisions to tackle the most harmful types of avoidance and to acknowledge the fact that most OITs are not tax motivated. To ease the burden on tax administrations, gateway conditions can be set within the legislation that could, for example, allow a developing country to tax the OIT where the share sale has occurred in a jurisdiction which does not tax the direct transfer of shares because it is a tax haven and that there is clear tax avoidance motive involved. (The use of a white list could be considered to help identify countries that have acceptable regimes for these purposes). Considering the G20 work on Tax Certainty, it would be helpful if the analyses regarding implementation approaches could include input on tax rulings and potentially cooperative compliance. Such instruments could equally help tax administrations triage any potentially abusive situations and provide as much upfront clarity as possible around treatment of OITs that were driven by legitimate business purposes. The Toolkit specifically highlights the problem of “round-tripping” and we would fully support robust anti- avoidance measures to prevent this. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? In our opinion the “deemed disposal” method (Model 1) has significant weaknesses. Accordingly, we prefer Model 2 – taxation of the non-resident seller. 27 In the analysis set out in the Toolkit, it is acknowledged that Model 1 has disadvantages including the possibility of double taxation without foreign tax relief available, e.g. because two different parties are effectively taxed on the same gain (the asset owning company which is deemed to have disposed of the immoveable property and the non-resident company which has made an actual disposal of shares). In our view, this weakness should carry significantly more weight than it does in the analysis set out in the Toolkit. We have set out further concerns below relating to the risk of double or multiple taxation. We do not consider that a method which has, as an inevitable feature, an increased incidence of double taxation is an acceptable approach. We strongly consider that any method chosen should minimise the incidence of double taxation as far as possible. Another stated disadvantage of Model 1 is that the asset owning company is not the company that has received actual sales proceeds. We think the likelihood of tax being collected would be much greater under a model where the person taxed has received sales proceeds. In the Toolkit it is mentioned that “practically speaking, however, it is expected that the parties (particularly the purchaser) would take steps to ensure that the local asset-owning entity had sufficient funds to discharge its tax liability to prevent the tax authority from taking enforcement action against locally held assets.” This may well be the action which businesses would need to take under Model 1. However, it is far from easy to arrange for substantial funds to be provided as may be the case for transfers of immoveable property. Accordingly, for business this is far from a preferred solution. Another key concern regarding Model 1 is in relation part disposals, e.g. if there are two non-resident shareholders (NR1 and NR2) each with a direct interest in Company A (a company with a direct interest in immovable property). NR1 is a majority shareholder and disposes of its interest in Company A to an unrelated party P. Under Model 1, the sale of shares by NR1 is deemed to be a (part) disposal of the immoveable property of Company A. This gain is assessed on Company A. Accordingly, the value of the investment in Company A held by NR2 (a minority shareholder) has decreased. Alternatively, if NR1 is to somehow compensate the minority and fund Company A to allow it to pay any tax due, a question arises as to how NR1 is supposed to inject funds into Company A? A fresh equity injection is not practical as it distorts the existing shareholding percentages and may not be allowed for regulatory/legal reasons, and a loan does not achieve the economic transfer of value (and may also create challenges if there are currency or other restrictions on loan funding and/or interest deductibility). Comparatively, Model 2 aligns with long established international practice of taxing real economic events rather than a deemed disposal and better manages the risks of double taxation. Accordingly, this model is preferred as a means of taxation of OITs, but again only targeted where there is a clear and aggressive tax avoidance motive involved. The disadvantages of Model 2 include it being harder for tax authorities to enforce and collect the tax. We recommend that further work is undertaken to support developing countries in having robust enforcement and collection procedures. 9. Are the complexities in the taxation of these international transactions adequately represented? There are a number of areas of complexity of taxation of OITs which we feel the Toolkit could better address. Risks of double or multiple taxation The Toolkit considers some aspects of the risk of double taxation, for instance, the stated disadvantages of Model 1 described above. However, the risks of double taxation identified in the Toolkit are limited compared with what would be faced in reality by businesses where OITs are taxed in the asset owning country. For instance, page 23 outlines the desire for “direct and indirect asset transfers to be treated identically for tax purposes”. Two routes to achieving this objective are outlined. One, is foregoing taxing rights on either direct or indirect transfers. The other, is taxing indirect transfers which is pursued in the rest of the Toolkit. It would be helpful if this analysis acknowledged that one disadvantage of the taxation of indirect transfers is that the same gain can effectively be taxed twice (or multiple times) by the country in which the underlying assets are located. Firstly, the indirect transfer of immoveable property can be taxed. Secondly, a subsequent direct disposal of the underlying asset can be taxed (and there would be no increased base cost resulting from the indirect transfer). Also, there could be taxation on another indirect transfer occurring at a higher tier of ownership (again without base cost reflecting the fact that the first indirect transfer was taxed). 28 Double or multiple instances of taxation is damaging to economic growth and is widely acknowledged that it should be limited as far as possible to promote international trade and investment. We strongly recommend therefore that more consideration is given in the Toolkit to the instances of double or multiple taxation that may occur as a result of domestic legislation enabling taxation of OITs. More consideration should be given to the tax base cost that may be used for future transactions as well as managing the number of tiers of ownership that can be “looked through” to determine if an OIT has occurred. Further details on this latter concern are outlined below under “commercial transactions”. Valuation complexity As already mentioned above, page 41 outlines the practice of valuation attribution that needs to be followed in determining if an OIT occurs with taxing rights preserved by the asset-owning country under a double tax treaty. That is, “In order to determine whether the value of the interest is principally (more than 50 percent) derived from that immovable property, a comparison is ordinarily required to be made of the value that the immovable property (relevant asset) bears to the value of all the property owned by the entity (all assets) without taking into account debts or other liabilities.” For businesses that are already well-experienced in dealing with taxation of OITs, valuation exercises such as this are known to be a complex task to undertake. In the extractives industry the analysis is further complicated by the impact that large fluctuations in commodity prices can have on the value of immoveable property. More generally, complexity arises from the need to determine how much value is driven by centrally provided inputs, e.g. management expertise, rather than from local expertise. Complications also arise in determining the correct amount of consideration attributable where a company is indirectly transferred with such immoveable property but where it is part of a number of companies or is in a chain of other companies that are disposed of simultaneously (for instance in a take-over of a large multinational with operations and businesses throughout the world). An appropriate calculation could encompass extensive valuations of multiple businesses and assets. Valuation is an important area of administering taxation of OITs that requires judgement and is highly subjective, it can lead to inconsistency in approach between different tax administrations. We strongly recommend that the Platform supports tax administrations and tax payers by developing clear and precise guidance on this topic. Commercial transactions In our opening remarks, we noted that most businesses will be undertaking ordinary commercial transactions and the fact that their choice to undertake an OIT or an asset transfer will be driven by commercial considerations as opposed to an aggressive tax avoidance motive. Accordingly, potentially unintended consequences of rules to tax OITs should be carefully considered. There are a number of areas where we think the Toolkit can go further in exploring mechanisms to ensure that the tax outcomes from ordinary commercial transactions are not over-complicated and uncertain. These areas include: • Considering how many tiers of ownership an asset-owning country should look through in seeking to tax an OIT. For a large multinational group, it is plausible that a divestment could include tens or even hundreds of subsidiaries. In this instance, if some immoveable property is owned at the bottom of the corporate structure it can be very difficult to identify that an OIT has occurred. • There is an acknowledgement in the Toolkit that certain exemptions for inclusion in domestic provisions could be explored. For instance, exempting transfers between related parties or transfers of listed stock or indeed transactions that are undertaken for bona fide commercial reasons rather than an aggressive tax avoidance motive. We strongly recommend that the Toolkit is expanded to list out and fully describe suitable exemptions for developing countries to include in domestic provisions. • Page 50 acknowledges that de minimis thresholds could be further explored. We think this is a helpful suggestion in managing the compliance burden for businesses and recommend that this is further explored in developing the Toolkit. 29 Other issues • Step-up in tax base: The Toolkit rightly introduces the opportunity of a step-up in tax base. It is, however, light on guidance as to how countries are to apply this on an annual basis, e.g. how depreciation or capital allowances are to be determined, how the stepped-up tax base is to be considered over time, etc. • Other options for neutrality: Options such as spreading the capital gains over a number of years, offset by the taxable profits as they arise, or suspension of taxation subject to the underlying assets remaining in the country are not touched upon. We understand the Tooklit is trying to move to uniformity but it would be good to understand why such options are not to be considered. • Minority shareholders: Especially in the extractive industries, but also regarding other long-term licences, investments are often held under shared ownership. In the case of joint investments, the Toolkit is not clear on how its options would affect other shareholders in a venture in case one shareholder decides to reduce or leave the investment. We have shared some of the complexities that can arise elsewhere in this paper. Given the common occurrence of this ownership structure, we think the topic of Minority shareholders must be given full consideration in the Toolkit. • Capital losses: Generally, systems that tax capital gains also allow the deductibility of capital losses. Considering the reduction in commodity prices in the past decade, it is important to be clear on how to treat capital losses when allocated to a country following an OIT. 30 China’s Feedback on The Taxation of Offshore Indirect Transfers-A Toolkit The State Administration of Taxation (SAT) of the People’s Republic of China would like to provide the following comments on The Taxation of Offshore Indirect Transfers-A Toolkit (“the toolkit” or “the draft” as the context requires). I. General Comments On the whole we find the draft well-structured with a clear logic, touching upon key international tax issues relevant to OITs in a holistic and systematic way. We appreciate that the Big Four (IMF, OECD, UN and WBG) put off such an excellent work. We believe that with input from interested parties, the toolkit will be a valuable resource for countries opting to tax OITs. II. Specific Suggestions for Consideration  On Page 18 it is mentioned that Capital gains on onshore direct asset transfers are taxable by the country in which the asset is located (even though the seller—and, likely, also the purchaser—may be non-resident); The term of onshore direct asset transfers is not defined in the draft and we find it difficult to come up with a scenario that an onshore direct asset transfer could involve a non-resident seller or purchaser. We recommend that the draft provide definitions on onshore direct transfer, onshore indirect transfer, offshore direct transfer and offshore indirect transfer.  Based on Box 4 on page 44, one could draw the conclusion that the realizing of assets and liabilities triggered by ownership change is not limited to OIT. The same result can also be achieved by offshore direct transfer, onshore direct transfer and onshore indirect transfer. However, in the case of onshore transfer, whether direct or indirect, the country in which the asset is located can tax the transfer directly without resorting to the Model 1 approach, leaving only offshore transfer exposed to Model 1. We recommend that more thoughts be given to the legality and rationality of Model 1, for example whether this approach is in line with the non-discrimination principle in tax treaties.  The draft points out the pros and cons of the two Models. In terms of the double taxation issue caused by stepping up tax basis or not, it seems that Model 1 does not take into account of the situation in which multiple transfers occur under multi-tiered structures when reaching a conclusion favoring this Model on page 31 47; whereas on page 55 a conclusion is reached disliking Model 2 based on the multiple-transfers-under-multi-tiered-structures premise. We recommend that the pros-and-cons analysis be based on the same premise.  The draft indicates that Model 2 is more widely adopted in practice and one of its disadvantages is the potential double taxation issue discussed on page 55, especially under multi-tiered structures. But the draft is short of providing a solution. We wonder whether solutions could be found to avoid or mitigate double taxation in Model 2.  We think that there is a possibility that a country adopts the two Models at the same time, at least in theory, for example Model 1 for indirect transfer of immovable assets while Model 2 for indirect transfer of other taxable assets. On page 48 it is mentioned that adopting them together may lead to double taxation. We wonder if further analysis could be provided on the mechanism of double taxation, the method to avoid double taxation and pros and cons of adopting the two Models at the same time. If you find the need to adopt the two Models at the same time valid, we recommend that more study be done in designing the rules in that regard.  As to China’s practice in taxing OITs on page 64, we think that the following presents a more accurate account: According to the Corporate Income Tax Law, capital gains is taxed as follows: Gains derived from assets transfer, including movable and immovable assets, by residents or permanent establishments of non-residents, is subject to a 25 percent corporate income tax. Gains derived by non-residents from transfer of onshore immovable assets, or equity interests of residents, not attributable to their permanent establishments, is subject to a 10 percent withholding tax . Subject to the general anti-avoidance rule, China does not levy corporate income tax on the offshore indirect transfer of assets. When the offshore indirect transfer is found to fail the reasonable business purpose test, SAT would re-characterize the nontaxable transaction as a 32 taxable transaction and levy corporate tax based on the rules provided above. Considerations determining whether a transaction meets the reasonable business purpose test include: i) whether the value of the shares transferred offshore is sourced mainly from China; ii) whether the investment and income of the offshore enterprise transferred directly or indirectly is derived mainly from China; iii) whether the offshore enterprise transferred directly or indirectly undertakes substantive functions; iv) the tax consequence of the indirect transfer offshore and the duration time of the indirect shareholding structure; v) the substitutability of the offshore indirect shareholding structure and transaction; vi) the applicable tax treaty and the like. 33 20 October 2017 The Platform for Collaboration on Tax By email to: taxcollaborationplatform@worldbank.org Dear Sir / Madam, DISCUSSION DRAFT: The Taxation of Offshore Indirect Transfers – A Toolkit Thank you for the opportunity to provide comments on the draft toolkit captioned above. Our comments are provided from the perspective of the UK. The taxation of offshore indirect transfers of assets (‘OITs’) has become a focus of significant attention over recent years and we welcome The Platform’s effort to encourage greater clarity and consistency in the approach of policy makers, legislators and tax authorities. We note that the draft presents a wide-ranging and detailed analysis of policy issues. We are also pleased to see suggested language for inclusion in legislation, however, we feel that practical aspects of implementation might be discussed in more detail and make some suggestions in this regard in the attached document. We consider it important that proposed legislation: • Provides certainty to enable businesses to plan confident of the tax implications of their decisions; • Is consistent with tax treaties and norms of international taxation; • Taxes economic gains proportionately; • Eliminates the risk of multiple taxation of the same transaction; and • Does not discourage economically beneficial activities. Tax certainty is particularly important for companies in the extractive industries because of the large investments required and the long lives of projects. Whether the toolkit achieves this, however, will depend on the extent to which low income countries follow the guidance presented in the final version. We also note that the draft presents two quite different approaches for countries to choose from, which may not help in achieving the objective of a coherent approach if retained in the final version. Moreover many countries have already introduced legislation (some of it specifically targeted at the extractive industries), and may be reluctant to change their tax laws, even after finalisation of the toolkit. Please find attached our responses to the specific requests outlined in the press release of 1 August, 2017, announcing the publication of the draft toolkit. We also have some supplemental comments and suggestions which may be found at the end of the attachment. If you have any questions or wish to discuss any aspect of our comments further please do not hesitate to contact Bill Page (bpage@deloitte.co.uk ) or me. Yours faithfully, W J I Dodwell Deloitte LLP 34 DISCUSSION DRAFT: The Taxation of Offshore Indirect Transfers – A Toolkit Deloitte UK comments Responses to questions raised in the press release of 1 August, 2017 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? We understand that the toolkit is being developed in response to requests on practical guidance on taxing OITs, so the question of whether this is desirable from a policy perspective will likely already have been addressed by users of the toolkit. The draft acknowledges the argument that it is economically inefficient to tax a capital gain realized on the direct or indirect disposal of an asset if the gain merely reflects the acceleration of profits that will subsequently be taxable in the host country. The arguments against imposing tax on capital gains include the fact that disposals do not affect the country’s overall share of location specific rents, and that taxing them heavily may inhibit transfers of assets to those most willing and able to develop and operate them. This is particularly problematic for exploration and development of mineral resources, where the triggering of tax liabilities early in the life of a project has an impact on net present value which may be large enough to render a project uneconomic. We acknowledge, however, the attraction of taxing such gains for low income countries which may have difficulty in raising finance from the international capital markets or other sources. We are also well aware of the domestic pressure that governments of developing countries come under if they do not apply tax to such gains. We suspect that these factors are likely to be decisive in most cases. Ultimately this is a policy issue for each country to address. Indeed many countries have already introduced legislation to capture tax on such transactions, as mentioned above. As practitioners our main concern is that any tax should be applied in a predictable and consistent manner, that is should reflect the actual economic gain realized, and that it should not stifle new investment and other potentially beneficial activities. The draft toolkit implies that all OITs are primarily tax-motivated. This is not the case and we suggest that tax policy recommendations should acknowledge this and should not advocate fiscal policies that would penalise groups which hold assets via such structures by imposing disproportionate taxation. Investment structures can be influenced by many factors other than tax planning, for example the requirements of project and other types of financing can be a significant influence; investments in a particular region are often held and managed from regional hubs with good infrastructure and communications such as South Africa or Dubai; access to bi-lateral investment treaties is also seen as desirable, particularly for natural resource projects, to protect large, long term investments against aggressive resource nationalism. The sale of a company owning an asset may be preferred by buyer and seller as it preserves all existing licences, permits and third party contracts for sales and purchases that a company has entered into, which will minimise the impact on day-to-day operations and risk of loss of value. Preservation of such non-tax attributes may result in an OIT being favoured over a direct sale of underlying assets, regardless of tax considerations. It is important to note that the fact that buyers and sellers will always have differing views on the value of any asset to their future business independent of tax considerations. For example, a mature oil field may be deemed ‘non-core’ by an oil and gas major focusing on gas, whilst it may be very desirable for a start-up funded by private equity, regardless of any specific tax attributes of the asset or differences in the tax positions of buyer and seller. It is argued on page 15 of the draft that because an acquirer takes into account the taxation of future revenue to be generated by an asset it is acquiring, any gain realised by the seller must reflect “changes in earnings that would otherwise be untaxed.” The justification for this statement seems to be the subsequent statement (on page 19) that “the exploitation of avoidance opportunities may diminish the effective power of the country in which the underlying assets are located to tax future earnings…” These assertions are questionable and not substantiated in the draft and imply that OITs (and other transfers) are motivated solely by tax considerations, which is not the case in our experience. 35 We note that the statement referred to in the previous paragraph seems to be slightly at odds with the subsequent statement on page 15 that provided the purchaser receives a step-up in basis, the impact of taxation is expected to be neutral, ignoring timing effects, as the purchaser will be able to deduct the purchase price against future revenues for the purpose of calculating taxable profits. The justification for applying tax to any gain is to realise a timing benefit for the host government, which is likely to be attractive to a low income country. This line of argument does not fully acknowledge the difference between a share purchase (where the purchaser would only be able to offset the cost against a future sale, which may not happen) and a purchase of the underlying asset (where a step-up in basis would often, but not always, be given for tax depreciation purposes). Nor does it acknowledge the certainty that whatever expectations of future revenues are at the date of sale, the reality will be different: in effect the taxation of a gain is the taxation of ‘hope value’ and may act as a disincentive to otherwise commercially beneficial reallocations of assets between those for whom they are non-core and those for whom they represent an attractive investment opportunity. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? Please see our answer to the previous question. 3. Is the definition of an offshore indirect transfer of assets satisfactory? In general we find the explanation in the draft adequate, but we note that the ‘stylized’ OIT presented in the diagram on page 12 of the draft may be confusing for some users of the toolkit. In practice many non-OECD countries will tax some or all direct disposals by foreign shareholders of local entities. Deloitte research in 2017 showed that 10 out of 13 Latin American countries reviewed had rules that taxed such transactions in certain circumstances, and 10 out of 14 Asia Pacific countries reviewed also took this approach. An OIT which is mainly motivated by an intention to minimize tax might involve the sale of Corp B, rather than Corp A, in such cases. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? As noted above, we understand that the toolkit is being developed in response to requests on practical guidance on taxing OITs, so the question of whether this is desirable from a policy perspective will likely already have been addressed by source countries. The draft briefly addresses the question of whether some part of OITs might be attributable to increased value created by management and technical expertise provided by the parent company. Whilst this argument is considered to have some merit, the point is made that in many cases the special purpose entity making the sale will have little function other than as a holding company, and relevant expertise is likely to lie elsewhere in larger groups. On the other hand, in natural resource projects in particular, head office teams can play a critical technical, management and finance role in discovery and successful development and this should be compensated. Given the complexities of allocating a gain between the source and residence country, this might be better addressed by re-considering the transfer pricing aspects of intra- group charges for those services, which is outside the scope of this toolkit. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? The scope rules for taxing OITs has focused historically on physical assets such as land and mineral deposits but there are more recent cases involving other kinds of assets, such as telecom licenses, and we understand the rationale to broaden the definition to include these. The concept of location specific rents is helpful in identifying the types of assets that may be included, but we agree that these may be a difficult to define in legal language without creating vagueness and uncertainty. Our preference would be for the approach taken in the draft, i.e. the identification of specific asset categories to be included in any model legislation proposed. 36 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? Please refer to our answer to the previous question. 7. Are there other implementation approaches that should be considered? Both models potentially tax more than the actual gains arising in respect of the relevant immovable property. Model 1 deems the disposal of all assets and liabilities, not just the immovable property. It also does this in any case where more than 50% of the underlying ownership has changed, in other words a taxpayer will pay tax on 100% of latent gains in respect of all assets and liabilities, even if 49.99% of the underlying ownership of the relevant immovable property has not changed. Model 1 would penalise a minority investor in a joint venture entity, where the majority owner is subject to acquisition by a third party, as the value of its interest would be impacted by the requirement for the entity to pay tax in respect of the deemed disposal. It should also be remembered that the entity, not having made an actual sale, will not have generated cash to pay any tax considered to be due. Presumably the buyer (rather than the seller which realised the gain) is expected to fund the taxpayer to settle its obligations absent a funding mechanism agreed before the transaction. Model 2, on the other hand, introduces an extraterritorial element of taxation: in any case where more than 50% of the value of shares or other interests sold derives from immovable property in country X, all the gain on the sale is taxable in country X, regardless of any gains deriving from other countries, or from assets other than immovable property. We also note that the more complex version of Model 2 seems to conflict with the language of article 13.4 of the OECD model tax convention as it taxes disposals where 50% or less of the value of the shares derives from immovable property in the relevant jurisdiction. Article 13.4 only imposes tax where more than 50% of the value derives from such immovable property. It appears to us that both approaches may lead to taxation which is disproportionate to actual economic gains in relation to the immovable property in the relevant jurisdiction and also give rise to risks of extraterritorial taxation and economic double taxation. We believe that the additional tax burden created will be an undesirable disincentive to the efficient functioning of a market that allocates assets to those most willing and able to invest, which in turn could actually damage economic development in low income countries by inhibiting the entry of new investors. Our recommendation is that both approaches should be modified to apply tax on gains attributable to the underlying immovable property in question. In the case of Model 1, the deemed disposal and reacquisition should be applied only to the immovable asset and only to a proportion of the assets reflecting the change in the underlying ownership. For example, suppose that group X owns a gold mine in country Y via a subsidiary Z. The gold mine asset has a tax depreciated value of US$50 million and a market value of US$100 million. If 60% of the shares of Z are transferred, the gain in respect of the gold mine to be taxed will be US$30 million (being 60% x US$ 100 million, less 60% x US$ 50 million). In the case of Model 2 (which is preferred to Model 1 for reasons set out in our answer to the next question), a similar approach could be adopted. Taking the facts set out above, let us suppose that subsidiary Z, in addition to the gold mine, holds other assets outside country Y with a market value of US$70 million. More than 50% of the underlying value of the shares is therefore attributable to immovable property in Y. The gain realised by group X on the disposal of Z is US$50 million. In order to determine the gain attributable to the immovable property, the taxable proportion of the gains could be calculated by applying the ratio of the market value of immovable property in Y to the total market value of assets held by Z (i.e. 100/170 x US$50 million = US$ 29.4 million). Our preference is for a hybrid approach combining features of both models. The intention of this is to apply tax only to the latent gain on the relevant immovable property in proportion to the change in control and to 37 do this in a way that minimises the risk of double taxation, and which provides a clear mechanism for the acquirer to achieve a step-up in base cost. This would be achieved as follows: a) In the case of a change in underlying ownership of immovable property located in jurisdiction L of more than 50%, the gain on the transfer of the relevant shares will be taxable in L. b) For these purposes, the gain will be computed as the difference between the market value of the relevant immovable property in L, less its tax base cost. c) In any case where the change in underlying ownership is less than 100%, the taxable portion of the gain will be the same as the percentage of underlying ownership that changes as a result of the transaction. d) To the extent that a gain has been thus taxed, the base cost of the relevant immovable property will be stepped-up to market value. This approach would not tax latent gains on any assets or liabilities other than the immovable property, would not tax actual gains disproportionately, but it would attach the tax liability to the share disposal, thus providing a basis for double tax relief and minimising the risk of double taxation. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? The draft favours Model 1 over Model 2 as it is stated that Model 2 would require extensive enforcement and collection machinery, for example requiring the local entity to report changes in shareholdings and act as agent for payment of any tax due. We do not agree. In our view adherence by taxpayers to the requirement to report the relevant transaction is the critical issue and this is the same under each model. The additional enforcement and collection machinery contemplated for Model 2 does not seem excessive, and the proposal to require reporting and payment by the entity that actually holds the relevant immovable property (as adopted by Kenya for example) seems reasonable. A default would put the purchaser at risk of loss of the immovable property which would seem a powerful lever to ensure compliance. In our experience it is highly unusual for multinational groups consciously to evade their tax obligations as the broader costs of losing a social licence to operate in a country and the global impact of adverse publicity on a group significantly outweigh and supposed benefit. Regulations for the extractive industries in particular usually impose reporting requirements for changes in the underlying ownership of oil and gas projects or mines, and there is an increasing tendency for OITs to require government consent in the same way as direct transfers. Any breach of these requirements, or failure to pay associated tax, puts rights to the assets at risk. Generally tax is applied to actual rather than deemed transactions. A further drawback to Model 1 is that it does not, as it stands, provide any method for double taxation relief, if the actual seller is subject to tax on the real gain that it realises. This because the tax triggered is payable by a different entity. The effect of this is potential economic double taxation which we consider to be undesirable from a fiscal policy perspective. Our strong preference is for an approach based on Model 2 which taxes the actual gain attributable to immovable property in the jurisdiction, so the ultimate liability should be linked to the economic gain actually realised. The major disadvantage for the purchaser is that the base cost of the shares is only available to offset in the event of a future sale of the same shares. There is no step-up in basis of the underlying asset available, which is a potential attraction of Model 1 for purchasers. Of course, the option of a direct purchase of the asset itself could be pursued if access to the step-up is a key value driver for the purchaser. Alternatively, the hybrid approach suggested in our response to question 8, provides symmetry by stepping up the basis in the underlying immovable property proportionately. 9. Are the complexities in the taxation of these international transactions adequately represented? We appreciate that it is difficult for the toolkit to provide solutions for all conceivable commercial situations, however we feel that the draft could be made even more useful to low income countries by incorporating more detailed, practical guidance on how to apply tax to OITs. In the interest of minimising areas of subjectivity and uncertainty, we agree with the toolkit’s recommendation to adopt specific rules for taxing OITs rather than the application of anti-avoidance legislation. 38 Model 1 In relation to Model 1, it will be important to ensure legislation clearly provides for a consequent step-up (or down) in tax basis of the assets and liabilities deemed to be sold and reacquired, including basis for future tax depreciation. This is assumed by the toolkit, but experience of working with fiscal policy and tax authorities in developing countries suggests it should be flagged more prominently as a requirement to ensure a reasonable tax result. It is also important to consider the specifics of tax depreciation machinery used in the relevant jurisdiction: for example, a pooling mechanism where the tax depreciated value at the start of the tax period is increased by new expenditure, and disposal proceeds deducted from the resulting total will prima facie not give rise to any tax liability as the deemed proceeds and deemed cost of reacquisition simply cancel each other out. Clearly this is not the intention. This might be addressed by ending a tax reporting period at the point that underlying ownership passes and deeming the reacquisition to take place at the start of the next reporting period. To ensure symmetry it is also be important to explicitly recommend that the cost of reacquisition is subject to tax depreciation on the same basis as the original expenditure and without restrictions. Actual disposals of assets may give rise to transaction taxes (e.g. stamp duty) and indirect taxes (e.g. VAT) as well as taxes on the repatriation of profits (via WHT on dividends or branch remittance tax). The draft toolkit does not mention any of these, so it is not clear what the authors view is on these being applied in the case of a deemed disposal. Our strong recommendation is that these are explicitly excluded by the toolkit. The draft does however mention the importance of amending source rules to exclude taxation of the disposal of the shares or other interests in addition to the deemed disposal. This is a key issue in practice, for example Tanzania’s source rules will potentially tax the direct sale of a local subsidiary and at the same time apply change of control provisions to deem a disposal and reacquisition of that entity’s assets and liabilities at market value. There is no specific definition of what is ‘an entity’ for these purposes, but it seems to include a foreign legal entity as well as a tax resident local subsidiary. It seems that it is not the intention to tax foreign legal entities on a deemed disposal of assets and liabilities in other jurisdictions, but this is not explicit and could be a problem area in some jurisdictions where the approach to taxing foreign legal entities may not be clear- cut in law or practice. It is clearly important that domestic law also provides a clear definition of ‘underlying ownership’, though this is not offered by the draft. The intention seems to be to link this to the ownership of the ultimate parent company which could conceivably give rise to the triggering of deemed disposals as a result of normal trading given the three year ‘window’ and would certainly do so in the case of the takeover of a listed entity. However it would seem to provide automatic exemption for corporate reorganisations which do not give rise to a change in the ultimate owner. Experience suggests that the definition will need to be quite elaborate given the broad definition of interests and entities suggested, and it might also need to deal with rights to acquire shares under certain circumstances as commonly found in shareholder agreements. Model 2 The more elaborate version adds a requirement that in the case of shares or interest deriving more than 20%, but not more than 50% of their value directly or indirectly from immovable property, a proportion of the gain should be taxed, based on the ratio of the value derived from the immovable property in the country to the total value of the interest. As noted above, this is not consistent with article 13.4 of the OECD model convention. Whilst this provides some chance that tax will be proportionate to economic gains in the case of such disposals (though the gross value of assets in a jurisdiction is not necessarily indicative of any latent gain), it is clear under this model also that tax will apply disproportionately in the case of shares and other interests that derive more than 50% of their value from assets in the country (though not at all if the value so derived is 20% or less of the total). This could result in double or triple economic taxation. For example, 39 suppose that company X holds interests in oil and gas fields in country Y and country Z. 60% of the value of company X derives from the fields in Y and 40% from Z. In the event of a sale of X, if country Z has introduced the more complex version of Model 2, one could find that the whole gain on the disposal is taxed in country Y and 40% of the gain in country Z. This is a clear case of economic double taxation. In the event that the gains is also taxed in X’s home jurisdiction, there is even the possibility of triple taxation of some of the gain, depending on whether and how the home jurisdiction relieves foreign tax. The draft considers the option of requiring a purchaser to apply withholding at source in respect of share transactions. We do not favour this approach. In the case of a final withholding tax, this creates a tax liability that bears no relationship to the actual gain and would even apply to proceeds in the case of a loss. We consider that tax should be proportionate to economic gains realised. In the case of withholding tax which is treated as a prepayment of any actual tax due from the seller, our concern, based on extensive experience of emerging markets, is that any repayment due may be delayed due to the relevant jurisdiction’s cash-flow exigencies. Where such amounts are denominated in local currency, the resulting foreign exchange losses may be very material. As noted above, we consider the optimal solution would be to adopt a requirement for tax to be reported and paid in-country as agent for the seller by the direct owner of the relevant immovable property. Other issues There are a number of areas which we feel should receive greater emphasis in the final toolkit. Valuation Valuation is fundamental to the successful application of both models. Model 1 requires the taxpayer to agree with the tax authorities market values for all assets and liabilities which may differ significantly from the values recorded in financial statements or tax returns. Model 2 requires agreement on the proportion of the value of shares and other interests deriving from immovable property in the relevant jurisdiction in order to determine whether to tax the gain, and in the case of the more complex version, how much of the gain may be taxed, in defined circumstances. The question of how to value assets and liabilities is not addressed by the draft at all. In the case of a single company with a single material asset (eg a mine or oilfield) it would usually be relatively straightforward to link the sale consideration to the market value of the underlying asset. Transactions are frequently much more complex however, involving multiple jurisdictions, subsidiaries and assets, and could involve swaps of different packages of assets. In such cases our clients have encountered significant difficulties in agreeing market values with tax authorities. For example, in the case of oil and gas assets, tax authorities in developing countries might have difficulties in assessing the reasonableness of reserves estimates, production profiles, price forecasts, cost estimates (including decommissioning), discount rates and any adjustments to the pricing of comparable historic transactions needed to take account of the rapid decline in hydrocarbon values since 2014. It is also unlikely that they will have access to funding to hire third party experts to carry out a valuation on their behalf. This is a reflection of capacity constraints, best addressed by training and access to expertise from other tax bodies under the Tax Inspectors without Borders initiative. However it is important to ensure that users of the toolkit appreciate the need for such support and we feel that the toolkit will be an appropriate place to draw attention to the complexities of this very important issue. We recognise that the purpose of the toolkit is not to serve as a comprehensive manual in valuation techniques, but it would be helpful for the toolkit to provide additional guidance on appropriate valuation methods, and the importance of using multiple valuation methods to validate proposed market values of assets. Base cost Costs are also a potentially contentious issue. Model 1 relies on the application of the normal principles of domestic tax law, which should (at least in most cases) be reasonably well-understood and tested. Model 2, however, will apply to transactions in shares and other interests which may not previously have been within the scope of the relevant tax law, so that issues like determination of base cost may not be so straightforward, for example what is to be the base cost of a share acquired via a share for share exchange, 40 or which has been subject to reinvestment relief in the jurisdiction where it is located? More guidance on these issues seems necessary. Reorganisations and reinvestments The draft mentions that OITs as part of reorganisations may often be treated as tax-free, subject to the continuity of substantial underlying ownership. This is common in legislation and logical, given that a reorganisation does not give rise to an economic gain (for example when the mechanism used includes an exchange of shares in one company for those of another). In the case of Model 1, a tax liability is not triggered where there is no change in the underlying ownership of the relevant assets, so the relief should be automatic. On the other hand, the wording provided to implement Model 2 does not provide any language to exempt such group reorganisations, though these may be carried out for bona fide commercial reasons (e.g. to facilitate financing arrangements). This seems to be an important gap in the proposal and it is to be hoped that it is addressed in the final version of the toolkit. The provision of relief for reinvestment of proceeds is not addressed in the draft, though this too is often found in capital gains tax regimes with similar relief also being provided for the non-cash element of farm-in transactions. It is frequently the case in the extractive industries that the holder of a licence will make a part disposal of its interest in a project (directly or via an OIT) to generate proceeds to finance its obligations in relation to the retained interest. Extractive companies also manage and share risk by diluting their interests in larger or riskier projects. Imposing a tax cost on such behaviour may inhibit transactions which encourage investment and maximise the long term benefits oil and gas projects bring to host countries. Where transactions do take place it will reduce the funds available for future investment in the project with the same effect. Again, it is to be hoped that this will be suggested in the final version of the toolkit. Transactions in listed shares and securities The draft offers countries adopting Model 2 the option of excluding transactions in listed shares and also suggests that certain transactions may be considered de minimis, for example, disposals of shares constituting in total less than 10% of the total issued share capital of the relevant entity. We consider that these should be more strongly recommended, with appropriate anti-avoidance provisions. No similar exemptions are suggested for Model 1, though it is possible that the underlying ownership of an entity could change by more than 50% during a three year period simply as a result of trading on the stock market, particularly in times of price volatility, we would therefore recommend a blanket exclusion of taxation of OITs arising from stock market transactions. Interaction with tax treaties We favour approaches which are consistent with the accepted principles of international taxation, as found in the model tax conventions issued by the OECD and UN. It should be noted that the language of 13.4 of the model convention (now adopted by both bodies) does not state that all the gain will be taxed, so it would be possible for domestic law to apply tax only that portion of the gain arising in the relevant country in line with our recommendation in response to question 7. It should also be noted that it does not provide a right to tax gains in cases where the value attributable to assets in the country lies in the 20% - 50% band contemplated by the more complex version of Model 2. Moreover it does not address at all Model 1’s deemed disposal approach for taxing OITs. Conclusion We appreciate the opportunity to comment on the draft toolkit and we applaud the intention to introduce more clarity around the taxation of OITs. Our key concern is that taxation is applied proportionately to real economic gains in a way that does not penalise investment and we have made a number of suggestions that we feel would improve the draft toolkit in this regard. As valuation is critical to the taxation of OITs we also strongly recommend that the toolkit provides, or is supplemented with guidance on how this important issue should be approached. 41 ICC Comments on the Platform for Collaboration on Tax Discussion Draft: The Taxation of Offshore Indirect Transfers The International Chamber of Commerce (ICC), as the world business organization speaking with authority on behalf of enterprises from all sectors in every part of the world, welcomes the opportunity to comment on the Platform for Collaboration on Tax Discussion Draft: The Taxation of Offshore Indirect Transfers – A Toolkit (hereafter referred to as the Discussion Draft or toolkit). ICC appreciates the work of the Platform to collectively produce “toolkits” for developing countries for appropriate implementation of responses to international tax issues under the G20/OECD Base Erosion and Profit Shifting (BEPS) project, as well as additional issues of particular relevance to developing countries that the project does not address. However, in this instance, ICC believes that the Platform is likely not the best forum to address such substantial changes within the international tax arena. Decisions in taxing rights for “source” and “residence” would cause significant shifts across markets and ICC would therefore recommend that such matters be addressed in a multilateral forum where the tentative recommendations included would be subject to open dialogue as well as legal and economic analysis. In this case, the aim of the toolkit is to provide the analysis, options and recommendations for the tax treatment of offshore indirect transfers. ICC‟s general comments on the Discussion Draft are outlined below. Is the definition of an offshore indirect transfer of assets satisfactory? ICC considers that the first step of the analysis of this topic should be a clear definition of indirect transfers. We believe that it is essential to have an accurate definition of indirect transfers as this would help countries adapt it to their own circumstances and be in a position to establish their legal and tax regimes while remaining consistent with the rest of the tax system. It would also be useful to consider offshore derivative instruments which typically do not give an indirect ownership interest in the underlying asset. The definition provided by the toolkit is as follows: “An indirect transfer involves the disposition of an indirect ownership interest in an asset, in whole or in part. It is the underlying asset that is being indirectly transferred.” (page. 11) According to this definition, if an individual is selling four shares of an entity with immobile assets all around the world; is he or she selling part of the immobile assets and, consequently, should pay taxes in each country where the immobile assets are? It is clear that this is not the aim of the toolkit. If, as opposed to an individual, an entity with liquidity and profitability problems is obliged to sell a package of 5% of their shares, is this company under the scope of the indirect transfer definition? Is the toolkit referring to an indirect participation of X % or is it referring to the control of the immobile asset? Are listed companies in the scope of the definition? 1 42 ICC believes that the definition that the Discussion Draft provides could be a major obstacle for international investments and the financing of these investments as it does not prevent double/multiple taxation and increases the price of investments, apart from other considerations. We recognize that further technical and policy work with a realistic approach is required. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? No. The Discussion Draft assumes that the “source” country has the primary right to tax the gain on the underlying property and does not explore the rationale for residence based taxation of shares. Furthermore, it misstates the current treaty rule. The country of residence has the right to tax capital gains other than those explicitly enumerated by the treaty. The political economy argument focuses on a few high-profile cases that are not necessarily representative of the vast majority of asset transfers, whether direct or indirect. It is ICC‟s view that the high- profile cases might be more appropriately addressed with narrower targeted rules. In addition, the Discussion Draft does not effectively lay out the rationale for limiting the tax treatment to only immovable assets (with its suggested possible expansion). The three illustrative cases provided in the draft toolkit correspond to cases where government licenses were granted, however some countries have implemented law providing for taxation of indirect transfers irrespective of whether the value has been derived from immovable property (including the expanded proposed definition). Does it lay out a clear principle for taxing offshore indirect transfers of assets? ICC holds that tax on indirect transfers of assets can be an impediment to business restructuring. If the disposal is taxed but the acquisition qualifies for tax relief there is a certain degree of consistency (subject to timing), however taxing gains on indirect transfers (where there is clearly no tax relief for the acquirer) would make transactions more expensive. This would mean that in some cases, a transfer which would be economically rational could not take place. The two proposals outlined are clear enough in their general outlines, but, as noted below, there are various challenging issues that are disregarded or treated cursorily. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? No. The Discussion Draft abandons the treaty definition of immovable property and advocates an expansive definition of immovable property, which the Draft acknowledges would be difficult to capture in legislative language. ICC believes that this could be a prescription for uncertainty and double taxation. In addition, the draft toolkit could be improved by inclusion of examples, such as the types of licenses which could be considered in the possible expansion. For example, there could be certain areas which could require licensing from the Government but still not be regarded as 2 43 using the country„s natural resources (such as banking licenses, running hospitals and schools, etc.). Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? The concept of LSRs in the Draft is not helpful in addressing these issues. The Draft acknowledges that access to a local market could be considered to generate location specific rents. ICC would like to note that a concept that is intended to be interpreted expansively, but is inadequately defined could be interpreted in ways that would reduce certainty and deter investment. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? No. The main reason for this is that the local entity would not have the cash from the sale but would be responsible for payment of taxes. While this could create tax credit issues (as also mentioned in the Draft Toolkit), it could also result in inability to get a reimbursement of the taxes from the offshore seller due to reasons such as exchange control issues and reimbursement of taxes which could result in additional taxes, etc. It could also be difficult for the local entity to be aware of changes in the indirect shareholding and be penalized for failure to pay taxes. The approach in the Draft seems to ignore the difficulties associated with imposing a tax on an entity that has no proceeds and may be unable to pay the tax. Moreover, depending on the thresholds, it may be difficult for the entity holding the local property to know that a transfer triggering gain recognition has occurred. Are the complexities in the taxation of these international transactions adequately represented? No. The simplified example that forms the basis of the analysis contained in the Discussion Draft does not take into account the complexities involved in determining whether the transaction should be subject to tax. The Discussion Draft also disregards or glides over the difficulties dealing with minority shareholders, valuation issues, the treatment of losses, and how economic double-taxation would be avoided. There are some complexities in taxation of international transactions which should have been covered – for example, issues that arise due to multiple holding company structures where each country taxes the ultimate seller for indirect transfer of assets, issues arising in determining location of certain intangibles such as patents registered in multiple countries, exploitation of rights granted by local authorities where knowledge and knowhow has been developed in a foreign jurisdiction, computational challenges, etc. The Draft Toolkit should also have provided that tax should not be levied on internal reorganisations including successive transfers pursuant to a restructuring. The Draft Toolkit should not apply to a global sale/acquisition of a company. The Draft Toolkit should also provide recommendations concerning the applicability in the event of private equity structures where the indirect holding structures are not targeted towards avoiding tax arising on direct transfers; and small shareholder exemptions including threshold limits, etc. 3 44 ICC respectfully notes that the Discussion Draft does not appear to address other key concerns for business, given that, from the onset, the document presents the idea of avoidance of direct tax and/or the simplicity of the “stylized three-tier ownership structure” pattern used. ICC believes that the Discussion Draft does not fully address a relatively common situation when all three entities - the owner of an immovable property, the seller and the purchaser- are all non-residents of Country L. In such a case Model 1 (deemed disposal by a local entity) would not provide a satisfactory taxing mechanism for all industries. ICC offers its knowledge and experience to assist in presenting business views on further issues and discussion drafts presented by the Platform. 4 45 Comments of India on the discussion draft titled “The Taxation of Offshore Indirect Transfers - A Toolkit” General Comments 1.1 India welcomes the work undertaken by the platform for collaboration on tax to address the challenges of addressing the base erosion and profit shifting arising from indirect transfer of assets, particularly movable assets, the capital gains on the transfer of which is taxable in the country where the actual assets are situated, and from where the indirect or the derivative assets derive their value. 1.2 India considers this work to be of immense importance for developing countries, and essential for their domestic resource mobilization. Thus, India strongly supports this initiative and urges the platform to clearly and unambiguously recognize indirect transfer of assets for minimizing taxes, as an abuse of both the domestic law as well as tax treaties, in the same way as other tax avoidance practices like treaty shopping have already been recognized as impermissible by the Final Report on Action 6 of the BEPS Project. India strongly urges the collaboration to highlight in the report that with the inclusion of modification of the preamble of the treaty specifying that the treaty is not intended to permit abusive behavior, such artificial arrangement aimed at minimizing taxes have already become impermissible under the treaties, once they are amended by the Multi-lateral instrument. 1.3 India strongly supports the work undertaken by the platform of collaboration to develop tools that will facilitate the developing countries in addressing the challenges of tax avoidance posed by indirect transfer of assets. India would like to suggest that it may be clearly noted in the report that every State has the sovereign right of taxing of capital gains arising from assets that are likely to generate income or profits within that state, in accordance with the global consensus that profits and gains should be taxed in the jurisdiction where the economic activity giving rise to them is located, and not in jurisdictions where a secondary asset without any real economic activity is located by an artificial arrangement, merely for the purpose of avoiding tax liabilities. 1.4 India would like to point out that the models proposed in the draft tool-kit need clarity as the domestic law measures and the tax treaty obligations have been combined. India would like to point out that is not a practical approach since domestic laws and tax treaties are determined in very diverse ways, under completely different circumstances and considerations. Whereas a State has only one set of domestic laws applicable on all taxpayers, which is decided by the legislature, every State has several tax treaties determined by negotiated processes that often tend to differ from each other. Advising one set of tools that include both domestic and 46 treaty remedies would not to be very helpful. Accordingly, India strongly suggests that the domestic law measures and the tax treaty measures for addressing this challenge may be considered separately, and completely independent of each other. 1.4 At places, the draft tool-kit appears to be aimed at addressing only the capital gains arising from indirect transfer of immovable assets. India strongly urges that in the tool-kit being developed no difference should be made between the taxation of capital gains arising from immovable property (which in any case has already been dealt in BEPS project) and movable property. Alternatively, the focus of the draft report could be to deal with the capital gains on indirect transfer of movable assets (that have not been given sufficient attention during the BEPS project). 1.5 With respect to the domestic law measures, India would like to point out that these measures need to be applied within the Constitutional and legal framework applicable in a particular jurisdiction, which may differ significantly from one State to another. Therefore, there is a need to provide greater flexibility in the suggested tool-kit, by providing more options and improving the flexibility of each of the model law. 1.6 With respect to the tax treaties, India would like to point out the need to introduce changes in the tax treaties to address this issue. India strongly suggests that for this purpose, the anti-avoidance rule in paragraph 4 of Article 13 in the Model Tax Conventions, which currently prevents tax avoidance by artificial arrangements involving indirect transfer in respect of immovable property, should be extended to cover the capital gains that are taxable in the source country under paragraph 5 of Article 13 in the UN Model Tax Convention. India believes that the need to address tax avoidance in respect of capital gains covered under paragraph 5 of Article 13 is as important as the need for addressing capital gains by transfer of immovable property covered under paragraph 1 of Article 13. 1.5 India would also like to suggest that since the tool-kit is being developed for the developing countries, it is essential that this exercise is undertaken with the inputs and participation of developing countries. India suggests that the platform may invite Government representatives from developing countries, and engage with them in detailed discussions before proceeding and finalizing this report, so as to ensure that their concerns and views are appropriately taken into account. 1.6 India does not agree with the stated Conclusions of the draft report focusing exclusively on taxation of capital gains from immovable property, which in the view of India, are not in accordance with the preceding discussion or details. India would like to point out that the issues related to taxation of capital gains from immovable property have already been dealt with in the BEPS project and are in the process of implementation in the MLI. 47 Hence, restricting the conclusions and recommendations to capital gains from immovable property will not result in a meaningful exercise. Specific Comments 2. With reference to the issue of “B. How the taxing rights should be allocated” India would like to point out that it has already been widely accepted in the recently concluded BEPS project with the endorsement of all countries of G-20, OECD and other associates, that income should be taxed in the jurisdiction where the economic value is created. Since the capital gains of an asset represent the accumulated value of its returns over time, they should be taxed only where the actual assets from which the gains are expected to be derived in the future are located. As pointed out in the draft report, the rights for taxation of capital gains from direct transfer of those assets provide a clear indication that the indirect transfer should be taxed similarly. India would like to point out that since the economic impact of indirect transfer is same as direct transfer, and it is fully recognized that gains should be taxed where the value is created, there cannot possibly be any argument that economically there is a difference between the direct transfer and the indirect transfer of assets. 2. India fully agrees with and strongly supports the view taken by the collaboration that the taxation rights to tax capital gains should be allocated taking into account the tax jurisdiction where the location specific rents are generated by an asset, irrespective of whether it is an immovable asset or a movable asset. India also strongly supports the view that the taxation rights should be allocated keeping the “benefit principle” of taxation in view, and the State that facilitates the generation of such ‘location specific rents’ with public resources must have a right to seek taxation of capital gains arising from them. 3. With reference to the counterargument on page 21 that “The increased value of the entity sold may reflect in part managerial and other expertise contributed by the seller, beyond what has been recovered in managerial fees, royalties and other explicit payments”, India would like to point out that it is a flawed and self contradictory argument , since the legal entity owning the assets directly and the related entity that derives the capital gains are legally distinct entities and any transactions between them are required to be undertaken at arm’s length price. Once that condition is fulfilled as per the tax treaties, this argument loses relevance. Thus, India fully agrees with and strongly supports the conclusion drawn by the collaboration of not taking this flawed argument into account. 4. India fully agrees and strongly supports the need to maintain tax neutrality between direct and indirect transfer of assets. India would like to point out that artificial arrangements 48 aimed at indirect transfer of assets amount to a costly economic exercise that utilizes factors of production without producing anything in return. India strongly urges the collaboration to point out that such unproductive exercises impose an avoidable cost of real resources not only on the developing countries, but the global economy as a whole, and hence must be strongly deterred. 5. With reference to the difference between immovable property and movable property, India would like to point out that economically there is no difference between the two. This non existence of difference is already evident from the very broad definition of immovable property in Article 6, which includes rights, such as mining rights, which are in the nature of intangible property and hence not different from movable property. India would like to point out that the source of difference between movable and immovable property is the difference between the taxation of capital gains in the OECD and the UN Model Tax Convention, as apparent from the differences in paragraph 5 of Article 13 of the two model conventions. 6. With respect to the Vodafone case which has been given as an example in Box 1 at page 26 in the report, India will like to bring to notice that the facts in second and the third paragraph of the Vodafone box is not an accurate position. Firstly, as per Indian law the purchaser is required to deduct tax at source while making payment to non-resident seller and secondly, the amendment through Finance Act, 2012 was brought in as a clarification to explain the intent of Indian legislation in situations of indirect transfer. Since the report is in a draft stage, the second and third paragraph in the report should be amended as under. “As per the Indian Law, the purchaser is required to deduct tax at source while making payment to the non-resident seller. Accordingly, the Indian Tax Authority (ITA) held the purchaser, Vodafone’s Dutch subsidiary, liable for failure to comply with its obligation to withhold tax from the price paid by it to Hutchison on the ground that the capital gains realized by the seller were taxable in India. This sparked a protracted court case, with the Supreme Court of India ruling in 2012 in favour of the taxpayer. The Supreme Court denied the ITA’s broad reading of the law to extend its taxing jurisdiction to include indirect sales abroad, though it took the view that the transaction was in fact the acquisition of property rights located in India. The government of India subsequently brought in a clarificatory amendment with retroactive effect to overcome the technical difficulty arising out of the Supreme Court ruling so as to allow taxation of offshore indirect sales and to validate the tax demand raised against the Vodafone's Dutch subsidiary. The legality of a retroactive effect of the law was subsequently not challenged by Vodafone in the Indian courts and instead it has submitted the action of the government of India to arbitration under the India-Netherlands Bilateral Investment Treaty". 49 7. With respect to the domestic law measures, India agrees with the observation of the collaboration that the tool-kit is only indicative in nature and needs to be adopted in accordance with the specific needs of the applicable Constitutional and legal framework of a tax jurisdiction. India suggests that the collaboration may consider including more options for addressing this issue, and also provide greater flexibility in the applications of the various models suggested in the draft tool-kit. 8. With respect to the tax treaty measures required for addressing tax avoidance by indirect transfer of movable assets, India would like to point out that where the treaty provides taxation of capital gains from transfer of movable property under Article 13(5) as in provision based on UN Model, such intent to allocate taxing rights to the source state should be preserved by extending the scope of Article 13 (4) to such capital gains in Article 13 (5). India would like to point out that this can be achieved relatively easily by amending the text of article 13(4) as under: 4. Gains derived by a resident of a Contracting State from the alienation of shares or comparable interests, such as interests in a partnership or trust, may be taxed in the other Contracting State if, at any time during the 365 days preceding the alienation, these shares or comparable interests derived more than 50 per cent of their value directly or indirectly from immovable property, as defined in Article 6, situated in that other State or from shares referred in paragraph 5 . 8.1 India would also like to point out the following justification for recommending the amendment in Article 13 of the UN Model and tax treaties based on it: • The purpose of Action 6 Report was to prevent the granting of treaty benefits in inappropriate circumstances, and for this purpose, certain changes have been recommended in the OECD Model Convention and Commentary in the Final Report on Action 6. Since that exercise was undertaken keeping only the OECD Model Convention in view, it will be appropriate for the Platform for collaboration on tax, constituted specifically for addressing the concerns related to developing countries to consider treaty abuse concerns that arise only in the UN Model Convention. • Paragraph 41 to 43 of the Report on BEPS Action 6 address the issue of transactions that circumvent the application of Article 13 (4) of the OECD Model Convention. It does not refer to the UN Model as the practice followed during BEPS was to focus exclusively on the OECD Model Convention, though it was expected that concerns specific to the UN Model can be subsequently dealt by the Committee of Experts. • Unlike OECD Model Convention, paragraph 5 of Article 13 grants certain taxing rights to the country of source for taxation of capital gains from transfer of shares. While Article 13 (4) addresses abusive transactions with a purpose of avoiding tax in respect of 50 paragraph 1 of Article 13, it does not prevent similar abusive transactions in respect of taxation of capital gains under paragraph 5. Since the nature of transactions that circumvent taxation of capital gains under paragraph 1 and paragraph 5 are exactly the same, there appears to be every reason and justification for preventing abusive transactions that circumvent the application of paragraph 5 of Article 13. • Given the emphasis on preventing granting of treaty benefits in inappropriate circumstances and all possible measures being recommended by the global community for addressing abusive transactions for avoiding tax, we should consider measures that will prevent transactions that circumvent the application of Article 13 (5) of the UN Model Convention. The need for such measures is further highlighted by the urgency shown in further strengthening of the anti-abuse provision in Article 13(4) in the OECD Model Convention. 9. India would also like to request the collaboration to consider the possibility of recommending that the concept of “beneficial owner” which has already been introduced in Model Conventions dealing with interest, dividend and royalty income, should also be introduced in Article 13 of the Model Tax Conventions dealing with capital gains, to prevent tax avoidance in respect of capital gains. 10. India does not agree with conclusions of the draft report that the primary issue relates to taxation of immovable property, or that the ‘key issue’ is the appropriate definition of “immovable”, which are clearly not based on the detailed discussion and economic analysis included in the rest of preceding draft. India would like to point out that the issues related to taxation of capital gains from immovable property have already been dealt with in the BEPS project and are currently in the process of implementation in the MLI. The definition of ‘immovable property’ is also available in Article 6 of the Model Conventio ns as well as their Commentaries on Article 6. Hence, restricting the conclusions and recommendations to capital gains from immovable property will convert this report into a meaningless exercise. 10.1 Accordingly, India strongly urges the platform for collaboration on taxes, which was constituted on the request of G-20 to address the challenges faced by developing countries that were not taken up during the BEPS project, to focus equally on the capital gains from indirect transfer of movable assets, to the extent the capital gains from the direct transfer of those asset are taxable under Article 13 (5) of the model conventions or analogous provisions in the treaty. 51 September 25, 2017 To: Platform for Collaboration on Tax From: International Tax and Investment Center (ITIC) Oil and Gas Taxation and Regulatory Dialogue Re: Request for Comments – Discussion Draft: The Taxation of Offshore Indirect Transfers (OIT's) - A Toolkit The Oil and Gas Taxation and Regulatory Dialogue of the International Tax and Investment Center (ITIC) is pleased to submit the following comments on the Platform for Collaboration on Tax - Discussion Draft: The Taxation of Offshore Indirect Transfers (OIT's) - A Toolkit (hereafter “Discussion Draft”). We appreciate the opportunity to provide our views and look forward to working on elements of future work. In addition, we would be pleased to meet at any point with representatives from the Platform to provide further background on our comments and to discuss particular points we have made or questions you may have. *********************************************** Executive Summary of Comments I. The Platform group, and individual countries, should focus on what they consider to be abusive transactions and employ more focused anti-abuse rules, rather than take the very broad and expansive approach suggested in the draft. The Discussion Draft notes concern about investors avoiding capital gains taxation in countries where assets are located by selling entities indirectly holding such assets (and located outside of that country). It further notes the “technically highly complex” nature of the issues, both in terms of their economics and legal aspects. International oil and gas companies, and many other multilateral enterprises, have ownership structures that have developed over decades with many reasons other than avoidance of tax in countries where their producing assets are located. Transactions 1634 Eye Street, NW, Suite 500, Washington, DC 20006 Tel: +1 202 530 97 99 / Fax: +1 202 537 7987 / Washington@iticnet.org / www.ITICnet.org Almaty • Astana • Baghdad • Baku • Dubai • Jakarta • Kiev • Lagos • London • Moscow • Nay Pyi Taw • São Paulo • Washington, DC 52 between companies, and even restructurings within organizations, often involve many parts of the ownership structure and many countries (rather than the one on one correlation of assets and indirect owner as illustrated in the stylized example in Discussion Draft Figure 1.) Thus, not only are the issues "technically highly complex" in terms of economics and legal aspects, but even more so in terms of the practical application of the rules. Several examples are provided in our comments to expand the “stylized” fact pattern to more complex and, in the case of multinational oil and gas companies, more realistic ones. When these realities are added, the complications compound exponentially. If the concern is really one of a country being deprived of tax revenue based on a tax avoidance motive, anti-abuse rules provide a much more targeted way of addressing such cases. The complexities of implementing the proposals contained in the Discussion Draft create uncertainties in multiple jurisdictions at multiple levels separately and in the interaction between one another. While the Discussion Draft speaks of the need to avoid creation of double, or multiple, taxation, changing rules to provide or expand taxing powers among jurisdictions creates greater risks of multiple taxation. This is a major concern in doing business in the international economy. Given the complexity of the fact patterns involved, and the uncertainties and additional risks associated with a broad application of, and imprecisely defined or administered, “extra-territorial” taxation, we strongly recommend that the Platform, and individual countries, focus on what they consider to be abusive transactions and employ more focused anti-abuse rules, rather than take the very broad and expansive approach suggested in the draft. This is particularly appropriate when focusing on oil and gas projects, since the practical reality is that the country in which the assets are located ultimately exercises complete power over them. We also offer a proposed framework for analyzing this complex issue. II. The toolkit should contain a full discussion of the additional option of a country not to tax OIT's. The Platform group has decided to make some specific policy recommendations in this toolkit. Rather than making a recommendation, we believe to be most valuable, the toolkit should contain as wide an array of “tools” as may be useful to developing countries, allowing individual countries to determine what may work best in their particular circumstances. For example, we believe that the option of not taxing OIT's should be included in the toolkit and discussed—providing both pros and cons. As noted in the Discussion Draft, the United States under FIRPTA does not reach foreign indirect sales of U.S. property held by a foreign corporation. Norway effectively takes this approach with respect to oil and gas 2 53 assets, demonstrating that it is not unreasonable for a country to decide not to tax OIT's (or even some direct transfers). Although realizing an acceleration of tax from an OIT might appear beneficial to a country, it would not be unreasonable for the country to recognize that the long-term consequences outweigh the short-term benefit. Therefore, the toolkit should note that a country might appropriately determine that other “taxing” or even “non-taxing” options may be best under its unique circumstances. III. IF a country, after considering the pros and cons of taxing OIT's, decides to do so, THEN it must provide equivalent treatment for losses and provide and fully describe a mechanism to step up the basis both in the underlying assets that will continue to be used in generating taxable income within the country as well as the local country stock, if any exists, in order to prevent double taxation. The basis step up must apply not only for future changes in control, but also for determination of taxable income from operating the assets, or the sale of some/all of the assets as well. Throughout the Discussion Draft, the importance of a step up in basis of the assets “indirectly” purchased is recognized to avoid double taxation in the country in which the assets are located. However, there is never an explicit recognition of the step up in basis for purposes of future cost recovery (depreciation, depletion, etc.), or future sale of some/all of the assets, but rather only for purposes of future capital gains on changes of control. With the position of Model 1 of the Discussion Draft being a deemed disposal and reacquisition of all of the immovable assets by the local country company, the treatment should be the same as an actual disposition and reacquisition and therefore a step up in the basis of the underlying assets is required. This should be the same result under Model 2 for reasons described in our comments and under the analytical framework we believe is appropriate. We do not believe that a recommendation as to the method of imposing a tax should be made. IV. The toolkit should be more balanced and precise when describing the level of potential base erosion in an income or profits tax applicable to oil and gas projects. It cuts too wide a swath with its comments, is misleading in several important respects, and can lead to invalid policy responses. The report devotes a great deal of analysis to the issue of allocation of the taxing rights on OIT's, considering equity, efficiency and political economy. With respect to equity (ensuring a country has the ability to tax “location specific rents”), for significant oil and gas projects, it is commonplace for their development to be done in joint ventures with other international oil companies (and often with participation by the host country’s national oil company). One of the unique but universal features of this structure is the “no profit” or “no markup” rule under which charges to the joint venture by the operator must be made without any markup. With respect to the revenue side, again particularly with respect to crude oil, widely available index prices and quality and 3 54 location differentials are key parts of a government’s toolkit in testing the arm’s length nature of the revenue reported by taxpayers with respect to the crude oil lifted and sold from a joint venture. Therefore, from both the expense side and the revenue side, in the case of an upstream petroleum project there is a significantly lower risk of profit shifting than where these features do not exist. The Beer and Loeprick study, cited in support of widespread base erosion and profit shifting in the oil and gas industry, is a flawed study which should not be cited with approval and endorsement. It creates misconceptions and fuels misleading and inappropriate speculations about this issue. In addition to the above “equity” related comments, we also offer comments on the efficiency and political economy points contained in the Discussion Draft. In particular, we point out that not taxing OIT's does not mean not taxing the rents from the local assets, especially in the oil and gas sector, unlike how these situations are often portrayed. *********************************************** Comment Discussion Overview Comments: The Discussion Draft is intended to provide “analysis and options and recommendations for the tax treatment of offshore indirect transfers.” In particular, it notes concern about the possibility of investors avoiding capital gains taxation in countries where assets are located by selling “interests” indirectly holding such assets but located outside of that country. It further notes the “technically highly complex” nature of the issues, both in terms of their economics and legal aspects. To illustrate the issue, the Discussion Draft provides a stylized example of an OIT structure in Figure 1 on page 12. While it notes that more complex structures are possible, and even common, most of the analysis that follows is based on a highly simplified fact pattern. For example, the stylized example consists of a transfer of shares in an offshore company whose only asset is shares of an onshore company which owns the assets in question. It further seems to assume that structures put in place (even more complex ones) are done so primarily for tax purposes, and hence to provide mechanisms to avoid tax due in the country where the assets are located. (See footnote 11 on page 14 implying the interposition of additional tiers of entities “perhaps for tax planning purposes…as in some cases, [a local country] taxes gains up to the first tier of ownership.)” See also the discussion on page 15 which, after recognizing the highly stylized nature of the example, nevertheless concludes: “Not coincidentally, however, many indirect transfers are indeed structured so as to bring precisely the features assumed in the example of Figure 1 into play.” 4 55 In reality, international oil and gas companies, and most other multilateral companies, have ownership structures that have developed over decades with many reasons other than avoidance of tax in countries where their producing assets are located. Further, transactions between companies, and even restructurings within organizations, often involve many parts of the ownership structure and many countries (rather than the one on one correlation of assets and indirect owner as illustrated in the stylized Figure 1.) Therefore, not only are the issues "technically highly complex" in terms of economics and legal aspects, but perhaps even more so in terms of the practical application of the rules. If the concern is really one of a country being deprived of tax revenue based on a tax avoidance motive, anti-abuse rules provide a much more targeted way of addressing such cases. The Discussion Draft makes note of these rules (and China’s adoption of that approach) but really gives very little credence to them. In fact, the suggestion is that the drafters believe taxation rights regarding indirect transfers should extend far beyond cases of potential tax avoidance. For example, in discussing an anti-avoidance approach, the Discussion Draft states: “This type of rule would only reach the gain in question if intentional tax avoidance regarding the transaction could be shown. Such rules would therefore not provide that the gain in question should be taxed as a matter of principle on the basis of a substantive right to tax in the location country and would be much more limited in scope.” This makes it quite clear that the issue is not “avoidance” of tax, but expansion of taxing rights and jurisdiction. When the Discussion Draft further suggests expanding the scope of the coverage beyond “immovable property”, which has up to now been the focus of the model tax treaties (and as noted, have only been picked up in 35% of the treaties), we are concerned where this line of thought might lead—e.g., no limit to the nature of the assets, or to transactions involving changes in control? In fact, would this lead to no limits on the application of such an OIT except in a case with the most minimal of connection with a country? Looking at things quite differently, consider cases like the Exxon-Mobil or BP-Amoco mergers. These cases involved mergers of top level companies in each organization, with hundreds of subsidiaries in dozens of countries. The transactions were driven by overall economic and business considerations of the managements of the organizations. But the rationale of rules suggested would create the right of every country to essentially “mark to market” assets within their countries and impose a tax because of a “change of control” event. One must question the overall benefit, compared to the compliance cost, of forcing a detailed analysis in every country to determine if, and the extent to which, such a transaction could trigger taxes on all of the “indirect” transfers that are deemed to occur. Other transactions occur frequently where a company decides to divest from a general area of business, transferring a holding company with subsidiaries in multiple jurisdictions. Again, this is hardly an attempt at avoiding local country tax, but simply an efficient and effective way of 5 56 reorganizing a business. But all along the ownership chain, the tax rules of every single country would be implicated under the rationale of the Discussion Draft. Another case that may exist is a legal entity incorporated in one country (e.g. country L) that has operations and assets in multiple other countries, through branches with no legal entity structure in the countries of operations. The practical application of the proposed rules to this case, in the event of the sale of the legal entity in country L, or even a higher tier entity, in each of the countries with operating assets, opens a "Pandora's Box" of issues to be resolved. This despite the transaction having no conceivable impact on the future taxation rights of the countries with operating assets, or any conceivable tax avoidance motives. Finally, assume a case just like that in Figure 1 except that Corporation B, in addition to owning shares of Corporation A (with assets in Country L), also owns shares in Corporation C (with assets in Country M). When Corporation B shares are sold, both Country L and County M, under the proposal, will invoke jurisdiction, they will “compete” over allocation of value, and the prospects of multiple taxation will grow. The illustrations above are simply intended to expand the “stylized” fact pattern beyond the “all assets in one country,” held indirectly by a single parent, which in turn owns no other assets to more complex and, in the case of multinational oil and gas companies, more realistic ones. When these realities are added, the complications compound exponentially. The complexities of implementing the proposals create uncertainties in multiple jurisdictions at multiple levels separately and in the interaction between one another. While the Discussion Draft speaks of the need to avoid creation of double, or multiple, taxation, changing rules to provide or expand taxing powers among jurisdictions creates greater risks of multiple taxation. This is a major concern in doing business in the international economy. Focusing on oil and gas projects, the facts actually call out for a narrower focus, because the country in which the assets are located ultimately exercises complete power over them. In fact, as will be commented upon more below, if anything, the “captive” nature of such assets, after a substantial investment has been made that established value in the immovable assets, makes investors the ones most “vulnerable” to economic changes. Irrespective of what may happen outside the country, the fact remains that as production occurs, it is visible, and taxed, and mechanisms exist to ensure the local country obtains the tax revenues due. Additionally, even before production, or irrespective of profitability, many other fees, taxes, and charges are due and payable. Given the complexity of the fact patterns involved, and the uncertainties and additional risks associated with a broad application of, and imprecisely defined or administered, “extra- territorial” taxation, we strongly recommend that the Platform, and individual countries, focus on what they consider to be abusive transactions and employ more focused anti-abuse rules, rather than take the approach suggested in the draft (which we fear is really intended only to be further expanded in scope in the future). 6 57 We would suggest the following framework for analyzing the offshore indirect transfer issue, particularly as it relates to oil and gas projects (clearly immovable assets in a country in which a company simply cannot ultimately avoid taxation). Framework for Analyzing Capital Gains on Offshore Indirect Transfers The purpose of subjecting to tax certain “extraterritorial” activities or transactions is to keep a country from being deprived of the tax revenues it would receive in the absence of the transfer (or in the case of an actual asset transfer). The purpose should not be to artificially increase the revenues a country would otherwise receive. If the effect of taxing such OIT's in fact is a revenue increase—then it is a clear change in law of the country, arguably amounts to double taxation, and in any event moves the tax rate on such transactions well above the no transfer case (or an actual direct sale of assets within the country—if that is used as a comparison model). It amounts to a tax surcharge on a limited class of transactions, which we believe is improper. The analysis of whether and if so how to subject OIT's to taxation should be evaluated against this principle. IF a country ensures itself that it will get the same revenues if an OIT occurs,either in the same or earlier timeframe, than in a no-transfer or actual asset transfer case,then that will achieve the goal and should be a neutral outcome. This result can be achieved in two basic ways. One can decide not to tax an OIT, on the basis that the activities within the country will be the same as in a no-transfer case, and timing of taxes will be the same. This is clearly what should apply in a “non-abusive” transaction case. A country could also decide to tax the transfer, but provide a step up in the basis of assets in country for purposes of the income tax. This results in an acceleration of the income tax, and thus the country receives that revenue earlier than in a no-transfer case (or a no taxation of OIT case). This could be applied in an “abusive” transfer case—effectively penalizing a seller by accelerating tax that otherwise would only be due over the project life. The mechanism for taxing an OIT can be tailored by the country based on what it feels is most efficient. For example, the Discussion Draft considers two approaches, one a deemed asset transfer and the other a deemed in-country sale of stock. However, whatever the mechanism chosen, it is imperative that the basis step up be applied at the asset level and be applicable for ongoing depreciation, depletion, or asset sales purposes. Without this, the basic principle expressed above—i.e., of ensuring the country receives what it would have in the absence of a transfer (or on an actual asset transfer) is violated. It is further critical that if a country decides to tax certain OIT’s, its rules regarding the scope of transactions that will be covered, and those that will not, need to be clear, as straightforward as possible, and in place prior to when initial investments are made. Ambiguity on this issue 7 58 increases risk, which reduces investment attractiveness. Further, sound and equitable tax administration simply requires such clarity. We now turn to some additional general comments on the Discussion Draft. *********************************************** General Comments General Comment 1: The toolkit should contain a full discussion of the additional option of a country not to tax indirect transfers. The Platform group has decided to make some specific policy recommendations in this toolkit. Rather than making a recommendation, we believe that to be most valuable, the toolkit should contain as wide an array of “tools” as may be useful to developing countries, allowing individual countries to determine what may work best in their particular circumstances. Therefore, we believe that the option of not taxing OIT's should be included in the toolkit and discussed—providing both pros and cons. For example, as noted in the discussion draft, the United States takes this approach with respect to OIT's (see page 63 where it is noted: “Importantly, however, a foreign corporation can hold U.S. real property and the disposition of its stock by a foreign investor is not subject to U.S. tax; FIRPTA does not reach foreign indirect sales of U.S. property held by a foreign corporation.) In addition, in some sectors, in some countries, even direct transfers of certain assets are not always subject to taxation at the time of transfer. Norway takes this approach with respect to transfers of oil and gas assets. It recognizes that it will receive the same total amount of revenue by not taxing such transfers and therefore not providing a step up in basis in the assets involved. While it may appear to experience a timing detriment, it has determined that in not accelerating a tax liability, it likely encourages better and more efficient development of its resources resulting in a net present value increase in revenue. The very recent transaction between Maersk and Total is a prime example of this—one can only wonder if that transaction would have occurred if a large upfront acceleration of tax had been due. The point of these two illustrations is simply that it is not unreasonable for a country to decide not to tax OIT's (or even some direct transfers) and thus a discussion of this approach belongs in the Toolkit. Finally, we believe that even if it contains specific recommendations, the toolkit should note that a country might appropriately determine that other “taxing” or even “non-taxing” options are best under its unique circumstances. The additional sub-points below are relevant to this issue: 8 59 1. The point made above in the Norwegian illustration suggested Norway sees benefits to not immediately taxing direct transfers of oil and gas. This approach encourages transfers and promotes efficiencies, such as where a new investor may be a better operator of the assets at a particular stage in their development, or may be willing to make further investments that the original holder will not make. An additional net benefit may accrue to a country in not taxing such transfers. Depending on relative “discount rates” investors use for evaluating projects in a particular country, there may be an actual present value loss from a country imposing a tax at the time of transfer. While the Discussion Draft notes that a country forgoing immediate taxation will see a time value of money detriment (see Discussion Draft page 16), it does not address how investors may evaluate their time value of money detriment from an acceleration of tax. In fact, if investors’ present value costs are greater than the country present value benefits, an actual overall financial loss will occur in valuing the natural resource activity. This will affect the overall attractiveness of the country’s resources, compared with other countries, and can actually reduce the actual, and present, value to a country. [See for example the IMF Fiscal Analysis for Resource Industries (FARI) model assumption that an investor will do project specific economics, requiring a return to compensate for the risks of that specific project, including specific country risks as well, rather than using as its discount rate an overall enterprise weighed average cost of capital. This can often result in the investor discount rate on a specific project being higher than the applicable country’s cost of funds rate. Additional background on the FARI model is available at: https://www.imf.org/external/pubs/ft/tnm/2016/tnm1601.pdf.] 2. Countries do in fact compete for investment capital. Policy recommendations that ignore this fact of life may actually be harmful in practical terms. The Platform should help countries understand the full implications of the policy options available to them, and then allow them to construct a framework that works best under their own circumstances. One size does not fit all just as, in the area of natural resources, no two countries are alike in terms of geography, geologic potential, or the economic, political, and multiple other factors contributing to an investor’s risk. A country with high geologic potential and many interested investors can command tougher fiscal terms than a country with much less geologic potential and many other risks. The decision to impose a tax at the time of a transfer rather than deferring it (like the cases of the US or Norway) is a key fiscal term. Particularly in the oil and gas business, where investors seek to diversify risks and optimize their “portfolio” of projects on a worldwide basis, rules in a particular country that make this more difficult or more expensive to manage can in fact be detrimental to investment in that country. This simply needs to be understood as a consequence of a policy choice to tax certain transfers. 3. The Report notes the numerous complexities involved in adopting, implementing, and administering a system that taxes OIT's. One can avoid these resource consuming 9 60 requirements by simply relying on the regular income tax system to generate the same (or perhaps, on a present value basis, even more) tax revenues. On the other hand, if a country decides to tax certain OIT’s, it is important that it address all of the complex issues presented, such that clarity is provided to investing taxpayers, and to the country’s own tax administrators. The task of considering and deciding how to deal with all of these issues, however, is time and resource consuming and far from simple. This again raises the cost-benefit issue which a country should consider. 4. Not taxing OIT's provides tax receipts on a more regular basis, a key benefit in a sustainable budget system for governments. 5. In concluding that that countries should tax OIT's, the Discussion Draft references the OECD and UN Model treaties as suggesting "wide acceptance" of the principle, yet acknowledges that the relevant article of the treaties is only included in 35% of all Double Tax Treaties. 6. The Discussion Draft concludes on page 58 that a more uniform approach to taxation of OIT's needs to be adopted by countries. While this may be a desire of the Platform members, it is hardly a rationale for a particular country to adopt an approach that may be harmful to its own circumstances. Countries need to be allowed to decide on an approach to this issue which may not necessarily be uniform, either for competitive reasons or because of interactions with other elements of their tax systems that would cause this tax to create double taxation. *********************************************** General Comment 2: IF a country, after considering the pros and cons of taxing indirect OIT's, decides to do so, THEN it must provide equivalent treatment for losses and provide and fully describe a mechanism to step up the basis both in the underlying assets that will continue to be used in generating taxable income within the country as well as the local country stock, if any exists, in order to prevent double taxation. The basis step up must apply not only for future changes in control, but also for determination of taxable income from operating the assets, or the sale of some/all of the assets as well. The Discussion Draft does not adequately address the treatment of losses on OIT's. If the application of the rule is limited to anti-abuse situations the need for application to losses would no doubt be eliminated. However, if the provision is subject to general application, it is imperative that equivalent treatment for losses be explicitly provided. The Discussion Draft concludes as stated on pages 5 and 59 that the favored method of imposing tax on an OIT is a deemed disposal of the underlying immovable assets by the local entity and a reacquisition of those assets at market value. As discussed below, we believe it is imperative that a complete description of the mechanics of whichever taxing method is chosen be provided, and most beneficially through examples. 10 61 While as previously stated the decision of whether to impose a capital gains tax on OIT's and, if so, how to impose the indirect tax is a complicated policy and procedural question to be decided on by a country, we do agree with the conclusion in the Discussion Draft that a decision to tax the gain must provide for basis step up (adjustment of basis) of all the assets deemed acquired and on which gain was taxed. Throughout the Discussion Draft the importance of a step up in basis of the assets purchased is recognized (see pages 5, 4, 16, 42, 43, 45, 47, 59 of the Discussion Draft) to avoid double taxation in the country in which the assets are located. However, there is never an explicit recognition of the step up in basis for purposes of future cost recovery (depreciation, depletion, etc.), or future sale of some/all of the assets, but rather only for purposes of future capital gains on changes of control. Footnote 12 on page 14 of the Discussion Draft makes the point that an acquisition of the immovable asset directly would generally result in an increase to depreciation allowances that would yield deductions sooner than the basis in shares could be offset against future gains. With the position of Model 1 of the Discussion Draft being a deemed disposal and reacquisition of all of the immovable assets by the local country company, the treatment should be the same as an actual disposition and reacquisition and therefore a step up in the basis of the underlying assets is required. This is somewhat affirmed on page 45 of the Discussion Draft which states that "...the nature of the disposal is only a deemed (as compared to an actual) disposal for tax purposes. Therefore, the local asset owning entity will still be the legal owner of the assets after the disposal is deemed to take place. In order to protect against double taxation, the model treats the local asset owning entity as reacquiring the assets for their market value. This means that its tax cost in those assets is stepped up to market value---which is important to ensure that double taxation does not arise in the location country in the event that another subsequent change of control occurs." However, double taxation will clearly also occur if the basis step up does not occur for purposes of cost recovery (depreciation, depletion, etc.), or for a future sale of some/ all of the assets. Therefore, it is critical that the Discussion Draft clarify that there is a step up in the basis of the assets for both cost recovery and future sales. The position of neutrality between a direct and indirect transfer is endorsed by the Discussion Draft on page 23. The Discussion Draft states that "One natural requirement for neutrality is that direct and indirect asset transfers be treated identically for tax purposes. That is, transferring an asset or transferring shares deriving their value from that asset, to the extent they represent the same transfer of ownership, should ---all else equal---attract the same tax treatment." If a country chooses to tax an indirect transfer, we agree with this concept at least to the extent that if the immovable assets triggering the indirect tax would have been stepped up with a direct sale, then they should likewise be stepped up with an indirect sale. We further agree with the conclusion stated on page 16 of the Discussion Draft: "So if, for instance a future sale will be taxable in the same jurisdiction as today's sale, and at the same tax rate, then the total nominal (undiscounted) revenue raised from the capital gains tax over time will be zero: that is, the same as if there had been no sale or no capital gains tax." 11 62 However, this statement is true only if there is a step up in basis for cost recovery or future asset sales—such that the actual capital gains tax paid on the change in control is offset by future reductions in income taxes—leading in the aggregate to no net tax as a result of the capital gains tax. If the step up were allowed only for future changes of control, this statement would not be true. In that case the net (undiscounted) tax raised would be the tax on the first sale plus any tax on additional gain realized in the second sale—this is tax in addition to the tax due under the income tax and thus is a net additional total amount of tax. The "Effects on other tax payments" discussion, also on page 16 of the Discussion Draft, is inconsistent with the previous statement on page 16 above. The first "effect" listed is that "the transfer has no direct impact on country L's future receipts of corporate income tax (or in the case of extractive industries, any royalties or rent taxes)". This clearly takes the position that there is no step up in basis for cost recovery purposes. This should be corrected to reflect the impact on income tax as a result of a step up for cost recovery with respect to the second "effect", although as stated on page 17, while there may not be a change in the rate of withholding tax on future dividends, the effect of the step up in asset basis and resulting increase in cost recovery will reduce distributable income and therefore the amount of dividends subject to withholding tax. Therefore the "effect" should be corrected to reflect this. The Discussion Draft should be clear throughout the document that the step up in basis is with regard to the underlying immovable assets for future cost recovery and sales of some/all of the asset as well as for future changes of control. Without this basis increase, there will be double taxation on the future earnings of the buyer, either through tax on recognition of the anticipated revenue stream or a sale of some or all of the assets. These conclusions can be demonstrated by simple examples. For this, we reference a Tax Notes International article Capital Gains Issues in the Extractive Industries1 which clearly articulates these points in example form. Applying the principles of those examples to the structure on page 12 in the Discussion Draft we provide the following: Corp. A is expected to generate net cash of 100,000 for each of the next 10 years. For tax purposes, Corp. A's assets are fully depreciated and there are no other differences between net cash and taxable income over the next 10 years. Assuming a Country L tax rate of 50%, Corp. A expects to generate after-tax cash of 500,000 over the next 10 years (50,000 after tax cash per year X 10 years). Similarly, Country L will receive 500,000 of tax over the next 10 years. Presumably Corp. B will require an after-tax sales price from Corp. P2 of 500,000 and Corp P2 should be willing to pay not more than the 500,000 of anticipated after tax cash to be realized over the next 10 years. If the tax law of Country L allows for basis step up, there should be no impediment to Corp. B and Corp. P2 reaching a deal. Consider the following cases: 1 Karl Schmalz, "Capital Gains Issues in the Extractive Industries," Tax Notes International (3 October 2016). 12 63 Case 1: Corp. B's gain taxed/Corp. P2 gets basis step up for cost recovery/future asset sales Assuming Corp. P2 is willing to pay 1,000,000 for the Corp. A stock, and gets a step up in basis, it will have depreciation to fully offset the 100,000 pre tax cash flow each year. Therefore Corp. P2 will pay zero tax over the next ten years and after ten years will break even on its investment. Corp B.'s 1,000,000 sale will be taxed at 50% resulting in after tax cash of 500,000, the same it expected to realize over the next ten years. And Country L realizes the same 500,000 of tax revenue it would have received over the next ten years. Case 2: Corp. B's gain not taxed/Corp. P2 gets no step up In this case Corp P2 should not be willing to pay in excess of 500,000 for the Corp. A stock, the projected after tax cash over the next ten years. And Corp. B should be willing to accept a 500,000 sales price, its same anticipated after tax cash over the next ten years. Corp. P2 essentially steps into the shoes of Corp. B. Corp. P2 will realize 100,000 pre-tax earnings and pay 50,000 of tax each year for the next ten years realizing a breakeven total cash flow of 500,000. Likewise, Corp. B will realize a 500,000 sales price with no tax and Country L will realize the same 500,000 of tax revenue. As can be seen from a comparison of Case 1 and Case 2, as long as there is symmetry (tax and step up in basis, or neither) between the buyer and seller, the net result for all parties is the same. And as indicated on page 16 of the Discussion Draft, the net undiscounted tax is unchanged, irrespective of whether the OIT is taxed or not. However, in a case that does not provide symmetry, i.e., no step up for cost recovery/future asset sales, the results are flawed since they change the total amount of taxes a country would receive from that of a direct asset sale. Case 3: Corp. B's gain taxed/Corp. P2 no basis step up for cost recovery/future asset sales Under these rules, Corp. P2 will only be willing to pay 500,000, the net after tax cash flow over the next ten years. However, Corp. B will not be willing to accept less than 1,000,000 because the sales price will be subject to 50% tax resulting in after tax of cash of 500,000, the amount it would realize over the next ten years without a sale. If Corp B did accept a 500,000 sales price it would only realize next cash of 250,000 and Country L would realize tax revenue of 750,000 over the next ten years, resulting in double tax. While the result in Case 3 might appear favorable to Country L, in reality, the tax implications would likely prevent the buyer and seller from reaching agreement on the sales price. This could limit development in those cases where the buyer might have been a more effective operator of the assets, or been willing to make additional investments to enhance the operation. This is clearly not an “efficient” outcome. As depicted in the above examples, if the Discussion Draft recommends the deemed sale method of taxing OIT's, we believe this method must allow a step up in basis for cost recovery/future asset sales. 13 64 We also believe use of examples as provided above should be incorporated into the discussion draft to clarify the intended application of the principles promoted. The Discussion Draft on page 15 states that: "Since the value that any actual or potential holder places on an asset can be expected to take into account any future corporate, withholding or other taxes due---including capital gains tax on any future sales---capital gains tax reaches income not taxed by these other instruments. Viewed one way it is a form of double taxation. More economically relevant, however, it is a way to capture changes in earnings that would otherwise be untaxed." We agree that capital gains tax is clearly a double tax to the extent it is taxing income retained in the business that has already been taxed and will potentially be taxed a third time when ultimately distributed to the shareholders. However, we disagree that it is "taxing earnings that would otherwise be untaxed", but rather it is at best accelerating a tax on changes in earnings that will be taxed when realized and at worst creating double taxation to the extent no basis step up is allowed. *********************************************** General Comment 3: The toolkit should be more balanced and precise when describing the level of potential base erosion in an income or profits tax applicable to oil and gas projects. It cuts too wide a swath with its comments, is misleading in several important respects, and can lead to invalid policy responses. The report devotes a great deal of analysis to the issue of allocation of the taxing rights on Overseas Indirect Transfers, considering equity, efficiency and political economy. Equity: On equity, when focusing on the importance of “immovability” of an asset, and the implication that its value reflects its location specific rents, it states: “…in practice there is also a widespread …recognition that it is appropriate for revenue from taxing … [location specific rents] to accrue to the place of location. The most obvious examples of such assets are often thought of—and in the resource case generally are—owned collectively by the nation.” [p.19]. The report further states that the “best way to tax such rents is by a tax explicitly designed for that purpose…” and then makes two additional points as to why a capital gains tax on offshore indirect transfers may be needed in addition to such “explicitly designed” rent taxes within the country. First it asserts that such taxes are not invulnerable to profit shifting (citing a 2015 draft of an article authored by Beer and Loeprick) and then concludes therefore that a capital gains tax can “be a useful backstop when implementation of such taxes is imperfect--though clearly inferior to an ability to effectively tax them as they accrue.”[emphasis added). Where it can be demonstrated that there is a low chance of “profit shifting,” and thus such taxes are far from “imperfect” but instead are effective in achieving their designed outcomes, then the superior manner of taxing location specific rents is under the “explicitly” designed rent tax, as they accrue. In the case of an upstream petroleum project, particularly an oil project, 14 65 this is precisely the case. Under this rationale, a country could reasonably forego a capital gains tax (even on a direct transfer, much less impose one on an indirect one which raises so many additional issues of complication and administration). Again, this is precisely what Norway has done. [Note that this approach also solves the issue of having to deal with tax treatment of losses that occur when a prior gain which was taxed under a capital gains tax turns out to have been overestimated—another aspect of “equity” which the Discussion Draft should address.] With respect to significant oil and gas projects, it is commonplace for their development to be done in joint ventures with other international oil companies (and often with participation by the host country’s national oil company). One of the unique but universal features of this structure is the “no profit” or “no markup” rule. Under longstanding practice in the oil and gas industry, one of the partners to a joint venture is appointed as the operator to manage and oversea joint venture activities, BUT costs it charges to the joint venture must be made without any markup. It is in the interest of all of the non-operating members of the venture to keep these costs low, and to audit such costs to ensure they are consistent with the “no profit” or “no markup” terms of the contract. Thus, on the cost side of oil and gas joint venture projects there is an alignment of interest between the country and the non-operating partners and there is little or no room for “base erosion.”2 With respect to the revenue side, again particularly with respect to crude oil, widely available index prices and quality and location differentials are key parts of a government’s toolkit in testing the arm’s length nature of the revenue reported by taxpayers with respect to the crude oil lifted and sold from a joint venture. Therefore, from both the expense side and the revenue side, in the case of an upstream petroleum project there is a significantly lower risk of profit shifting and consequently not a good rationale for imposing a capital gains tax as a back stop for the rent tax. In practice, given the “immovable nature” of an oil and gas project developed in a country, it is actually the taxpayer that is more vulnerable with respect to application of the fiscal terms, including rent taxation, than the government. Investors see being subject to this vulnerability— after they have borne the development risks-- as the opposite side of the “coin” of governments’ claims (supported by NGO’s and others) that they must take certain actions to ensure investors pay their taxes. Extractive industries view this with special concern because of the enormous amount of investment required upfront to establish the existence of the resources that create value in the asset. At that point in time the investor may find itself at risk of expropriation either directly, or through increases to tax and/or royalty rates. 2 It is often stated that key areas for “base erosion” are inter-affiliate management or service costs and financing. An operator’s “inter-affiliate management or service costs” are precisely the types of costs that are covered by the “no profit” or “no markup” rule, and financing costs are often not permitted within a joint venture, or if they are— as in a project financing case—they are either with or based upon unrelated third party arm’s length arrangements. Furthermore, deductibility of financing charges, even if permitted, is one of the easiest areas for governments to set specific limitations on and to police. 15 66 Finally, the ITIC wants to go on record as challenging the validity of the 2015 Beer and Loeprick report finding “evidence of extensive profit shifting in the [oil and gas] sector, with signs that developing countries are especially vulnerable.” One need only look at the fact that 60% of their sample data is from three countries (the UK, Norway, and the Netherlands) and another 20% is from Russia to see that any “conclusion” with respect to “developing” countries is unsupportable simply by the amount of the data they have used. More important, to assert that a country like the UK is losing up to a third of its tax base from base erosion is simply preposterous. Further, our understanding is that the conclusions are not a reflection of taxpayers utilization of aggressive base erosion techniques which may go undetected on audit, but also consist of policy choices a government has made which, if made differently could, repeat could, have generated more revenue. This in our view is truly a flawed study which should not be cited with approval and endorsement. It creates misconceptions and fuels misleading and inappropriate speculations about this issue. Efficiency: The arguments in the Discussion Draft regarding location specific rents are the same as have been outlined above. But another efficiency consideration, that of neutrality between direct and indirect transfers, is an interesting one. Arguably, the Norwegian approach is the most “efficient” system with respect to neutrality, since it treats direct and indirect transfers equally and it avoids the inefficient and resource consuming effects of imposing an additional tax which arguably does not economically—even on a present value basis given different views of investors and governments—provide additional revenues. But, even if a direct transfer is subject to a tax under the country law, it may still be efficient not to seek to impose that tax on an indirect transfer. While this may not be “neutral”, it may still be more efficient to the country in terms of revenues per administrative effort. Finally, in order to be neutral—if that is the definition one uses to define “efficiency”—then since a direct transfer results in a basis step up, a taxed indirect transfer must also provide a basis step up for further depreciation, depletion, or asset sales purposes. Political Economy: The Discussion Draft suggests that not taxing OIT's can lead to domestic dissatisfaction and harm efforts to build a taxpaying culture. This may be true if the tax system is opaque and political leaders are unwilling or unable to explain their system and dispel inaccuracies. Not taxing OIT's does not mean not taxing the rents from the local assets, especially in the oil and gas sector, unlike how these situations are often portrayed--even by those with knowledge. Political leaders, and commentators, should be prepared to educate their constituents (or other actors) to keep such misconceptions from driving economic and tax policy. Also, worthy of discussion are the investors’ views of the political economy. For a country to insert a tax on OIT's on existing projects alters the fiscal terms for investors after they have completed their long-term capital commitments. This is the type of fiscal instability that can reduce future investments and ultimately have a negative impact on long term fiscal stability of the country. 16 67 In summary, we are not suggesting it would be unreasonable or inappropriate for a country to decide to tax certain OIT's. We are only suggesting that it would likewise not be unreasonable or inappropriate for a country to decide NOT to tax certain OIT's. Further, if a country decides to tax such transfers, it should appreciate that such a decision will have economic consequences in terms of investor reactions and it should carefully design the scope and mechanism for such taxation. For example, in the oil and gas industry it is commonplace to bring additional investors (partners) into a project via “farm-in” arrangements. Placing a tax cost on such transactions can lead to inefficient resource development. Similarly, mergers and acquisitions not necessarily specifically targeted to changing the ownership of one country’s local assets often qualify for “tax free reorganization” treatment because discouraging them may lead to inefficiencies and far from a “cashing out” of an investment, they contemplate continued operation of the businesses involved. Therefore, rather than recommending a very broad rule of application, we suggest that taxation of OIT's should focus on abusive transactions. Further, a country should not subject an OIT to taxation if such a transaction would not have been taxed if done directly. Where it does decide to tax a certain class of OIT's, it needs to provide a basis step up—not just for purposes of resale of such assets but also for ongoing operation of the assets. Thus, the step up needs to reset depreciation and other cost recovery levels as well. The Discussion Draft should be clearer on this point. *********************************************** Additional Specific Comments Ø On the bottom of page 18 of the Discussion Draft, the last sentence should be corrected to read, "...which the location country could have or did tax in the past and may, or will tax in the future..." Ø On the top of page 19, continuing with the thought from page 18, we disagree that the gain "reflects earnings that the location country has in a sense simply chosen not to tax." Rather it reflects earnings previously taxed by income tax, or to be taxed by income tax when realized and likely taxed a second time when distributed to the investor. Ø On page 46, the Discussion Draft states that liabilities must be restated with the assets in the application of Model 1, but that "no gain or loss in a liability would be expected to be realized in the ordinary case...". In fact, there will almost always be a gain/loss on fixed rate debt and almost as likely even on floating rate debt and the Draft should be revised to reflect this fact. Ø On page 48, the Discussion Draft acknowledges that the entity directly holding the assets will be liable for the tax on the indirect transfer, but may not have the liquidity to pay the tax. To the extent the local entity borrows the necessary funds, either from the 17 68 purchaser, or elsewhere, the OIT legislation should specifically provide for deductibility of the related interest expense. Ø On page 56, the minimal definition of immovable property appears to be broad enough to apply to mineral resources, making any change unnecessary. *********************************************** Administrative Comments We note that certain format items in the draft will need to be conformed in the final product. Just by way of example, which you may have already adjusted, the introduction of the Discussion Draft indicates that the Draft is arranged by section and specifically identifies sections III through V, however the table of contents is not arranged by, nor does it identify, sections. Also, the body of the Draft does not include section numbers, although there is a reference on page 18 to section 5, which presumably should be section V. Submitted by Daniel A. Witt, David Delahay, and Karl B. Schmalz; International Tax and Investment Center; USA; +1 202 530 9799; dwitt@iticnet.org; ddelahaysr@aol.com; kbschmalz@gmail.com 18 69 From: Kate Zeller [mailto:katejbu@gmail.com] Sent: Friday, October 20, 2017 4:11 PM To: Platform for Collaboration on Tax Subject: Comments on a Draft Toolkit on the Taxation of Offshore Indirect Transfers of Assets To Whom It May Concern: On behalf of Jubilee USA, an alliance of more than 700 US faith groups, I'm writing to offer comments on the Draft Toolkit on the Taxation of Offshore Indirect Transfers of Assets. As the toolkit notes, tax avoidance schemes like offshore indirect transfers deprive governments of the resources required to provide necessary services like education, food and healthcare to those in need. We support the creation of this toolkit to address this tax avoidance method that has previously not received much attention. You have sought comment, among other topics, on whether the draft toolkit effectively addresses the rationale for taxing offshore indirect transfers of assets. In articulating the importance of this and other tax avoidance schemes, we suggest discussing the high amount of money lost to countries through corporate tax avoidance. According to a 2016 report from Oxfam America, corporate tax avoidance drains approximately $111 billion a year from the United States, and an estimated $100 billion from some of the world's poorest countries. In terms of the broader issue of illicit financial flows, between 2000 and 2008, $6.5 trillion left the developing world completely untaxed. These resources can be used to provide vital services to poor communities. We applaud the focus of the toolkit on equity and fairness. The recommendation that source locations be protected and facilitated in their right to taxes on location specific rents and capital gains is not only fair, but will significantly help lower income countries raise much-needed tax revenue. It will also help reduce tax avoidant corporate maneuvers and bolster countries’ ability to take such entities to court and recover lost tax revenue. We recommend that the implementation of these norms be prioritized in future treaties as widely as possible in the international tax arena and that greater tax equity remain the focus of these readjustments as we move to a more comprehensive and fair international tax system. Sincerely, Eric Eric LeCompte Executive Director Jubilee USA Network eric@jubileeusa.org 70 To To The Platform for Collaboration on Tax – Date 3 October 2017 taxcollaborationplatform@worldbank.org From KPMG International cc Christopher Morgan, KPMG – christopher.morgan@kpmg.co.uk Comments on Discussion Draft: The Taxation of Offshore Indirect Transfers – A Toolkit KPMG welcomes The Platform for Collaboration on Tax Discussion Draft “The Taxation of Offshore Indirect Transfers” (“the Draft”). The issue is important both for countries looking to increase or protect their tax base and for investors who require certainty regarding the cost of an investment. Reaching agreement on a framework for deciding which gains should be taxed and which should not be taxed and on the methods for applying taxation will facilitate better tax policy and greater certainty. Our response is divided into a summary, some general observations about the structure of the document, some specific comments on the drafting and replies to the questions. Summary The taxation of offshore gains should fit within the overall domestic policy for raising revenue and may be viewed as also needed to stop potential avoidance. At the same time rules should provide equality of treatment between domestic and overseas investors and should give clarity and certainty, while being as simple as possible to administer for both taxpayers and tax authorities. The impact on inward investment also needs to be considered in crafting any regime. We believe the Draft could be improved by: • Recommending that policy makers first consider what the overall domestic policy aims to achieve, including what sort of assets should be subject to capital gains tax, and then how any taxation of indirect offshore gains fits within that policy; • Analysing in greater depth the reasons for taxing offshore gains associated with “immovable property”, including how to define “immovable property”; • Raising the issue of a country’s requirement for inward investment in infrastructure type projects and how the proposals impact on this; • Containing a framework which recognises the impact on multinational enterprises (“MNEs”) and sets out rights and duties of governments; • Examining in more detail the consequences of the various proposals for taxing offshore gains with either suggestions for more targeted rules or an explanation of why some of the counter-intuitive results are acceptable within the overall policy; • A change in the tone in some places where, currently, it could be perceived as suggesting that that offshore gains are the result of unacceptable tax planning – rather than a commercial reality. 71 General comments Over all structure In helping governments consider all the implications in developing local policy, and to assist the debate over all, the Draft could perhaps be structured to raise the following issues: i) What assets should be subject to capital gains taxation under local rules? What other implementation issues arise such as depreciation / amortization rules, step-up of base cost on sales, etc.? This would be a new section but would not have to be a detailed analysis. It should raise the issues to be considered as a starting point before examining how to tax offshore gains. ii) Given the local regime, what are the issues around, and the best practices regarding, rules on capture in cases of indirect transfer? iii) What are the issues regarding the wording of the language in existing model tax treaties? iv) What are the possible options and issues regarding implementation? v) Valuation issues around such assets. This would not have to be an in-depth discussion but should raise the issues of how to identify the value of the different assets – eg the need to distinguish value added by management of assets as opposed to assets themselves. Should gains realised offshore be subject to tax in the ultimate source country where the underlying assets are based? The conclusion of the Draft is that offshore gains from immovable property should be taxed but not offshore gains in general. There is some discussion about expanding the definition of immovable property so it would cover things like telecommunication licences or even “location specific rents” in general. The Draft could usefully examine the rationale for distinguishing between immovable and other property in more detail, especially if the definition of immovable property may be expanded beyond the traditional treaty and domestic law definitions. Apart from pragmatic reasons, the main argument put forward for taxing immovable assets is that they derive their value, in part, from their location. Without greater clarity there is a risk that some countries could use the reference to location specific assets to justify imposing a charge on the offshore transfer of a wider category of assets. For example, the offshore sale of a subsidiary which carries on manufacturing would not, under current practice, give rise to a tax charge in the local country – unless more than 50% of the value was derived from, basically, real estate. If the test was widened to include gains from location specific assets it would still appear that such a gain was not taxable. However, in certain cases a country may argue that the specific location increased the value of the manufacturing facility because, say, the local work force was highly skilled or subject to very low wages. Therefore, it may be argued that the offshore transfer of the subsidiary should be subject to local capital gains tax on a location specific asset. Such reasoning would introduce subjectivity and a lack of clarity into the rules and could easily lead to double taxation or international disputes. Uncertainty could negatively impact on investment decisions. The rationale for taxing immoveable property appears to be that real estate, natural resources and the like are to a certain extent part of the common patrimony of the country and therefore any gain realised from the sale should be taxed. Furthermore, natural resources are wasting assets and once they are exhausted a country has lost the ability to tax them. This should be highlighted because it helps delineate what should and should not be included in the taxable category. By contrast, an investment in renewable energy – such 72 as equipment used in a wind or solar farm – does not have the same characteristic of being a natural wasting asset (and the land itself on which any plant is built is covered by the general definition of immoveable property). Furthermore its value comes more from the investment than a location specific property. These factors indicate such assets should be treated differently from immovable property such as a mine. The Draft suggests that immoveable property could include certain licenses – such as for telecommunications. There should be more analysis of the rationale for such an inclusion. In some cases the government might be preventing competition from other enterprises or providing some kind of price support and therefore contributing to the value of the business. However in other cases government regulation may protect the consumer and reduce profitability. This suggests that the licence in itself is not something which should be subject to any offshore gains rule. Impact on inward investment The Draft should raise the question about the impact of tax on investment. Countries should consider their infrastructure needs and policy and the extent to which these require funding by inward investment. Such an analysis becomes even more important if the definition of immoveable property is expanded to include a wide range of infrastructure projects such as power supply, water, and telecommunications. Taxing indirect offshore transfers may increase the after tax cost of a project and reduce investment or, in some cases, all or part of the cost may be passed on to the end consumers which may not be desirable where the project is providing public utilities. These factors need to be weighed against the benefit of increasing the tax take and the Draft could help jurisdictions develop a framework for weighing the relevant considerations. The need for a framework The Draft concentrates on how to tax the offshore gain. It would be more complete if it contained a framework which addressed the rights and obligations of the taxpayer as well as governments. For example, however the rules are drafted, they should: • Only tax gains attributable to immovable property (however defined); • Provide neutrality so that an offshore gain is not taxable where the same gain would be exempt if made on-shore – eg due to a participation exemption; • Provide for dispute resolution to avoid double taxation and correct apportionment of the gain; • Not be retrospective or as a minimum provide for assets to be rebased to market value at the date of introduction. • Ensure that double taxation is avoided by providing either credit relief or exemption in the state in which the offshore gain is realised; • Ensure that double taxation is avoided by providing for assets to be rebased so that the same gain is not taxed a second time on a future sale; • Contain an exemption where the disposal is part of an internal reorganisation within a group; The first four are not covered in the Draft. While the final three are mentioned, they could be covered in more detail. 73 Specific comments on the paper 1. On page 13, in describing the example it states “imagine in Figure 1 the owners of P1 want to realise a capital gain reflecting an increase in the value of the underlying asset”. The reference to wanting to realise a gain could be interpreted as suggesting that the structure has been driven by a desire to avoid tax on gains. It would be better to use a neutral phrase. In the vast majority of cases the sale will be effected for commercial reasons: the MNE may wish to exit the particular country or to rationalise its business and focus on one division rather than another. The intention of making the sale is not, at least primarily, to make a tax exempt gain. If there is a proposal that any rules should apply different treatment between different types of investors according to the length of the holding or their investment profile, this should be clearly examined and justified. 2. On page 16 in the first paragraph it says that “the total nominal (and discounted) revenue raised from the capital gains tax over time will be zero.” This needs to be qualified as it will not always be the case. The situation is complex and the result will depend upon such issues as local depreciation rules, the step up in the base cost of assets and the method of imposing the tax on the offshore transfer. Suppose the subsidiary in question opens a mine and exploits it until it is exhausted. It then closes the site and the subsidiary is liquidated. In this case the subsidiary will be taxed on its profits from the extraction business according to local law but there will be no capital gains tax on the value of the mine. If however the subsidiary is sold while the mine is still in production phase, any capital gain will become taxable. The local country will still be able to tax all the profits of the production. Therefore the tax on the gain is effectively a “windfall” unless local rules allow the new owner to depreciate the market value of the mine for tax purposes and so set off an amount against its income. 3. At the top of page 22 it states “the reason for this is that any such changes mean that resources are being used in a way that are socially efficient, but are privately profitable only because of taxation.” It appears the word “not” is missing from the before “socially”. 4. On page 28 some comments are made about the three case studies. It notes that in each case the amount of tax at stake is very large and in one case amounts to “nearly 50% of public spending on health”. It is unclear why it is necessary to make such comparisons. If the point is that taxation of offshore gains can raise a significant amount of tax it should also be highlighted that the other side of the coin is this may be a significant cost for MNEs doing business. There may be a knock-on effect to the level of investment or a disincentive for MNEs to restructure their businesses to focus on their areas of core competency, which could lead to a lack of efficiency in the provision of infrastructure or services to the consumer. 74 5. On page 29, referring to the Peruvian case study, it notes “the transaction became linked with corruption scandals, leading to the dismissal of the Prime Minister and Cabinet.” It is unclear why the statement is relevant to the Draft. It is unhelpful if it somehow suggests that offshore capital gains are (often) associated with corruption. 6. The table on page 31 sets out the allocation of taxing rights under the UN and OECD model treaties depending upon whether the transfer is onshore or offshore. The table suggests that where the assets are moveable and “Seller has PE in Country L to which the assets are located”, tax can be levied in Country L even where it is an indirect offshore transfer. This is not very clear as it does not explain why the seller – of the indirect interest – also has a PE in country in L. In any case there would have been no direct transfer of the movable assets in any PE and any gain would only be taxable in the hands of the party disposing of the shares in the offshore holding company. 7. Page 31 notes that where there is not an adequate treaty in place, double taxation may occur “though taxpayers would presumably avoid structuring transactions in ways subject to such treatment.” The Draft should not assume this is the case. There will be many cases where structures have grown organically, or have been acquired by acquisition, and will not be tax efficient. Requiring the taxpayer to structure themselves in a way to avoid potential double taxation is contrary to the principle of efficiency (tax structuring may impede the commercial requirements). The Draft should emphasise the need for appropriate double tax relief in treaties and domestic law. 8. Page 32 refers to the “2001 UN version”. Should this be the 2011 version? 9. Model 1 (taxation of a deemed direct sale by a resident) taxes an offshore transfer by deeming the resident company to have sold and reacquired its assets. This may give a satisfactory result where the vendor owns 100% of the local subsidiary. However what happens if there is say a 60/40 joint venture and the 60% partner sells their share. The deemed sale rules will be triggered on hundred percent of the gains thereby reducing the value of the subsidiary to the joint venture party which had not benefited from the sale. It is true that conditions to deal with such a situation could be included in the joint venture agreement but it is bad policy to require taxpayers to address this issue rather than ensure that the tax rules do not create this effect. Consideration should also be given to how the provision applies where the offshore investment vehicle is an open ended fund. Units will be issued and repurchased on a regular basis. It would not be equitable to impose a tax charge which would affect all the continuing unit holders at the date the 50% ownership test was triggered. 10. Model 1 can also have the effect of taxing movable property and even gains on liabilities. Assume that the vendor wishes to sell a subsidiary in country A which is held through an intermediate holding company. The subsidiary in country A has immovable property standing at a gain of 100 and movable property standing at a gain of 50. Model 1 will treat the subsidiary as having sold and reacquired all its assets and therefore realised a gain of 150 – even though only 100 of the gain is attributable to the immoveable property. The Draft should either provide for an apportionment or valuation mechanism to avoid this result or explain why there is no policy inconsistency in taxing the gain on the movable property in this case. 75 11. Page 40 notes that under Model 1 the company bearing the tax does not receive the actual proceeds of the disposal but that it is expected “that the parties (particularly purchaser) would take steps to ensure that the local asset-owning entity had sufficient funds to discharge its tax liability.” Given that it is the vendor which would have made the gain one would expect that the purchaser would require the vendor to put the target company in funds, rather than bearing the cost itself. The paper should, however, discuss more the difficulties which could arise from having to finance the tax liability - for example would injecting cash through equity result in stamp duty or similar tax charges; if a loan was made and later written off, would this create taxable income? 12. Page 50 has a table showing when the full gain or a pro rata amount of the gain is to be taxed under Model 2. If the shares derive more than 50% of their value from immovable property the whole of the gain is taxable. However in such a way a chargeable gain could arise on movable property and even property not situated in the relevant territory. For example assume that an MNE has an intermediate holding company which has two subsidiaries in countries A and B and more than 50% of the value in the shares in the intermediate holding company is derived from real estate in country A. According to the proposed rules, country A is entitled to tax 100% of the gain realised on a sale of the intermediate holding company. However part of that gain could be attributable to an increase in value in the subsidiary in country B or from movable property in country A. It would therefore be better to apply the apportionment formula in all cases so that it is only the part of the gain attributable to the immovable property which is taxable. 13. Further clarification should be given on how the apportionment mechanism works. As drafted the gain is multiplied by a fraction which is found by dividing the value of the shares derived from the immovable property by the total value of the interest. Assume that the value in the subsidiary in question has three components: an immovable asset worth 50, a movable asset worth 50 and a liability of 50. The value of the shares in the subsidiary would be 50 and so the formula would be calculated as 50÷50 which means that 100% of the gain would be taxable. However the gain may in reality be attributable to the movable, not the immoveable, asset. The denominator in the fraction (C) should therefore be the sum of all the assets rather than the total value of the interest. In this case, in the example given, the fraction would be 50÷100 or 50%. This corresponds more to reality as 50% is an estimate of the proportion of the gain which is derived from immovable property. 14. Page 55 notes that one of the disadvantages of Model 2 is the difficulty of enforcing the tax charge on the non-resident vendor. Could this be avoided by having a secondary rule which imposed a tax charge on the resident company if not paid by the vendor within a prescribed period? 15. Model 2 is only triggered if a certain percentage of the value of the shares sold relates to immovable property in the source state. This could mean that even where the value in the local subsidiary comes 100% from such assets, this could be swamped by other investments held by the offshore entity which is sold. Consideration could be given to focusing the test on the percentage value of the local entity which is made up of immovable property in that country rather than the percentage value of the off shore entity which is sold. 76 Answers to the questions 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? As set out above, a deeper analysis is needed of the reason for taxing offshore gains, why this should be limited to “immovable” property, and whether or not it is appropriate to extend taxation to assets such as government licenses. Such an analysis is needed to help governments decide the most appropriate policy for taxing gains, to give an overall cohesion to international practice and to give investors clarity about what category of gains may become taxable. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? The operation of Models 1 and 2 are clear. However as set out above they could be more focused. Both Models can result in taxation on movable property (which seems to be against the rationale for focusing on immovable property). Model 2 could lead to gains on property situated outside the source jurisdiction being subject to tax (as could Model 1 in exceptional cases). The Models should either be amended to exclude these possibilities or there should be an explanation of why this is acceptable given the stated policy. 3. Is the definition of an offshore indirect transfer of assets satisfactory? Yes. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? See answer to 1 above. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? See answer to 1 above. More detail and justification is required. The Draft should also discuss the need for countries to consider their infrastructure requirements and policy and the potential effect on inward investment of taxing offshore gains. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? See answer to 1 above. 7. Are there other implementation approaches that should be considered? There should be more focus on ensuring that only the part of the gain attributable to the immovable property is taxed. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? Model 1 has the advantage of being easier to enforce and the benefit that the assets can easily be rebased following a sale to avoid a future double tax charge. However the downside is that, where there are multiple shareholders and not all sell their 77 shares, the tax charge will economically be borne by the remaining shareholders. Conversely, no tax charge would arise if the transfer does not result in a change of control but the vendor nevertheless realises a substantial gain. It also has the problem that the entity paying the tax is different from the one which realises the taxable gain. Model 2 has the benefit of taxing the proportionate amount of any gain in the hands of the actual vendor. However more work is required on avoiding double taxation on future sales, especially if a future sale is made at a different level in the holding chain. Consideration should be given to combining Model 2 with an anti-avoidance rule to impose the tax on the source state company where the vendor fails to pay. 9. Are the complexities in the taxation of these international transactions adequately represented? There are various comments in the Draft that oversimplify the situation and also make it look as if offshore sales are (largely) driven by tax avoidance. The paper should recognise that there are many commercial reasons why offshore sales may take place and efficient tax structuring may or may not be one of the factors. It should consider in more detail how taxation of offshore gains fits in to the source country’s overall tax policy and its requirements for infrastructure investment. The paper should also discuss the potential for the taxation of offshore indirect transfers to restrict the ability of MNEs to restructure for commercial purposes and the need to ensure that there are adequate rules to prevent double taxation. Finally there should be more discussion about how to ensure that the rules only tax gains associated with the immovable property in accordance with the policy rationale. It would be useful for the paper to contain a set of high level guidelines which address the different rights and duties of MNEs and governments. 78 Platform for Collaboration on Tax By email to: taxcollaborationplatform@worldbank.org 20 October 2017 PwC’s Comments on the Draft Toolkit to help developing countries tackle the complexities of taxing offshore indirect transfers of assets PricewaterhouseCoopers International Limited, on behalf of the Network Member Firms of PwC (PwC), thanks the Global Tax Platform for the opportunity to provide comments on the draft toolkit to help developing countries tackle the complexities of taxing offshore indirect transfers of assets (OITs). We recognise many of the issues being raised in the draft toolkit and their significance to both developing and developed countries. Countries’ approaches to date have differed widely, in terms of both the legal approach taken and which assets are covered; we agree that greater coherence could help enhance tax certainty. In overview, we think that: • some of the recommendations risk disturbing the current economics of investment decisions with insufficient evidence of their impact and suggest more research is carried out • the question of whether there should be an international standard or what would constitute best practice as regards the extent of taxing OITs is one that needs to be more extensively discussed between countries before a methodology for doing so consistently can be considered • there is a lack of clarity in some of the recommendations, which we would like to see explained in more detail, particularly as to how countries should prevent economic double taxation occurring • there would seem to be an inherent risk of generating more uncertainty in encouraging countries to interpret the definition of ‘immovable property’ in different ways according to their local needs 79 • double taxation is considerably more likely, with its negative impact on investment, unless there is clear consensus and agreed guidance on the interpretation of existing treaties and the resolution of differences on the source versus residence debate in this area, and • the conclusion that Model 1 would be preferable to Model 2 does not seem to be proven. In a deemed disposal of assets under any of the provisions in a given tax regime, valuation of those assets is in practice one of the more difficult issues faced. Consideration should be given to providing clear guidance on how to perform such valuations, particularly if that country imposes a deemed disposal of assets in relation to an OIT. Numerical examples would considerably clarify some of the points being made. The purpose of the toolkit in relation to those countries that already have tax regimes which deal with OITs is also unclear. For example, where a country currently relies on a general anti-avoidance rule (GAAR), is the Platform’s recommendation that the country continue with that rule, withdraw it in favour of one of the alternative approaches (the draft recommendation being a rule based on Model 1 rather than on Model 2) or have both in future? Our more detailed comments are summarized below in response to the specific questions asked. 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? While the value of an asset takes into account all future potential net earnings from it, taxing the OIT could be a form of double taxation and it is arguably better to apply taxes – withholdings on royalties, export duties, profit taxes/ mining taxes etc – solely at the point of extraction, realisation, etc. Where earnings or similar amounts are untaxed, it is typically because of some tax holiday or up-front incentive offered by the source country, which would be better removed than ‘corrected’. When an investor is thinking about acquiring an ownership interest there is often a significant up- front cost and risk involved, which could be exacerbated by an additional tax charge at that point. It should not be forgotten that there are often significant restoration or rehabilitation costs at the end of the life of an ‘extractive’ asset. Unless you’re talking about a cash-flow tax model rather than taxing ‘gains’, it is hard to justify in accounting or tax terms the economic argument that if a future sale will be “ taxable in the same jurisdiction as today’s sale, and at the same tax rate, then the total nominal (undiscounted) revenue raised from the capital gains tax over time will be zero: that is, the same as if there had been no sale, or no capital gains tax.” 80 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? It should be remembered that the value of some assets is significantly affected by matters which are clearly separate from the source country. An obvious example is an asset deriving much of its value from a global brand. It does not seem appropriate to include such added value in the OIT regime and consideration should be given as to whether amounts derived from certain elements of an asset’s value should be excluded. The stated reasons for ignoring onshore transfers and the different domestic rules often applied to reconstructions, reorganisations, etc is that they don’t offer the opportunity for ‘significant abusive avoidance’ when beneficial ownership is actually substantially transferred. This seems incongruous when it concludes that more than an anti-avoidance device is required. It would also seem to be critical to the effectiveness of the measures to provide for clear exemptions to cover commercial reorganisations etc. 3. Is the definition of an offshore indirect transfer of assets satisfactory? There is little that would appear to depend solely on the definition of an offshore indirect transfer of assets. The draft toolkit doesn’t really focus on the position of individuals as taxpayers, though the principles applied by countries may well apply equally to them as to corporate and other taxpayers. The nature of assets covered potentially brings into scope such individuals, even though in practice it is likely to be multinational organisations that are affected, at least in relation to indirect transfers, by their very nature. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? There is some recognition that double taxation would result from the actions put forward, particularly in the recommendation that there should be a deemed disposal of assets and liabilities in the source country when there is a significant change in control in the ownership chain, with limited detail on how this might be reduced. The paper appears to justify this to some extent by the economic relevance of dealing with non-taxation although it doesn’t provide evidence on the economic effects on investment. Unless there is broad consensus in relation to the source and residence issues in this debate, multiple taxation could go unrelieved. It is not yet clear whether such consensus will be achieved. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? The draft toolkit helpfully recommends that a taxing right cannot be supported without appropriate definition in domestic law of the assets intended to be taxed and without a domestic law basis to assert that taxing right. However, it postulates that arguments for limiting the scope to assets which are immovable are unclear, and puts forward theoretical economic arguments for an extension. It is uncertain whether there is a broad consensus about some of the proposed items which might be 81 included but the proposal that countries should have the ability to define it according to their local requirements does not appear principled. A recent judgment of the Court of Justice of the European Union (CJEU) on 12 September 2017 highlighted the potential difficulty where the parties to a double tax treaty have different definitions of terms. The CJEU determined that even though Article 3 paragraph 2 of the double tax treaty between Austria and Germany sets out a rule of interpretation according to which a term not defined by that treaty must be given the meaning it has under the tax law of the State applying it, a particular concept – in this case “debt-claims with participation in profits” which the countries interpreted in different ways - must be interpreted according to the methods proper to international tax law (C-648/15 - Austria v Germany). 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? The concept of ‘location-specific rent’ is one has that been put forward in relation to various discussions concerning global tax policy, including tax incentives and transfer pricing. These discussions have not been conclusive so far as to the degree of importance to be attributed to such a concept. Reaching a consensus on its relevance and how to value it, might take the timing of this draft toolkit beyond its anticipated deadline. 7. Are there other implementation approaches that should be considered? Consensus and consistency of interpretation and application are critical. If one considers clarification of the existing international standard an option together with something broadly aligned with the underlying principles of Model 1 (deemed disposal) or Model 2 (taxing the non-resident seller), these would seem to cover the main choices available. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? The draft toolkit leans in favour of the ‘deemed disposal’ method for reasons, it states, of the relative ease of enforceability, and the logic and simplicity of basis adjustment it implies. It is likely that, if it were considered necessary to have an OIT rule, a number of commentators may favour Model 2 rather than Model 1, since that would mean the entity paying the tax would have greater access to the necessary funds to do so. There are a number of areas in which greater clarification would be needed of the final conclusion, particularly if an OIT rule is recommended. For example, various exclusions, exemptions or safe harbours would help to make any rule more administratively manageable for both taxpayers and tax administrations. Numerical examples would help clarify at least what has to be calculated and when, if not how any required valuation would be carried out. 82 9. Are the complexities in the taxation of these international transactions adequately represented? The Samples for Model 1 and Model 2 included in the draft toolkit are very generic. The draft toolkit notes that there remain potential issues with existing treaties and that domestic implementations would, in particular, need to: • take into account the specific legal tradition and system, as well as the political and administrative structure and fiscal policies of the country concerned (including common concessions that typically apply to remove, reduce or defer the recognition of taxable gains), and • limit economic double taxation so that the same gains are not taxed multiple times in the hands of different taxpayers through realisations of gains on intermediate shareholdings through multiple tiers of indirect ownership (the Samples cover asset transfers being taxed twice by the location country in the hands of the same taxpayer). These two issues are examples of particular complexities which are not fully dealt with in the draft toolkit. While it may be said that they are ‘represented’ in the draft toolkit, they are not resolved and remain areas which countries might ignore or for which an incoherent approach may result. The process for producing this toolkit, and other similar toolkits, is a little unclear. It does not appear to be a consensus document and discussions with individual countries are ongoing, it states. It is presumably then published more to stimulate discussion at this stage. It postulates a degree of acceptance of the rationale for its recommendations by drawing on other discussions, while it would seem to be more appropriate to have gauged reaction directly and in advance. We recommend, and are keen to participate in, further discussions involving business and wider stakeholder groups on the issues involved in this debate. If you would like to pursue in more detail some of the points to which we refer above please contact the undersigned or any of those people whose contact details are provided below. Yours faithfully, Stef van Weeghel, Global Tax Policy Leader stef.van.weeghel@pwc.com T: +31 (0) 887 926 763 83 PwC Contact Country Email Edwin Visser Global edwin.visser@pwc.com Philip Greenfield Global philip.greenfield@pwc.com Dave Murray Global david.x.murray@pwc.com 84 85 86 87 88 89 90 91 92 Dr. Sergio Guidas Sr Financial Director, Certified Public Auditor (Italy) Comments on a draft toolkit on ‘The Taxation of Offshore Indirect Transfers’ of assets. Questions to consider 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? Yes. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? Yes. 3. Is the definition of an offshore indirect transfer of assets satisfactory? Yes. Probably it is agreeable not to undertake there heavier descriptions of many details about transfer price practices. Indeed, I would suggest some links to Oecd ‘Common reporting standard’s’ recent developments (e.g. to clarify how a substantial part of capital gains are attributable to value- enhancement provided from abroad, etc.). 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? Yes. Probably it could be increased the attention to intangible assets in their widest acceptation. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? Yes. I think it would be useful to establish links with cadastral findings and local administrative authorities’ registries, too. This could be helpful also for a much effective way to approach the extensiveness of the concept of “ permanent establishment “in the country in which the asset is located, I think. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? Two points: a) Basically the impact of corporate income tax on the level of capital investment is measured through the user cost of capital (the pre-tax real required rate of return on an investment) and a firm will invest up to the point at which the marginal product of capital is just equal to the cost of capital so that the project just breaks even. The impact of tax on the cost of capital is measured by the effective marginal tax rate and normally multinational firms face a choice between alternative locations of production and choose that location (or locations) offering the highest post-tax profit. Many investments earn economic rents; that is, profits in excess of a market return, which can be characterized as either firm-specific (or mobile) or location-specific. Investment generating mobile rents (arising from factors such as management know-how, a brand or a businesses' possession of a particular technology) can be moved from one jurisdiction to another. Location-specific rents may arise from exploitation of natural resources, existing fixed investments (such as factories), agglomeration (where businesses obtain benefits from co-location such as economies of scale), 93 attractive local infrastructure, public services and institutions or consumer preference for domestically produced over imported goods. For a mobile rent, source-based taxes can reduce investment: investors will simply shift the investment to a lower tax jurisdiction so they can receive a greater share of the rent. In contrast, a source-based tax on a location-specific rent will not distort investment decisions. The draft shows how,despite the rules, in many cases multinationals are able to shift at least part of their profits to countries with low statutory tax rates. b) << In both model treaties, as seen above, a taxing right arises when over 50 percent of the value of the transferred stock or interest derives from immovable property in the location country.>> In my opinion, to indicate a limit (even a lower one) is not the best way to follow if we want to catch the transfers of value, because it can easily be circumvented. As said “ in order to determine whether the value of the interest is principally derived from that immovable property, a comparison is ordinarily required to be made of the value that the immovable property (relevant asset) bears to the value of all the property owned by the entity (all assets”, but the contribution can be given in a lot of different ways, often not immediately and directly quantifiable. Hence could be appreciable the reference to the most recent developments in Common reporting standards I mentioned above, too. 7. Are there other implementation approaches that should be considered? Maybe besides the efforts to adopt not only tax treaties but especially international-wide rules, it should be prosecuted more effectively the objective of harmonizing tax rate, too. Often low tax jurisdictions are not so collaborative both in principles and in facts boycotting inter-nation equity. 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? I’m convinced that the country in which the asset is located has primary taxing rights on an indirect transfer of its ownership that takes place outside the location country. So my answer is that such a preference is appropriate. I agree “the rule should not apply only as an anti-avoidance device to combat “double non-taxation,” but rather should constitute a fundamental aspect of the international tax architecture—in which rights to tax gains arising on such assets when subject to an indirect transfer offshore would be primarily allocated to the location/source country”. Indeed, we must always keep in mind the risks that designing and drafting legislative provisions that limit economic double taxation could paradoxically transform in tax avoidance opportunities. 9. Are the complexities in the taxation of these international transactions adequately represented? I think so. By reading the many examples and considering the many graphics and statistics, everyone can immediately understand the most common situation, as well as the “holes” in some countries’ legal environments, especially in the near past. Very important are the comparisons with best practices, too and both the links and the differences between BEPS and UN statements. Finally, I think that widening as much as possible the implementation of withholding prescriptions (for instance imposing of either a final or non-final nature and with particular care to non-resident seller’s gains) would be desirable for an increasingly effective struggle to tax avoiding strategies. Dr Sergio Guida Sr Financial Director, Certified Public Auditor (Italy) 94 October 20, 2017 VIA ELECTRONIC SUBMISSION: taxcollaborationplatform@worldbank.org Mr. Vitor Gaspar, Director Fiscal Affairs Department International Monetary Fund Mr. Pascal Saint-Amans, Director Ms. Grace Perez-Navarro, Deputy Director Centre for Tax Policy and Administration Organisation for International Co-operation and Development Mr. Alex Trepelkov, Director Financing for Development Office United Nations Ms. Deborah Wetzel, Senior Director, Governance Global Practice Mr. Jim Brumby, Director, Public Service and Performance, Governance Global Practice The World Bank Re: Comments on Discussion Draft: The Taxation of Offshore Indirect Transfers – A Toolkit Dear Sirs and Madams: We are writing to provide the comments of the Silicon Valley Tax Directors Group (SVTDG) on the Discussion Draft released on August 1, 2017 by the Platform for Collaboration on Tax. We appreciate the opportunity to comment. A list of SVTDG members is appended to this letter. The SVTDG represents U.S. high technology companies with a significant presence in Silicon Valley. The SVTDG promotes sound, long-term tax policies that allow the U.S. high tech technology industry to continue to be innovative and successful in the global marketplace. As companies with operations and investments all over the world, we have an interest in any tax policy development that could have a 95 significant impact on the way we do business, particularly one that is liable to affect adversely normal evolution in our business structures. SVTDG member companies support the work of the Platform to help tax administrations of developing countries improve their capacity to administer tax laws and treaties. We believe that stronger tax administrations provide benefits for governments and taxpayers alike, by providing greater certainty and consistency in the application of the law. The Discussion Draft on this particular topic, however, raises a number of important concerns for us as global companies engaged in a broad spectrum of cross- border transactions. This letter provides our general comments on the Discussion Draft, while the attached Appendix I sets forth our detailed comments. General Comments The press release accompanying the Discussion Draft presents nine questions for stakeholders to consider. We believe that these questions are well framed and they serve to identify most of our concerns regarding the Discussion Draft. As discussed below and in our more detailed comments, we respectfully submit that the Discussion Draft in its current form falls short on each of these measures. We also believe that there is another important question that should be considered: What is the likely effect of the measures proposed by the toolkit on the foreign investment sought by many developing countries? The Platform clearly articulates its rationale for proposing to tax offshore indirect transfers of assets. The press release accompanying the Discussion Draft describes it as a “toolkit designed to help developing countries tackle the complexities of taxing offshore indirect transfers of assets, a practice by which some multinational corporations try to minimise their tax liability.” The draft toolkit Intro duction further characterizes such transfers as involving the “significant abusive avoidance of the taxation of gains on local assets.” These statements are based on two presumptions: (1) that there is a tax due on gain from such transfers, and (2) that companies commonly engage in indirect transactions to avoid that tax. We submit that neither of these presumptions is correct. In fact, as the Discussion Draft itself implicitly acknowledges, in all but very rare cases, neither domestic laws nor treaty provisions currently permit the taxation proposed by the Discussion Draft. Even if tax were due, the great majority of indirect asset transfers occur in connection with routine acquisitions and dispositions within corporate groups in the ordinary course of business, and the indirect holding of assets within a multinational group is merely a feature of modern corporate management structures, not a tax avoidance scheme. Therefore, the Discussion Draft is advocating significant policy and legal changes that should be considered and evaluated as such, and not on the rationale that tax is being avoided at present. 96 We are concerned that the Discussion Draft proceeds from the assumption that the country in which an asset is located should impose tax whenever a foreign entity owning the asset is sold. This is consistent with the Platform’s goal of significantly increasing developing country tax revenues from foreign investors, expressed in its original Concept Note. However, the three principles selected for discussion - inter-nation equity, efficiency, and political economy - seem to be little more than a restatement of this goal. Taking as a given the goal of increasing taxation at “source” on indirect transfers by foreign investors has unfortunately also made the Draft’s analysis of these and other policy and administrative issues less complete and balanced than it might otherwise be. For example:  The Discussion Draft lacks the balanced analysis of the considerations that typically underlie policy decisions about allocation of primary and secondary taxing rights on international transactions.  It also contains an inadequate analysis of the double taxation implications of its recommendations, failing to fully acknowledge the domestic and international double taxation risks that could arise, including both juridical and economic double taxation, and the extensive coordination and consensus that would be needed to address them.  In addition, there is no consideration of the negative effect that the proposal would have on foreign investment or of its other potential economic effects. The Discussion Draft relies too much on “economic” theory, equating direct and indirect ownership of assets, without acknowledging the real, practical, and legal distinctions between the two, all of which should be taken into account in making policy determinations about international tax law.  The Discussion Draft’s recommendations that the scope of its proposed measures be as expansive as possible – perhaps even taxing access to domestic markets – creates an unacceptable level of uncertainty regarding their scope. This will surely lead to inconsistent implementation and spawn new cross-border controversies, rather than ensuring the certainty and consistency the Discussion Draft aims to promote.  The Discussion Draft also cites and relies on current political considerations to an extent that seems inappropriate for a major proposal. Such considerations should not be confused with tax policy and administration principles.  The Discussion Draft overstates the potential “abuse” nature of offshore indirect transfers, and it fails to identify simpler, more effective solutions to problems. Its citation of three cases featured in the media does not reflect the wide array of situations that would be affected. The 97 Discussion Draft then recommends offshore indirect transfer taxation rather than more conventional, administrable solutions to the cited problems that pose less risk of double taxation or other impediments to cross-border investment (e.g., appropriate enforcement of transfer pricing, CFC rules for “round-tripping” situations, etc.). Finally, in our view, the Discussion Draft vastly understates and fails to address adequately the complexity of the issues involved in trying to tax offshore indirect transfers. The Draft’s focus on highly simplified, stylized examples fails to alert tax policymakers and tax administrations to these issues, especially those with relatively limited experience, as in many developing countries. A more robust and balanced analysis, including a consideration of the potential effects on investment, is needed to better assist tax policymakers and administrations in the formation of well-informed and responsible measures. In light of these numerous concerns, we respectfully submit that the proposed Toolkit should be withdrawn. We hope that our comments will prove informative to your deliberations on this matter. Sincerely, Robert F. Johnson Co-Chair, Silicon Valley Tax Directors Group 98 Appendix I I. INTRODUCTION AND SUMMARY A. Description of SVTDG and its interest in the topic 1. The SVTDG represents U.S. high technology companies with a significant presence in Silicon Valley that are dependent on R&D and worldwide sales to remain competitive. The SVTDG promotes sound, long-term tax policies that allow the U.S. high tech technology industry to continue to be innovative and successful in the global marketplace. As companies with operations and investments all over the world, we have an interest in any tax policy development that could have a significant impact on the way we do business, particularly one that is liable to affect adversely normal evolution in our business structures. B. Identification of key points to be addressed 2. In our general comments, we address certain over-arching concerns about the Draft’s recommendations, including its apparent emphasis on “abuse” as a primary policy driver in this area, its acceptance of anecdotal political rumblings as the equivalent of policy analysis, its reliance on vague economic theories divorced from any link with the legal framework of international taxation, its failure to evaluate the merits of OIT taxation in comparison to other, more conventional tools, and its failure to address adequately the serious issues of double taxation, complexity, and potential adverse effects on inbound investment associated with its recommendations. 3. Section III of our comments considers the economic rationale the Draft puts forward in support of its recommendations. The Draft’s statement that OIT taxation is needed to ensure taxation of corporate earnings that would otherwise go untaxed is found to be inaccurate, or at best misleading, for a number of reasons. The three “principles” cited by the Draft in support of the allocation of taxing rights to the country of location of the underlying asset (i.e., inter-nation equity, efficiency, and political economy) are analyzed, and the Draft’s conclusions are found to be flawed. For reasons we explore, the Draft is misleading regarding the existence of an international consensus in favor of OIT taxation, it ignores the implications of the BEPS changes on the calculation of where and when taxation will occur, it overstates any “abuse” aspect to this policy decision, it uses dangerously broad and ambiguous language in its recommended expansion of the concept of “immovable property” to be subject to OIT taxation, it fails to take into account the legal and economic differences between direct and indirect investment in assets, it fails to analyze less disruptive alternative tools to address the concerns raised, it lacks a robust analysis of the potential risks of double taxation and compliance challenges and their potential negative effects on inbound investment, and it cites a reliance on political considerations divorced from any underlying policy considerations. 4. Section IV of these comments outlines deficiencies in the Draft’s analysis of treaty considerations, noting the mischaracterization of current treaty practice on a number of points. 5. Section V addresses the two potential models for OIT taxation outlined in the Draft, taxation of a deemed direct sale by a resident (Model 1) and more conventional approach of taxing the actual gain realized by the non-resident seller on the disposition of shares in a prescribed category of cases (Model 2). Model 1 raises concerns about compliance difficulties relating to the deemed transferor’s knowledge of the actual transaction, double taxation risks (both domestically within the country of the deemed sale and internationally), inequitable results where the actual seller is not the sole owner of the indirect investment, consequential taxation effects on assets other than the supposedly targeted immovable property, and potential treaty violations. Concerns identified with respect to Model 2 include significant 99 double taxation problems (both domestically in the country where the underlying asset is located and internationally), and substantial compliance complexities and difficulties. 6. Section VI addresses issues of legal implementation of the Draft’s recommendations, including at both the domestic law and treaty level. 7. Section VII identifies a number of concerns the Draft does not address (at all or adequately), including valuation issues, issues involving related party transactions, issues involving dispositions of shares in non-wholly-owned companies, and the treatment of losses. II. GENERAL COMMENTS 8. The Draft is presented as an initiative to assist developing countries in preventing the avoidance of tax on certain transfers, which are characterized as involving the “significant abusive avoidance of the taxation of gains on local assets”.1 In fact, the Draft advocates significant policy and legal changes to the current law of most jurisdictions and the provisions of most tax treaties with effect far beyond any abusive transactions, and it should be considered and evaluated on that basis. 9. The Draft acknowledges at the start that its goal is to enable the country in which an asset is located to impose tax when a foreign entity owning the asset is sold. While this is consistent with the significant increase in developing country tax revenues from foreign investors which was identified as a general goal of the Platform in its original Concept Note, this goal needs to be balanced by a full and objective evaluation of any new measure the Platform may recommend, including both its likely effect on investment and its administrability. 10. The given goal of increasing taxation at “source” on indirect transfers by foreign investors unfortunately makes the Draft’s analysis less complete and balanced than it might otherwise be. 11. The Draft cites and relies on current political considerations to an extent that seems inappropriate for a major tax policy proposal. At the same time, while the Draft indicates that the drafters consulted with selected country officials (to be named in the final Toolkit) and cites media reports of several particular transactions, the Draft was prepared by the Secretariat and staff of the Platform’s four international organization members and contains a caution that “[n]either this draft nor the final report should be regarded as the officially endorsed views of those organisations or of their member countries.” This raises the question of whose views are being reflected in the work of the Platform. 12. The Draft does not contain a balanced analysis of the considerations that typically underlie policy decisions about allocation of primary and secondary taxing rights on international transactions. It relies too much on “economic” theory, equating direct and indirect ownership of assets, without acknowledging the real, practical, and legal distinctions between the two, all of which should be taken into account in making policy determinations about international tax law. 13. Moreover, the Draft contains an inadequate analysis of the double taxation implications of its recommendations, failing to fully acknowledge the domestic and international double taxation risks that 1 Draft, Introduction, page 12, footnote 5. 100 could arise, including both juridical and economic double taxation, and the extensive coordination and consensus that would be needed to address them. 14. There is no consideration of the negative effect that the proposal would have on foreign investment or of its other potential economic effects. . 15. The Draft vastly understates and fails to address adequately the complexity of the issues involved in trying to tax offshore indirect transfers. Its focus on highly simplified, stylized examples is misleading to tax policymakers and tax administrations, especially those with relatively limited experience, as in developing countries. In addition, it overstates the potential “abuse” nature of offshore indirect transfers, and it fails to identify simpler, more effective solutions to problems. 16. The recommendations contained in the Draft could affect a multitude of transactions that take place as part of routine business expansions or contractions, or that result from business-driven internal restructurings. Simply citing a small number of cases featured in the media does not reflect the wide array of situations that would be affected. 17. The Draft recommends offshore indirect transfer taxation rather than more conventional, administrable solutions to problems cited which pose less risk of double taxation or other impediments to cross-border investment (e.g., appropriate enforcement of transfer pricing, CFC rules for “round- tripping” situations, etc.). III. ECONOMIC RATIONALE A. Discussion of need for capital gains taxation 18. The Draft says capital gains taxation is a way to capture changes in earnings that would otherwise be untaxed. This statement is incorrect, or at the very least, misleading, for several reasons. 19. First, gain on shares may reflect accumulated earnings that have already been taxed by the “source” country at the corporate level and can be taxed again by that country at the shareholder level upon distribution. Even where share gain reflects unrealized appreciation in an underlying asset, that appreciation may be largely due to the expectation of a future increase in earnings from the asset, which can likewise be taxed by the “source” country at the corporate level upon realization and at the shareholder level upon distribution. Share gain can also reflect asset appreciation which can be taxable by the “source” country upon disposition of the asset, as part of corporate earnings which are taxable upon realization at the corporate level and upon distribution at the shareholder level. 20. The statement also ignores the potential for there to be residence country taxation of the share gain realized by the shareholder; the Draft inappropriately dismisses that very real possibility and thus fails to acknowledge the interests of the residence country and to provide measures for avoiding double taxation. 21. The Draft also wrongly states that the cumulative revenue raised from capital gains tax in nominal (undiscounted) terms on consecutive sales of an asset is zero due to basis adjustments, and that taxing offshore indirect transfers (OITs) is necessary to avoid foregoing a timing gain in taxing 101 otherwise untaxed income.2 Cumulative revenue raised from capital gains tax in nominal (undiscounted) terms is equal to tax rate times capital gains realized; any basis adjustment by the source country after the first sale simply prevents double taxation by that country of built-in gain already taxed on first realization, but it does not prevent subsequent taxation of the later-accruing gain. Triggering a “source” country tax on an offshore indirect transfer of asset simply accelerates tax that country would subsequently receive upon a direct transfer of the asset; it isn’t needed to prevent loss of taxing jurisdiction over the asset’s built-in gain. 22. The statement that “this timing effect is a consideration of some importance for governments of lower income countries that face constraints on their borrowing capacity” implies that countries would be able to tax earlier if they didn’t have to wait for realization events occurring within their jurisdiction. While that may be true, a desire to accelerate taxation with respect to certain assets owned by foreign investors is hardly a sufficient justification for taxing offshore indirect transfers. B. Principles for Allocation of Taxing Rights on OITs 23. The Draft cites 3 “principles” in support of the allocation of taxing rights to country of location of underlying asset: inter-nation equity, efficiency, and political economy. 1. Inter-nation equity: 24. The Draft asserts that there is a broad consensus on allowing country in which asset is located to tax “onshore direct asset transfers”, “even though the seller may be non-resident”, and says this supports the view that a country in which an asset is located should be entitled to tax gains associated with it.3 There are a number of problems with this assertion. 25. First, the statement is unclear, since “onshore” transfers are generally defined as transfers by residents (or non-residents with a local PE), so rules relating to the taxation of “onshore” direct asset transfers provide no signal as to consensus regarding whether the country of an asset’s situs should be allowed to tax “offshore” direct asset transfers (i.e., transfers by nonresidents). Even if one read the sentence as referring to some consensus regarding the situs country’s right to tax offshore direct asset transfers, such a consensus exists (as reflected in the Capital Gains articles of tax treaties) only in respect of direct transfers of specifically defined categories of immovable property, property forming part of a PE, and in some cases substantial shareholdings in local companies. 26. The Draft cites a consensus that dividends received by a parent company abroad may be subject to withholding tax by the payor company’s country as supporting the view that this country should also be able to tax foreign investors’ capital gains “associated with a domestic source”. This statement ignores the existing international consensus that prohibits source country taxation of stock gains on offshore transfers except where the stock value is substantially attributable to source country immovable property and where such taxation is permitted by the applicable tax treaty, if any. Moreover, the statement also ignores the growing international consensus around prohibiting source country taxation of dividends paid to substantial corporate shareholders. 2 Draft, page 16. 3 Draft, page 18. 102 27. The Draft goes on to suggest that taxation of OITs is justified by the possibility that asserted avoidance opportunities diminish a source country’s effective power to tax future earnings. This suggestion is ill-considered as a policy and legal matter, given that there is no generally agreed right to tax potential future earnings currently. It also ignores the BEPS Project’s measures that will enhance countries’ abilities to appropriately tax earnings. In addition, it ignores the likelihood that enhanced enforcement of taxation of earnings is more straightforward and administrable to achieve and less likely to produce double taxation than trying to tax OITs. 28. The Draft further cites a consensus on allowing the country where immovable property is located to tax gain on that property as justifying a recommendation to tax OITs. This statement ignores the consensus in existing models which significantly limits source countries’ right to tax OITs, even in cases where there is gain inherent in underlying local immovable property (e.g., allowing such taxation only where such underlying property represents more than 50% of the value of the shares, and with recommendations to further limit that taxation in a variety of circumstances, including where the stock is publicly traded, where the transfer occurs as part of an internal corporate reorganization, where the transferor is a tax exempt entity, etc.). 29. The statement that immovable property can be seized and that this therefore makes taxing OITs pragmatic appears to be an economist notion, and it ignores the legal and practical distinctions between tax liability of the entity holding the immovable property and tax liability of entity’s shareholder engaging in the OIT. First of all, the fact that the immovable property is held by an entity, the shares of which have been sold by the taxpayer demonstrates that seizure of the property can only be effective where the taxpayer’s liability has been effectively shifted as a legal matter to the buyer (e.g., through the imposition of a withholding requirement), and even then only where the buyer effectively has 100% ownership indirect ownership of the underlying property. The disjunction is particularly acute where the selling shareholder is not the sole owner of the entity holding the asset in question, because then a seizure of the immovable property would economically disadvantage shareholders who have not sold their interest, an inequitable result. 30. The Draft says that location specific rents (LSRs) can be taxed at up to 100 percent without causing any relocation or cessation of activity. We seriously question the accuracy of this statement, as we doubt investors would bother to make investments that generate LSRs if they are taxed at up to 100%. Furthermore, the statement ignores the likelihood that taxation of OITs will, in the absence of highly complex international coordination, often result in double or multiple taxation of the same gains, potentially exceeding 100% of the LSRs. 31. The Draft’s suggestion that taxing OITs can be a “useful backstop” when implementation of specific taxes on LSRs is imperfect is no substitute for the much more direct solution of improving the implementation of such specific taxes. 32. There are also deficiencies in the Draft’s discussion of the scope of “immovable” property on which taxation by the “source” country is ostensibly justified. Even assuming a source country should be acknowledged to have primary taxing right over gains from direct transfers of local immovable property, the Draft makes minimal effort to address the issue of which indirect transfers, if any, should be treated the same as direct transfers of immovable property. 33. For example, problems exist regarding the potential for other assets to be held by the entity whose shares are being sold and thus to affect the outside gain on those shares. To the extent that 103 source country immovable property reflects less than 100% of the asset makeup of the entity whose shares are sold, the shares (and any gain or loss on them) become less closely associated with that immovable property, so there’s less policy justification to treat an indirect transfer the same as a direct transfer. As a practical matter, this problem is generally compounded where the indirect transfer occurs at a level more than one tier up from the underlying immovable property. 34. Distortions also occur where liabilities exist within the chain of entities through which the underlying immovable property is held. For purposes of determining whether shares of a company derive more than 50% of their value directly or indirectly from immovable property situated in a source country, there is a convention that compares the value of such immovable property to the value of all the property owned by the company without taking into account debts or other liabilities of the company (whether or not secured by mortgages on the relevant immovable property). This can mean that immovable property in which a company has minimal real investment (e.g., due to an offsetting liability) can cause share-level investment in the company to be treated as being disproportionately attributable to source country immovable property. 35. The Draft’s proposed expansion of the “immovable property” concept to cover other categories of assets said to generate LSRs creates additional difficulties. For one thing, the valuation of such assets (e.g., government-granted rights) may be particularly difficult due to lack of a market in them, and to the possibility that they were not acquired for a payment. 36. Moreover, the discussion of what constitutes an LSR is dangerously broad and open-ended. A category of taxable asset founded on a concept that the Draft itself acknowledges “can be difficult to identify in general” does not provide an appropriate foundation for a new basis of taxation that the Draft advertises as designed to provide greater consistency and certainty for taxpayers. The reference to “access to domestic markets” as a potential LSR is particularly concerning, notwithstanding the Draft’s acknowledgement of certain difficulties. 37. The Draft’s proposed legislative language designed to expand the definition of immovable property is also dangerously broad and ambiguous. For example, it refers not only to land or buildings but also to “an interest” in land or buildings, without specifying whether that could be so broa d as to include a mortgage interest on the part of a creditor. It includes a lease of land or buildings (presumably in the hands of the lessee), without any indication of how such an “asset” would be valued. It refers cryptically to “information” relating to certain rights over natural resources, without any explanation of what is meant by that. 38. The Draft asserts that its legislative proposal should be “a minimum domestic law definition of immovable property”, and it actively encourages countries to extend the definition to encourage various rights granted by governments to be a supplier or provider of goods, utilities, or other services, which could raise a multitude of issues (e.g., would this include patents? copyrights? drug approvals? airline landing rights? Etc., etc.) Responsible policymaking does not support making novel taxing recommendations based on such a poorly considered analysis. 39. The Draft also provides an overly weak summary of counter-considerations under the inter- nation equity argument. It fails to acknowledge the weaknesses of its supporting considerations outlined above. It undermines its reference to the possibility that source countries may have chosen not to impose OIT taxation (e.g., in order not to discourage inbound investment) by once again citing their purported inability to tax future earnings in more conventional fashion. 104 40. The Draft cavalierly dismisses the possibility that increased value may reflect managerial and other expertise contributed by the seller or other foreign parties, and it artificially limits its focus on that possibility exclusively to contributions from the jurisdiction of the entity whose shares are being sold, without taking into account the possibility of contributions from jurisdictions higher up within the group where taxing power also resides. 41. In summary, the Draft fails to make a convincing case based on inter-nation equity grounds that taxation of OITs is justified. 2. Efficiency 42. The Draft wrongly asserts that OIT taxation satisfies the principle of good tax design that a tax system should, so far as is practicable, not distort investors’ decisions. As outlined elsewhere in these comments, the Draft’s recommendations raise the specter of significant risks of double or multiple taxation, not to mention substantial uncertainty on issues of scope, implementation, and compliance. These are exactly the sorts of concerns that routinely play a role in businesses’ decisions about whether to invest in jurisdictions, and countries considering the Draft’s recommendations would be ill-advised not to give serious thought to the potential adverse effects these proposals could have on inbound investment. The Draft should not be finalized without a robust and objective analysis of those potential effects. 43. The Draft once again asserts that OIT taxation is a necessary tool to achieve source country taxation of LSRs, without providing any explanation of why more conventional means would not be more efficient with less potential for distortive double taxation. It overstates the neutrality principle that “direct and indirect asset transfers be treated identically for tax purposes”, failing to acknowledge the many different considerations that permeate that differing treatment within tax systems. 3. Political economy 44. The Draft cites as a policy reason for recommending OIT taxation the political pressure for greater source country taxation sparked by a handful of high profile cases of OITs which could not be successfully taxed by the “source” country. It argues this has led to uncoordinated unilateral actions that increase taxpayer uncertainty. 45. Political considerations divorced from underlying policy justifications, however, hardly seem a defensible basis for making “policy” recommendations for law changes. 46. The Draft refers to “highly publicized” cases of natural resource industry or similar OITs,4 without acknowledging that these would likely represent a tiny percentage of the cases likely to be affected by such a fundamental change to international tax rules. In any event, the Draft does not acknowledge the possibility that taxpayer certainty could be achieved through consensus around policy approaches that are not so tainted by poorly informed impressions of the general public. 4. Summary on principles for allocation of taxing jurisdiction on OITs 4 The Draft contains specific descriptions of the Vodafone case in India, the Petrotech acquisition in Peru, and the Zain acquisition in Uganda. 105 47. Given the weaknesses described here, the Draft wrongly concludes that the balance of the arguments on the three principles outlined above favors source country OIT taxation. It then overstates the rationale for expanding this type of taxation beyond indirect interests in immovable property. 48. For example, it improperly cites Article 13(5) of the UN Model (regarding dispositions of substantial shareholdings in local companies) as a justification for expanded OIT taxation, since that is an example of taxation of a direct transfer of a local asset (i.e., it is the stock in the local company, without reference to the location of the underlying assets, that is considered the local asset). 49. The Draft also baldly asserts “the importance to the location country of defining ʽimmovable assets’ in a sufficiently expansive manner”, suggesting the inclusion of “all assets with the potential to generate significant location-specific rents and over which the government can exercise sufficient control to ensure collection,” without making any serious attempt to define that category with any precision or to point to any existing, well-developed definition for the category. As indicated above, we believe this is an irresponsible approach to policy-making and in any event exceeds by PCT’s mandate. IV. TAX TREATIES AND OITs A. The Draft’s analysis of treaty issues relating to OITs is deficient. 50. The Draft’s summary of existing tax treaty practice5 is misleading in a number of respects. 51. First, it wrongly suggests there is a category of “offshore indirect transfer” of assets in cases where the seller has a PE in the country to which the assets are allocated that is taxable by that country. Treaties allow source country taxation of direct transfers of movable assets allocable to a local PE, but there is no common treaty practice of allowing a PE country to tax indirect transfers (i.e., transfers of stock in a company having a PE in a particular country to which movable assets are allocated). 52. Second, the Draft also improperly overstates the extent to which the UN Model authorizes “source” country taxation of OITs of movable assets located in the source country. The UN Model allows taxation only of substantial percentages of stock in a local country company (i.e., direct transfers of such stock) and that without reference to the location of the underlying assets. 53. Third, the Draft does not adequately examine the extent to which OIT taxation of the type authorized by Article 13(4) of the OECD Model (i.e., the provision allowing a State to tax a resident of the other State on gains from the alienation of shares deriving more than 50% of their value directly or indirectly from immovable property situated in the first State)6 can be distortive. For example, it fails to alert developing countries to the problem that determining the percentage of stock value represented by underlying local immovable property by taking into account the gross value of that property, without reference to offsetting debt, may substantially overstate the extent to which the stock value should be considered allocable to the country where that immovable property is located. 5 Box on Draft, page 31. 6 The counterpart in the UN Model is also Article 13(4). 106 54. Moreover, it does not provide any additional guidance on the considerations underlying the Models’ suggestion that countries may wish to consider various exceptions to the OIT taxation otherwise authorized by Article 13(4), including exceptions for:  Gains from the alienation of shares traded on certain stock exchanges;  Gains from the alienation of shares in the course of a corporate reorganization;  Gains on shares held by pension funds and similar entities; and  Gains from the alienation of certain interests in real estate investment trusts (REITs). 55. The Draft’s failure to provide guidance on these very practical issues is particularly noteworthy, given the PCT’s general objective to provide practice implementation guidance to developing countries on issues of relevance to them. 56. The Draft’s discussion of UN Model Article 13(5) (regarding the direct disposition of substantial interests in local companies) is also deficient. It merely suggests that Article 13(5) may be “unnecessary” if a country’s treaties have a sufficiently broad version of Article 13(4), without acknowledging that a widespread international practice of including both provisions in treaties could lead to serious cases of multiple “source” country taxation, as some countries would be taxing gains based on the local residence of the entity whose shares are being disposed of whereas other countries would be taxing the same gains based on the location of the underlying assets. V. IMPLEMENTATION CHALLENGES AND OPTIONS A. The Draft’s two options for taxing OITs 57. The Draft sets forth two models for taxing OITs and expresses a preference for the first model. 58. Model 1 (taxation of a deemed direct sale by a resident) would apply whenever an entity (anywhere in a chain) which derives more than 50% of the value of its shares from location country (Country L) immovable property undergoes a change of (direct or indirect) ownership of more than 50%. In such a case, any such entity would be deemed (for Country L purposes) to have disposed of (and reacquired) all its assets and satisfied (and reassumed) all its liabilities. Country L would then exercise its taxing jurisdiction over the deemed disposition by the (presumably local) entity in the chain with the direct interest in the Country L immovable property, and that basis of that property (and of all relevant shares up the chain) would be adjusted (for Country L purposes) to reflect the recognition of any gain (or loss). 59. Model 2 would involve the more conventional approach of taxing the actual gain realized by the non-resident seller on the disposition of shares in a prescribed category of cases. B. Issues with respect to Model 1 60. The Draft does not address how the local entity which is the deemed direct transferor would necessarily know of the share transfer occurring higher up in the chain, or know whether any such share transfer constituted a change in control. It also does not address how the local entity would fund its tax liability on the deemed disposition, given that it is not an actual seller receiving cash. 107 61. Secondly, if the local entity owns not only Country L immovable property but also other assets, potentially including assets in other countries, the deemed alienation and reacquisition of those assets may give rise to premature Country L taxation and to double taxation risks if there is a subsequent disposition of those third country assets for which there has been no basis step-up in the third country. The Model also incorporates a “cliff” effect, triggering a deemed sale of 100% of the local entity’s assets whenever 50% of the entity’s value is derived from assets in the nature of Country L immovable property. The deemed sale and reacquisition could also trigger currency gains and losses at the level of the local entity. 62. Third, if the entity whose shares are being sold is not wholly owned by the shareholder disposing of the shares, the triggering of gain and imposition of tax liability in Country L on a deemed disposition could result in a diminution in the value of shares held by non-selling shareholders, thereby violating horizontal equity principles. 63. The potential triggering of a deemed disposition by a local entity upon the change of control in the shares of a foreign parent may raise treaty nondiscrimination issues if there is no similar triggering of a deemed disposition by a local entity upon the change of control in the shares of its local parent. 64. The Draft acknowledges the likelihood that there would be international double taxation of the event, given that Country L would be taxing a deemed asset disposition by the local entity and the country of residence of the actual seller of shares further up the chain would be taxing the gain realized on the actual share sale. However, it does not propose any solution to that problem. 65. The Draft also acknowledges, without any real response, that Model 1’s deemed asset disposition could be seen as an effort in substance to tax the foreign seller on its disposal of shares, which would constitute a treaty violation. 66. Model 1 also involves the likelihood that there will always be multiple levels of Country L taxation, given that a deemed asset disposition will presumably create earnings which Country L will tax again upon their distribution. C. Issues with respect to Model 2 67. This Model, which is based on Article 13(4), has a cliff effect in the sense that it would allow Country L to tax 100% of the gain on a sale of shares whenever at least 50% of the share value was attributable to Country L immovable property. 68. International double taxation can ensue unless the country of residence of the Seller of shares acknowledges a primary right of Country L to tax 100% of the gain and unless any intermediary countries in the chain agree to recognize a step-up in basis (or otherwise not tax) stock that could be the subject of further dispositions. Furthermore, domestic double taxation (by Country L) could ensue unless the application of Country L taxation to the OIT was accompanied by a step-up in basis (for Country L purposes) of all the assets held by the entity directly owning the local immovable property and all the stock held by entities in each tier of the chain above that (including stock acquired by the Purchaser and stock held by entities in the chain above the Seller where the Seller has not fully disposed of its shares in the relevant entity). 69. The Model contemplates the imposition of OIT taxation on stock dispositions by small shareholders who may not be able to prove (or disprove) the legal basis for imposition of the OIT tax 108 (i.e., the fact that 50% of the share value is attributable to immovable property in Country L). If coupled with a withholding mechanism for enforcement, as mentioned below, this could hugely complicate trades in shares occurring on stock exchanges, which is one of the reasons the OECD Commentary on Article 13(4) notes the practice of some States not to apply that Article in such cases. 70. The Draft notes the serious enforcement / collection challenges posed by any attempt on the part of Country L to impose OIT taxation on transferors resident outside its jurisdiction. In doing so, however, it fails to further note that Country L typically could not rely, for enforcement / collection purposes, on information exchange or assistance in collection provisions of treaties unless its treaty partner had substantively agreed to its right to impose OIT taxation on residents of that treaty partner, which seems unlikely, at least where the asserted taxing right goes beyond that recognized in the applicable Article 13(4). 71. The Draft’s discussion of the potential withholding tax approach to enforcement of the Model 2 OIT tax does not adequately address the challenges posed by:  The Purchaser’s potential uncertainty as to whether application of the OIT regime of a particular country is triggered (which can require knowledge of the makeup and value of the assets all the way down the chain of the acquired stock), including potentially the status of that analysis for the year preceding the acquisition;  The Purchaser’s lack of knowledge as to the Seller’s basis in the shares being disposed of, and its consequent need to impose withholding on a gross basis;  The virtually inevitable need for the Seller to file refund claims with respect to any such gross basis withholding; and  The administrative burdens (on taxpayers and Country L tax authorities) of having to engage in a pre-clearance program to avoid the imposition of potentially confiscatory withholding taxes. 72. The Draft cites as a disadvantage of Model 2 that an agency approach to collection assumes that the direct owner in Country L can always make itself aware when there has been a transaction resulting in a triggering change of indirect ownership in the local entity, but this is also a challenge under Model 1’s deemed asset disposition approach. VI. LEGAL ISSUES A. Domestic law measures 73. The Draft’s recommendation that domestic law measures be adopted to avoid the need to amend treaty provisions limiting the taxation of indirect transfers constitutes the endorsement of an effective override, or at least circumvention, of the international tax treaty network. The recommendation that countries address any sourcing issues by amending their domestic law provisions regarding source is troubling for similar reasons. 74. The endorsement of purely domestic measures fails to address the disputes that will arise regarding existing treaty obligations and the barriers those measures may pose to the resolution of those disputes. It is difficult to see how the recommended domestic law approach would either achieve the greater international consistency or provide the greater certainty to investors promised by the Draft. 109 VII. COMPLEXITIES NOT ADDRESSED 75. The Draft is also deficient as a practical guide to the issues surrounding taxation of OITs by its failures to address a number of complexities involved, as outlined below. A. Valuation issues 76. Trying to design a regime for the taxation of OITs involves a series of difficult issues related to valuation. For example, both Model 1 and Model 2 require a determination of whether more than 50% of the share value of an entity is derived from immovable property in the taxing country (Country L). In both cases, this requires the determination of the current market value of assets (the immovable property and all other assets held within that corporate solution) which are not currently being sold. In some treaties, including all of those that incorporate the change to Article 13(4) set out in the OECD’s Multilateral Instrument, that valuation would have to be done not only as of the date of the share transfer but also for each of the 365 days preceding the transfer. In the case of Model 1, there would also have to be a valuation of the shares of each entity in the chain, from the one that directly holds immovable property in Country L up through the entity whose shares are actually being transferred to determine whether the 50% threshold is passed. This, too, can require the determination of the value of assets (shares below the top tier) that are not being sold, both as of the current date and for each of the 365 preceding days. Model 1 would also require the valuation of every other asset (and liability) held by any entity in the chain that passes the 50% threshold for purposes of applying the deemed sale and repurchase provision. 77. As indicated above, there is a convention (at least under the OECD Commentary) of determining whether the 50% threshold is passed by comparing the gross value of the Country L immovable property to the gross value of all other assets held by the relevant company. In other words, this is done without reference to liabilities. What this means is that a corporation with, say, two assets, one being Country L immovable property with a value of $5 million which is subject to a mortgage of $4 million (i.e., a net value of $1 million), and the other being a different asset with a value of $3 million, whose shares have a value of $4 million, will be treated as exceeding the 50% threshold, even though in reality only 25% of its net asset value is represented by Country L immovable property. Thus, this convention is highly distortive. On the other hand, difficulties and distortions could also arise by trying to ascertain on a net basis whether a particular asset represented more than 50% of outside share value, as that would require the allocation of liabilities (sometimes including liabilities incurred at a different corporate level than the underlying assets) among assets. 78. Another valuation aspect of any attempt to operate an OIT taxation regime is that it may not appropriately be limited to taxing gains attributable to the Country L immovable property being targeted. For example, at the time of sale of shares, the underlying assets may consist of Country L immovable property worth $5 million and another asset worth $4 million, and the shares would be worth $9 million. If there is gain of $2 million on the shares, that could be taxed entirely by Country L under an OIT regime. It’s entirely possible, however, that the $2 million of share gain was attributable to a $2 million appreciation in the value of the other asset, not to any gain attributable to the Country L immovable property. This is another indication of the distortions in the allocation of taxing rights that could arise from any effort to tax indirect transfers. B. Issues involving related party transactions 110 79. The Draft also fails to take into account the issues raised by the fact that a significant number of affected share dispositions may be between related parties. One obvious issue in this context is that transfer pricing issues may arise for purposes of determining the amount of the taxable gain on the indirect transfer. In other words, Country L may effectively have to depend on good transfer pricing enforcement by the countries of residence of the transferor and transferee in order to be sure that the taxable gain is properly computed. 80. Another huge issue in this area involves the appropriate tax treatment for corporate reorganizations, stock distributions, liquidations, mergers, etc. The Draft does not adequately acknowledge the extent to which routine internal group restructurings, which enjoy tax-free treatment under the law of many countries around the world, would be severely disrupted by the imposition of the OIT tax. Some mitigation of that disruption might be achieved if the OIT regime provided for exceptions covering these type of transactions, but the likely need for review and potential pre-clearance by Country L tax authorities would place an enormous burden on taxpayers and those tax authorities alike. C. Issues Involving Sales of Shares in Non-Wholly-Owned Company 81. The Draft also fails to address the substantial equity issues posed by the fact that not all structures will involve the highly simplified single owner scenario envisaged in the Draft’s examples, but many will involve structures having multiple shareholders. The Model 1 OIT recommendation puts a partial economic burden of the tax on non-selling shareholders, which violates the horizontal equity and ability-to-pay principles of good tax policy and administration. The Model 2 recommendation’s collection suggestion involving requiring the local entity to be an agent for the nonresident selling shareholder likewise puts partial economic burden of tax on non-selling shareholders. D. Treatment of losses 82. The Draft is virtually silent on losses, which raises a number of questions. For example, will the country of location of an asset recognize loss on an indirect transfer? Would that happen on the deemed sale of all assets under Model 1? Would that happen if the shareholder suffered a loss under Model 2? Against what categories of income could such a loss be taken? Since implementation and operation of the Models depend on country-level actions, how could symmetrical treatment of losses be required in practice? ***** 111 Accenture Intuit Inc. Activision Blizzard Intuitive Surgical Acxiom Keysight Technologies Adobe KLA-Tencor Corporation Agilent Lam Research Amazon Marvell Ancestry.com Maxim Integrated Apple MaxLinear Applied Materials Mentor Graphics Atlassian Microsoft Autodesk NetApp, Inc. Bio-Rad Laboratories Netflix BMC Software NVIDIA Broadcom Limited Oracle Corporation Brocade Palo Alto Networks Cadence PayPal Chegg, Inc. Pivotal Software, Inc. Cisco Systems Inc. Plantronics Dell Inc. Pure Storage Delphi Qualcomm Dolby Laboratories, Inc. Qualys, Inc. Dropbox Inc. Red Hat Inc. eBay salesforce.com Electronic Arts Sanmina-SCI Corporation Expedia, Inc. Seagate Technology Facebook ServiceNow Fitbit, Inc. Snapchat, Inc. Flex SurveyMonkey Fortinet Symantec Corporation GE Digital Synopsys, Inc. Genentech Tesla Motors, Inc. Genesys The Cooper Companies Genomic Health The Walt Disney Company Gigamon Theravance Biopharma Gilead Sciences, Inc. Tintri GitHub TiVo Corporation GLOBALFOUNDRIES Trimble, Inc. GlobalLogic Twitter Google Inc. Uber Technologies GoPro Veeva Systems Hewlett-Packard Enterprise Veritas HP Inc. Visa Indeed.com VMware Informatica Western Digital Ingram Micro, Inc. Xilinx, Inc. Integrated Device Technology Yahoo! Intel Yelp 112 19 October 2017 Platform for Collaboration On Tax c/o The World Bank Group 1818 H Street, NW Washington, DC 20433 Via email: taxcollaborationplatform@worldbank.org RE: Taxation of Offshore Indirect Transfers – A Tookit Dear Sir or Madam: The Platform for Collaboration on Tax (the Platform), a joint initiative of the Organisation for Economic Co-Operation and Development, International Monetary Fund, United Nations, and World Bank, released a document entitled The Taxation of Offshore Indirect Transfers – A Toolkit (the Draft Toolkit or Toolkit) on 1 August 2017. The Draft Toolkit was designed to help developing countries address the complexities of taxing offshore indirect transfers of assets, which the Platform states is a practice by which some multinational corporations try to minimize their tax liability. The Platform requested public feedback on the Draft Toolkit from interested stakeholders by 20 October 2017. On behalf of Tax Executives Institute, Inc. (TEI), I am pleased to respond to the Platform’s request for comments. TEI Background TEI was founded in 1944 to serve the needs of business tax professionals. Today, the organization has 56 chapters in Europe, North and South America, and Asia. As the preeminent association of in-house tax professionals worldwide, TEI has a significant interest in promoting tax policy, as well as the fair and efficient administration of the tax laws, at all levels of government. Our nearly 7,000 individual members represent over 2,800 of the leading companies in the world.1 TEI’s members are responsible for managing the tax affairs of their companies and must contend daily with the provisions of the tax law relating to the operation of business enterprises, including issues surrounding the tax 1 TEI is a corporation organized in the United States under the Not-For-Profit Corporation Law of the State of New York. TEI is exempt from U.S. Federal Income Tax under section 501(c)(6) of the U.S. Internal Revenue Code of 1986 (as amended). 113 complexities of offshore indirect transfers. We believe that the diversity and professional training of our members enable us to bring a balanced and practical perspective to the issues raised by the Draft Toolkit. TEI Comments Overview and Summary of Comments TEI appreciates the opportunity to comment on the Draft Toolkit and its proposed approach to the tax issues presented by offshore indirect asset transfers. As a threshold matter, TEI notes the status of the Toolkit is unclear. The Draft Toolkit does not appear to be an officially sanctioned or endorsed view of any of the contributing organizations that comprise the Platform, nor any of the member countries. However, in TEI’s view there is a strong possibility tax authorities, particularly in the less developed nations for which the draft Toolkit is being developed, will treat the Toolkit as authoritative guidance. Therefore, TEI recommends the Platform make clear that the Toolkit should not be treated as authoritative guidance and is not meant to override contrary guidance that is authoritative, including obligations imposed by bilateral income tax treaties. Changes to the fundamental policy underlying the capital gains articles of treaties should be the subject of discussion by countries, either bilaterally or in a multilateral framework. Overall, we believe the Platform should reconsider its suggested approach to offshore indirect transfers as set forth in the Draft Toolkit and focus on helping countries make informed decisions about how to treat offshore indirect transfers for tax purposes. This could be done by detailing the advantages and disadvantages of various approaches to taxing or not taxing offshore indirect transfers, and in particular detailing the issues that should be considered when making the decision whether to tax such transfer before deciding how to tax them. The Draft Toolkit should also ensure neutrality and symmetry for offshore indirect transfers when compared to direct asset transfers. Relevant issues to consider when assessing a tax on indirect transfers include how to determine the potential capital gain, how to ensure a step up in the basis of the underlying assets, whether deferral rather than recognition of gain is possible, how to limit the scope of the rules to ensure effective taxation while avoiding unintended taxation and other consequences, and how to address offshore indirect capital losses. It is critical that these issues be discussed in sufficient detail to ensure that the final toolkit provides helpful guidance to developing countries. More broadly, TEI believes the Platform’s agenda should be driven by the objective, also supported by the G20, of providing toolkits that increase certainty for taxpayers and tax authorities. Currently, how offshore indirect transfers will be assessed and taxed is often uncertain in various countries. Rather than trying to coordinate and recommend a consistent approach to taxation across countries, which in our view will fail, certainty can most likely be achieved in the Draft Toolkit by helping countries make clear and informed choices on whether such capital gains should be taxed, and, if so, how they should be taxed and what transactions will be subject to such a tax. Whether a transaction falls within the scope of an offshore indirect 114 transfer tax and the value of the actual result of the transfer, however both are determined, are the most common sources of disagreement between taxpayers and tax authorities, yet the Draft Toolkit offers very little to increase certainty in this respect. As both taxpayers and tax authorities are looking for certainty, it would be helpful to bring them together with Platform representatives to discuss concerns and potential solutions before finalizing the Draft Toolkit. Overall Approach of the Draft Toolkit It is unclear what goal the Platform has in mind for the Draft Toolkit. As the organizations comprising the Platform have different objectives and deliverables, further clarification of the approach and objective would be helpful. One issue, as noted above, is the status of the final toolkit as “final” or “authoritative” guidance for any of the Platform’s organizations. More broadly, TEI believes the United Nations’ (UN) approach to taxing offshore indirect transfers better suits the underlying technical issues. Generally, the UN provides different options for addressing tax issues and details the advantages and disadvantages of each. In general, the UN’s endeavors to place tax authorities in a position to make an informed choice when determining tax policy without attempting to choose the best approach on behalf of countries. This has been the approach, for example, of the UN Subcommittee on Extractive Industry Taxation Issues for Developing Countries, which has opined on the taxation of capital gains – direct or indirect – for such industries. In contrast, the Draft Toolkit appears to strongly recommend countries tax indirect capital gains as an initial matter, and then provides two specific options of how to do so. This goes beyond the general approach that at least some of the Platform’s contributing organizations take to such matters and may not be acceptable to countries who are members of those organizations, or other countries generally. The efforts of the organizations comprising the Platform should be commended, however, as the Platform appears to be the best way to align the approaches of these multinational organizations toward developing countries. We hope coordination between these institutions can increase, which in TEI’s view can be furthered with clarification of the Platform’s objectives, approaches, and deliverables. This coordination should include aligning terminology, definitions, and even abbreviations among the organizations to ease understanding of the Platform’s discussions and documents and reduce cross-organizational misunderstandings. Taxation of Capital Gains and Indirect Capital Gains in General The Draft Toolkit posits that a capital gains tax will not distort economic transactions. TEI believe that this is incorrect, particularly in extractive industries. In countries where capital gains taxation is present, introduced, or expanded, such taxation will likely result in fewer ownership transfers of business opportunities, which may leave them un- or under-exploited (either in efficiency or time). This approach may prevent the offshore business best suited to maximize the value of the opportunity from becoming involved, resulting in lower growth in the local country. 115 In general, investors assess after tax cash flows when determining the profitability of an opportunity. For this reason, the tax system, while not in itself decisive, will always be a factor in investment decisions. Moreover, tax systems are often specifically designed to encourage investment and increase employment. As capital is constrained via a capital gains tax, investors will compare alternatives. This is especially the case for business sectors whose profitability depends on the success of long term projects and prospects (e.g., the extractive industries); corporate taxation and capital gains taxation are only part of a tax system that can affect investment returns. Royalties and project bonuses bring forward the moment of taxation before any particular long-term prospect may become profitable. A capital gains tax and/or an indirect capital gains tax would further frontload taxation, pushing the point of an investment’s positive return further into the future. If a local tax system is already frontloaded, policymakers in that jurisdiction should consider whether to introduce or expand capital gains taxation in that context. In addition, introducing capital gains taxation may produce double taxation as the future income stream is taxed upon the disposal of a business and then again as the new owner generates income. An indirect capital gains tax only further increases the risk of double taxation. In TEI’s opinion, the above differences across tax systems in different countries makes the UN’s approach preferable. It is more useful for policy makers in developing countries to be fully aware of the advantages and disadvantages of a tax on offshore indirect transfer so they can better assess whether the implementation of such a tax in their current system would be predictable and clear. Having a tax regime that is based on consistent and predictable application of principles- based tax rules is the best way to promote and attract investment. In TEI’s view, tax principles and rules should be transparent, proportionate, administrable, fair, reasonably certain, conducive to timely determination of results, and avoid double taxation of profits or non-deduction of costs. Attempting to tax transactions in an ad hoc manner and potentially in contravention to agreed taxing rights (such as under a treaty) should be discouraged. Again, clearly presenting the advantages and disadvantages of capital gains taxation in the first instance, the same for indirect transfers second, and then options on how to implement such a tax is a better approach and will allow countries to make an informed decision on an indirect transfer tax that furthers their overall tax policy goals. Technical Considerations to Permit an Informed Decision on Capital Gains Taxation of Offshore Indirect Transfers Before considering options for the taxation of indirect transfers, TEI believes the developing countries that are the focus of the Draft Toolkit would be better served if the Toolkit addressed symmetry and neutrality in a broader sense. These aspects would help alleviate double taxation concerns that arise in many capital gains tax systems and should be analyzed in the Toolkit. While the toolkit early on recognizes that transfers of assets – either directly or indirectly – can generate capital gains as well as capital losses,2 none of the options set forth how to address offshore indirect transfers that result in a loss. The Draft Toolkit several times emphasizes the need for neutrality between direct and indirect transfers of capital gains, and 2 See, e.g., Draft Toolkit, p. 11. 116 indeed various capital gains tax systems currently exist that provide neutrality for share transactions, if the profit generating assets remain in the country, which ensures tax neutrality for such transactions. Although the Toolkit supports tax neutrality in principle, it does not provide enough detail on approaches that actually result in neutrality. TEI recommends that the Toolkit be modified to include examples of neutrality approaches. Symmetry in tax treatment is also not developed in the Toolkit. Generally, tax policy that taxes capital gains allows deductibility of losses. Should the Platform wish to make the case that where capital gains are taxed, indirect capital gains should also be taxed, it should detail how to deduct indirect capital losses. The options proposed in the paper only contain a short description of how offshore indirect transfers should be taxed and how that tax should be collected. A number of high level comments are then included on how to provide a basis step-up. Like the basis step- up, the treatment of a potential loss is not covered in detail at all, but merely mentioned, leaving countries, and in particular developing countries, to their own devices to address – or ignore – such critical issues. TEI also recommends the Platform exclude internal reorganizations from the scope of any indirect asset transfer tax. In an internal reorganization assets may be indirectly transferred to other parts of the group, for operational, legal, and other reasons. When countries tax indirect offshore transfers, such changes in shareholding within a group of companies would generally fall within the scope of the tax without any actual ultimate change of control of who owns the assets. The Toolkit also does not consider guidance on how to handle listed/publicly traded companies. When countries consider expanding their capital gains taxation to indirect transfers, all share transfers could create concerns. Shares in some companies may change hands frequently, so when assessing whether there has been a sufficient change in ownership to trigger an indirect capital gains tax, it would be extremely difficult to determine whether normal day-to- day trading of a company’s shares has triggered the rule. More fundamentally, public trading of a company’s shares does not present the same tax policy concerns as majority ownership transfers in an entity that holds local appreciated assets. For this reason, TEI recommends the final Toolkit exempt public trading on a stock exchange from the scope of any indirect transfer tax. The Toolkit should also specifically address how to treat joint venture (JV) partners of a company transferring its ownership. That is, the Platform should spell out at what point is a capital gains tax triggered when only one JV partner transfers part or all of its shares (at once or within a specified time period) and who is liable for that tax as it raises several questions. For example, do the indirect transfer tax models expect the JV entity – owned by parties who have not transferred anything – to pay the tax, which would impact all JV partners? How would a step up in basis work in a JV scenario? How do the models in the Toolkit envisage a JV obtaining the funds necessary to pay any such tax if all the proceeds from a sale go to the JV owner? What if nobody acquires control in such a transfer? For example, suppose two JV partners sell 20% and 30% respectively, but to different purchasers – would that trigger the tax? Should there be a need for someone to gain control, as opposed to just a change in control, for the capital gains tax to be 117 triggered? Moreover, many JVs are organized as pass-through entities, with the tax burden of the JV’s operations falling directly on the JV’s owners. Thus, any tax levied on the JV entity for a sale of an interest in the entity would have an immediate deleterious economic impact in the JV’s remaining owner(s). The Platform should therefore also address how to handle JV’s formed as pass-through entities in the final Toolkit. These types of issues need to be addressed by any country enacting an indirect transfer tax because they can substantially impede a country’s ability to create certainty for investors, especially for large and long term investments. Elaborating in detail on some of these issues would therefore help improve tax certainty. With respect to the models generally, it would be very helpful to focus on clarifying the issues and options, as well as detailing their advantages and disadvantages. Potential Anti-Abuse Approach The Draft Toolkit describes the offshore indirect transfers that would be subject to tax as transfers made “offshore for tax purposes.” In case the Toolkit only covers transfers made offshore for tax purposes, it should include examples of offshore transfers that are not made for tax purposes and therefore not subject to the tax. Most indirect transfers are made for non-tax reasons and it is critical that the Platform state this in the Toolkit. For example, share sales may be preferred to asset sales because of the ease of transferring the underlying assets and local employee considerations. Moreover, offshore companies are often used for non-tax reasons, such as the absence of robust local corporate legislation and the protection of intellectual property. More broadly, the Draft Toolkit should avoid the inference that most companies illegally evade taxes. The vast majority of multinational companies are compliant and pay their taxes in accordance with all laws. To address those small number of companies who operate outside the law, countries with clear legislation and transparent tax policy could then tax offshore indirect transfers only in abusive cases. This approach would focus those countries on tax abuse while limiting unintended consequences to investments and the resulting negative impact on economic growth. Such a beneficial result would be furthered if application (or non-application) of such an anti-abuse legislation can be confirmed up front by tax authority rulings in a transparent manner. Should tax authorities decide to look through the corporate form and consider an indirect transfer as a direct asset transfer under an anti-abuse rule, the taxpayer should have the opportunity to provide evidence that the offshore transaction was made for non-tax purposes and therefore not within the scope of the rule. Specific Comments on the Options for Implementing an Indirect Transfer Tax The Draft Toolkit includes two pertinent options for implementing an offshore indirect transfer tax. The first option makes the direct corporate asset owner, who does not receive any compensation in the transfer, liable for the taxes on capital gains realized by another company. This option is problematic for a number of reasons, primarily because the asset owner may not have sufficient funds to pay what may be a substantial tax burden. The Draft Toolkit points this out on page 47, however, it does not address how such a tax payment could be funded nor, if the 118 payment is debt-funded, whether interest on that debt is deductible. The Toolkit also does not address the impact on JV partners of making the JV entity liable for the tax. JV partners are often not involved in indirect transfers (i.e., the transfer is made by the other JV partner) but may nevertheless be taxed on a non-existent gain. The second option has the advantage of taxing the party effecting the transfer. This should reduce or eliminate the issue of funding the resulting tax payment and will allow the party paying for the assets to obtain a step-up in basis. To be effective and provide more certainty, the Draft Toolkit should also discuss ongoing effects of a tax on an indirect sale, for example how should depreciation be allowed and on what basis. In both options, determining the value of the actual gain is often a primary issue. The Draft Toolkit appears to imply (correctly) that only the gain should be taxed, not the proceeds. The Toolkit should state this explicitly and also include further clarification as to what assets the tax should apply, and how the gain is to be allocated among them (e.g., based on their fair market value? Tax basis? Book value?). Additional Specific Issues The Draft Toolkit assumes that the “source” country has the primary right to tax the gain on the underlying property in an indirect transfer (i.e., a transfer of shares in a company that owns the underlying property) and does not discuss the rationale for residence based taxation of shares. Under current general international tax principles, the country of residence has the right to tax capital gains other than those explicitly enumerated by the relevant tax treaty. The “political economy” argument – positing that the inability of a country in which an asset is located to tax indirect tax transfers provokes “intense domestic dissatisfaction” and may harm efforts to build a “tax-paying culture”3 – focuses on a few high-profile cases that are not representative of the vast majority of asset transfers, whether direct or indirect. In TEI’s view, the high-profile cases are more appropriately dealt with through narrower, targeted rules – perhaps an anti-abuse rule as discussed above. The Draft Toolkit contains several examples of country practices in taxing offshore transfers, including a discussion of the U.S. taxation of dispositions of U.S. real property held by foreign investors. However, the examples deal with the simplest of cases and any rules promulgated to address offshore indirect transfers generally would need to adopt and define many thresholds and terms for such rules to be practically applied. We also note that the Draft Toolkit abandons, without justification, the general treaty definition of immovable property. Instead, it advocates an alternative, novel, and expansive definition of such property, which the Toolkit then acknowledges is difficult to capture in legislative language. It is unclear how this new definition would be interpreted by tax authorities and would create a significant future risk of economic double taxation. TEI therefore recommends the Draft Toolkit use the traditional treaty definition of immovable property. 3 Draft Toolkit at 23. 119 Separately, the concept of location-specific rents is not helpful. As the Draft Toolkit acknowledges, access to a local market could be considered to generate location specific rents. However, it appears that the concept is intended to be interpreted expansively yet is poorly defined. As such, in TEI’s view location specific rents will be interpreted in ways that will reduce certainty and deter investment. We recommend the final version of the Toolkit not include this concept. Finally, the deemed disposal approach ignores the difficulties associated with imposing a tax on an entity that has no proceeds from a sale and may be unable to pay the tax. Also, depending on the thresholds, it may be difficult for the entity holding the local property to know that a transfer triggering gain recognition has occurred. This approach should be discarded or tightly limited to a small number of clearly delineated fact patterns. Conclusion TEI appreciates the opportunity to comment on the Draft Toolkit regarding offshore indirect transfers. These comments were prepared under the aegis of TEI’s European Direct Tax Committee, whose Chair is Giles Parsons. If you have any questions about the submission, please contact Mr. Parsons at +44 1455 826561, parsons_giles@cat.com, or Benjamin R. Shreck of the Institute’s legal staff, at +1 202 464 8353, bshreck@tei.org. Sincerely yours, TAX EXECUTIVES INSTITUTE, INC. Robert L. Howren International President 120 Transfer Pricing Economists for Development Paris France www.tped.eu October 19th, 2017 The Platform For Collaboration on Tax Discussion Draft The Taxation of Offshore Indirect Transfers – A Toolkit Subject: Comments on the Discussion Draft on The Taxation of Offshore Indirect Transfers (“OITs”) Dear Sir or Madam, The purpose of this letter is to present TPED’s comments after the release by The Platform For Collaboration on Tax of the Draft on the Taxation of Offshore Indirect Transfers (“the Draft” or “the Draft Report”). Transfer Pricing Economists for Development (“TPED”) is a recently-launched Paris-based Think-Tank aiming to promote the development and sharing of business economics knowledge in transfer pricing as an enabler of development of emerging economies and developing countries. In line with the Association’s focus, TPED’s comments focus on the economic aspects of the Draft distinguished from, but in support of, the tax and legal considerations, which have been duly taken into account. Our comments will focus on: - the proposed extended definition of immovable property in article 13(4) MTC, including reference to government rights under which companies may operate; - the valuation of such rights for the assessment of the 50% criterion in said article 13(4) MTC; and - the taxation of the whole transaction in the State issuing the rights, and related “basis step-up” for the enterprise as a whole, once that the 50% threshold is satisfied, as opposed to pro-rated taxable asset rule. TPED – www.tped.eu 121 Page 2 October 19th, 2017 We use this opportunity to compliment the authors for using real cases examples in the Draft as opposed to hypothetical, simplified examples. 1. The Extended Definition of Immovable Property The discussion draft suggests the following terms: “Box 10: Extended definition of immovable property “Immovable property” includes… (e) A right granted by or on behalf of the government (whether or not embodied in a license) to be a supplier or provider of: (i) goods (such as radioactive materials) (ii) utilities (such as electricity or gas); or (iii) other services (such as telecommunications and broadcast spectrum and networks) country wide or within a geographical area of Country L” The Draft defines “in economic terms, the concept of ‘immovability’ might be most meaningfully thought of as proxying for the possibility of location specific rents”. […] This view suggests an expansive definition of ‘immovability’ capable of capturing at least the most likely sources of significant LSR.” Still, the Draft acknowledges “the concept of LSR has not been fully developed to be readily captured in legislative language” and that “LSRs can be difficult to identify in general” with the exception of “government-created rights, notably in the extractive industries and telecoms.” Our comments here focus exclusively on economic concepts and will not delve into legal analysis; we must note, however, that extending the definition of “immovable property” in the language of international tax treaties in order to allocate taxing rights may face legal challenges in numerous jurisdictions throughout the world. The physical or corporeal nature of “immovable property” and its inherent or construed connection with “land” can trigger challenges of interpretation which perhaps could be avoided if an entirely new “qualification” (other than “immovable property”) were created. Perhaps “government rights” would deserve a separate article in the Model Convention. Again, and although any legal analysis is outside the scope of this commentary we would like to point out that a different – and greater – definition of immovable property would have to be reconciled with the guidance developed under the G- 20/OECD BEPS Project, whereby in the new Chapter VI of the OECD Transfer Pricing Guidelines governmental rights are treated as intangible property, as discussed in the next section. TPED – www.tped.eu 122 Page 3 October 19th, 2017 The Draft, nonetheless, calls for the following comments from an economic and transfer pricing perspective: - (a) Government rights within the proposed extended definition of immovable property are intangible assets The “right[s] granted by or on behalf of the government (whether or not embodied in a license)” targeted by the Draft seem to correspond in part or in full with the OECD Transfer Pricing Guidelines’ “rights under […] government licenses”: “Government licenses and concessions may be important to a particular business and can cover a wide range of business relationships. They may include, among others, a government grant of rights to exploit specific natural resources or public goods (e.g. a licence of bandwidth spectrum), or to carry on a specific business activity. Government licences and concessions are intangibles within the meaning of Section A.1” (para. 6.6) They fulfill the criterion set by the OECD and the UN with respect to an intangible: - “In these Guidelines, therefore, the word “intangible” is intended to address something which is not a physical asset or a financial asset, which is capable of being owned or controlled for use in commercial activities, and whose use or transfer would be compensated had it occurred in a transaction between independent parties in comparable circumstances.” (para 6.24 of the OECD Transfer Pricing Guidelines) - ”For the purposes of this chapter the term “intangible” encompasses something which is neither a physical nor a financial asset, which is capable of being owned or controlled for commercial purposes, whose use or transfer would be compensated had it occurred between independent enterprises in comparable circumstances.” (para. B.5.2.3 of the UN TP Manual) The current Draft Report is, in our opinion, not explicit enough with respect to the exact nature of the assets included in the extended definition, and whether these assets are to be deemed intangibles or not. On the contrary, the Draft Report provides a relatively elliptic definition of intangibles, which does not include the government rights in scope of the extended definition: “Intangible Property. For purposes of this report, this term is defined herein as property which has no physical presence, for example, a financial asset such as corporate stock; intellectual property; business goodwill” (page 7), which is not necessarily aligned with either the definition by the OECD or the one by the UN. TPED – www.tped.eu 123 Page 4 October 19th, 2017 Instead of qualifying the rights in scope of the extended definition as intangible assets, the Draft Report links them to Location Specific Rents “(LSRs”), while acknowledging the inherent challenges of the concept (an economic concept rather than a legal one). Based on the above, it would be useful to leverage the developments made in the transfer pricing area about the concept of LSRs: The concept of LSR has been elaborated in the context of transfer pricing1. The concept of Location Specific Advantages (LSAs), which captures the same phenomenon, is widely used in international economics literature2, and is now better defined in the transfer pricing context. Various authors have contributed to the definition of LSAs3, the definition of Location Rents resulting from the exploitation of the sources of LSAs and the allocation of LSAs among various members of a Multinational Enterprise45. LSAs may involve the supply side, notably “factors of production and distribution that can be exploited to produce a particular product or service cheaper, better and/or with less risk, or to increase the ability of a company to sell more products, at a higher price and/or achieve a larger market share”.6 On the other hand, they may also relate to the wider market structure and demand side, for instance legal, regulatory or administrative restrictions, as well as physical or other constraints, limiting the number of competitors and inducing an artificial scarcity in the relevant market. These location specific advantages may then lead to excess demand and the capability for the incumbents to sell more products at a higher price7. Both the OECD and the UN have commented that LSAs, at the origin of the LSRs, do not constitute intangibles per se, are not capable of being owned or controlled, but still should be taken into account in a transfer pricing analysis. The exclusive reliance in the wording of the 1 See glossary to the 2017 update of the United Nations Transfer Pricing Practical Manual for developing countries. 2 The concept of location specific advantages is one widely used in international economics literature, see for example Dunning, J.H. (1977), "Trade, Location of Economic Activity and the Multinational Enterprise; a Search for an Eclectic Approach", in Ohlin, B., Hesselborn, P.O., and Wijkman, PJ. (éd.), The International Allocation of Economic Activity, London, MacMillan. 3 Patton M.F., Quick P.D., “Location Savings after Sundstrand v Commissioner: out of the BALRM and into the Game Room?”, Tax Management International Journal, July 1991, 20-7 4 McKee M. and M. McDonald, “Location Savings in competitive markets”, Tax Management Transfer Pricing Report, Vol. 9, No. 19, Feb 2001. 5 Gonnet, Fris and Coriano, “Location Specific Advantages-Principles”, Transfer Pricing International Journal (June 2011). 6 Patton and Quick, (1991) 7 Gonnet, Fris and Coriano, “Location Specific Advantages-Principles”, Transfer Pricing International Journal (June 2011). TPED – www.tped.eu 124 Page 5 October 19th, 2017 Draft Report on LSRs to justify the taxation in the local country of certain rights may create confusion. In TPED’s view, governments rights within the extended definition are indeed “market specific characteristics” but fall under the “intangible” category, as they are owned and controlled by the local entity. Still, they are indeed or may be at the origin of LSRs in the local country. As a matter of economic policy, it should be noted that the attribution of “immovable property” treatment to such government-created “intangibles” extends the “force of attraction” feature of immovable property to governmental licenses. This could represent an incentive for the “over- regulation” of domestic markets, which would be distortionary and particularly detrimental to developing economies. Immovable property exerts such “force of attraction” in respect to taxing rights because of its inherent connection with a nation’s territory irrespective of its regulatory or institutional environment; governmental licenses and other such “intangibles”, instead, are not inherent to land but “created” as a consequence of economic policy choices, governmental activity, domestic laws, and regulations within each country’s institutional framework. Therefore, by granting such extended taxing rights or “force of attraction” to “government-created assets”, by equating such “assets” to “immovable property”, states that are in fact “consumer markets” or “labor markets” would have a “tax incentive” for increased regulation and increased governmental interference in domestic markets. Over-extended regulatory activities triggered by such “tax incentive” could have detrimental welfare effects to the countries that take this approach: over-extended regulations could distort the allocation of factors of production and of capital, distort competition and reduce investment within such countries, and most critically reduce consumer surplus and labor welfare in developing economies. In light of the above, we recommend; - To confirm the intangible nature of the “government rights” within the extended definition of the Draft Report, and to align it with that of the OECD Guidelines / UN Manual, - To retain the reference to LSRs but, as a consequence of the exploitation of an intangible owned by a company (the government rights), while acknowledging that such an intangible/government rights may not be the only source of the LSRs that the company benefits from (see Section 2), - To further refine the nature of governmental rights and regulations (e.g. in extractive industries and utilities) that conform to the object and purpose of the rule, thereby discouraging an “over-regulation” of markets. The next section discusses the valuation of such rights, now properly defined, for the assessment of the 50% criterion in said article 13(4) MTC. TPED – www.tped.eu 125 Page 6 October 19th, 2017 2. The valuation of such rights for the assessment of the 50% criterion in said article 13(4) MTC and proposed “inside basis step-up” Article 13(4) MTC states that “Gains derived … from the alienation of shares deriving more than 50% of their value directly or indirectly from immovable property in (a) … State may be taxed in that …. State.” Where usually the value of an underlying real estate can be assessed in a relatively straightforward manner, this is very different in assets that are brought under article 13(4) by an extension of the definition of immovable property. The value is in many cases derived from exploitation activities in-country that involve more activities abroad, investment and knowledge than the mere holding of a governmental license as described in the extension of the article 13(4) definition. Such broader activities, investments and knowledge are often owned or performed by more than one party in multiple countries, and therefore critically linked to analyses that are ruled by the Transfer Pricing Guidelines. The Draft Report seems to make a clear and direct link between LSRs (or residual/excess value) and the government rights in scope of the extended definition, but acknowledges only in a few instances that other value creating activities could be involved in the value creation process: - while discussing taxation in the country where an asset is located (in the inter-nation equity section) , the Draft suggests cases where “perhaps, a substantial part of those [capital] gains are attributable to value-enhancement provided from abroad […] Establishing the extent of any such contribution, however, could of course be problematic” (page 18) - “The increase value of the entity sold may reflect in part managerial and other expertise contributed by the seller. […] It may indeed be that there are company-specific as well as location-specific rents at work” (page 21) In TPED’s view, the understanding of the origin of LSRs, and more generally of value and profit creation by the local company (owning the rights) is crucial as it forms the basis for the subsequent financial evaluation of the 50% threshold. It cannot be presumed that only the “government right” explains the local company’s flow of profits, and therewith its value. We suggest using economic theory concepts, already leveraged by the OECD and the UN in their definition and assessment of LSAs, to confirm the origin of value and profit in such situations. In transfer pricing, the treatment of LSRs demands 1. an understanding of LSRs in the broader context of how the MNE operates, and 2. an apportionment of such LSRs between entities (or even parts thereof) of the MNE involved in the exploitation of such LSRs. TPED – www.tped.eu 126 Page 7 October 19th, 2017 As LSRs result in some forms of super profits (or residual profits in transfer pricing terms), it is generally the case that the generation of such LSRs does not involve a passive behavior from the beneficiary of such rights. There will likely be counter examples to this active behavior, notably found in what the Draft calls the “minimal definition of immovable property”: a lease of land for instance may generate LSR and sees its value increasing by factors not related to any active management of the right, (for instance investments in the neighboring surroundings by the State; etc.). In general though, LSRs are the fruit of an active management, notably through people, investments, expertise, etc. Looking at the “extended definition of immovable property”, the government rights in the telecom industry (category (iii)) clearly fall in the category of active management. The government indeed allows to an operator to operate under certain spectrum usage rights, within the context of a specific national regulatory framework. Still, such an operator needs to make large investments in networks, needs to deploy relevant technologies, and needs to structure a suitable commercial offering, to attract and retain customers, often in a highly competitive market. In case such operator generates LSRs, it is generally a combination of all its assets (including the government rights) which are at the origin of such value. The LSA matrix provides an illustration that LSRs do not originate by the mere market feature or grant of a right but in some cases, also involve other intangibles: TPED – www.tped.eu 127 Page 8 October 19th, 2017 LSA Matrix Generally I – Market Power & General Access to LSAs: II – Perfect Competition in the end-product available Extra Profits, but no Location Rents market & General Access to LSAs: and No Location Rents exploited Without market power, the MNE would not be LSAs do not (or only temporarily)create extra able to convert LSAs into super-profits; market profits for the company because competition in the power of the MNE drives the conversion of end-product markets forces prices to reflect costs location advantages into super-profits, which in and all the relevant competitors share the same cost strict economic terms are return on the specific structure. Therefore the firms would be forced to sources of competitive advantage (e.g. pass the cost savings to the final customers. Availability and exploitation of LSAs intangibles), rather than location rents. . IV- Market Power & Exclusive Access to III – Perfect Competition & Exclusive Access to LSAs: Location rents LSAs: Location rents LSAs convert into extra profits as a Prices are determined by competitors’ prices, who do combination of some kinds of exclusivity in the not have access to location LSAs. Thus LSAs access to the LSAs and market power. Part of originate location rents for the MNE. these extra profits can be defined as Location Rents. Examples: Telecom Examples: Lease of a Location (e.g., factories) Exclusive access and exploitation Protection from Perfect Degree of competition in the end-product competition competition market (market power) Under such a matrix, we focus on quadrants III and IV where there is exclusivity in the access to, and availability and exploitation of, the sources of LSAs. In the context of a perfect competition, any LSR resulting from a specific right or a specific restriction is indeed explained fully or quasi-exclusively by the right or the restriction. A lease of land might be within this category. On the contrary, other rights (or restrictions) do not generate LSR (and then gain value) in the absence of an active management through people, investments, expertise, etc. TPED – www.tped.eu 128 Page 9 October 19th, 2017 This suggests that: - In the first case, the full value should be recognized locally, in the place where the LSR is realized where the right/the restriction (or even the -exclusive - market feature) is awarded. - In the second case, the LSRs result from exclusivity in the access to the LSAs (for instance a government right) and market power, generally related to the Company’s competitive advantage. Such competitive advantage cannot be obtained from a passive behavior by the Company or by the “group” or wider “enterprise” to which it belongs. Active management, notably through people, investments, belonging to a wider multinational firm, and non-local knowledge and expertise is common to multinationals in multiple sectors, which want to ensure their profitability and continuity not only in each local country but worldwide. Companies operating under certain government rights do have specificities in the sense that - they are (to a varying extent) regulated meaning that their tariffs, investments, returns are (more or less) regulated (controlled); - they operate under the strict supervision of the Government or a regulator; and - they may have special obligations or treatment, for instance of their employees. But there is a myriad of situations. The extent to which such regulated companies need to develop and maintain value-added intangibles varies depending on the markets they are in. Utilities such as gas and electricity do not face the same challenges as the telecom industry, and government concessions and licenses for manufacturing, production, oil refineries, or container terminal locations, to name a few, have their returns and value intertwined and inextricably linked with a broader business venture. All the other (non-governmental) assets and intangibles may have been developed locally and/or may have been contributed by other non-local entities of a multinational group. In the transfer pricing world, which entities are entitled to the LSRs‘ underlying profits depends on which entities contribute to 1. securing the exclusivity and 2. developing the competitive advantage. The above analysis should be straight-forward only in the case of a local company owning a lease of land, owned by an entity without any staff or substance. But for companies with more complex operations, such as the ones targeted by the extended definition, the above will form the basis of a broad understanding of the contribution of the various companies involved (including the local company, owning the government rights and possibly other intangible assets, its parent, and possibly other group companies). The subsequent step is the financial valuation of the asset in scope of the immovable property definition. We understand that the 50% threshold should be evaluated by comparing the fair TPED – www.tped.eu 129 Page 10 October 19th, 2017 market value of the (“extended”) immovable property with the total value of the assets, for which the transaction provides an arm’s length price. Within this financial valuation exercise, not only the immovable property should be revalued, but all the assets of the company, including intangible assets, some of them not appearing on the balance sheet of the company. This asset reevaluation is well-known to accountants who routinely perform Purchase Price Allocations post-transactions. In the context of the current Draft Report, a number of questions arise: - What is the value of the government rights in isolation from the rest of the assets of the Company? - What is the value of the other intangibles (such as customer relationships/clientele or the brand) in isolation from the government rights? - How to treat goodwill that accountants typically recognize in their PPAs? o In this respect, we note that the Draft Report includes business goodwill in the scope of intangibles. The OECD does not. Under the OECD Guidelines, goodwill, while not an intangible, should be taken into account in a transfer pricing analysis. Isn’t goodwill exactly what the Draft Report intends to relate to the government rights and tax in the context of the realization of a capital gain? The above has large implications in terms of the evaluation of the 50% share: Assume an individual “location specific asset” akin to an immovable property (as proposed in the Draft) has a historical value of $100, and that a transaction involving the sale of the shares of the parent of the local company, owning such asset (OIT) suggests an arm’s length price for the local company and its parent of $500. Assume further that revalued tangible assets are worth $150. Remains $250 of value, which should be split among the various intangibles of the firms (both the local company and the parent), including a revaluation of the immovable property. Depending the results of this analysis, the 50% threshold may or may not be reached. Furthermore, the Draft notes that in taxing the entire value of the transaction, even if source countries allow and recognize a full “step-up” in the depreciable or amortizable basis of “all assets”, a substantial “time-value of money” benefit would accrue to local tax treasuries. Given the multitude of intangible assets (separate from the government right itself) that could be “stepped-up” as a result, it is theoretically possible that a substantial “write-off” could be triggered on the same year when the “gain” is captured by the local country. This would reduce the present value of the tax revenue of the first year, while other “stepped -up assets” would provide further amortizations and depreciation in future years. Therefore, the net revenue effect to local treasuries arising from such broad, enterprise-wide “step-up” is hard to anticipate but may well be far less substantial than suggested in the Draft. If, instead, the taxing right and the “step-up” in question were limited to the “government right” itself (as opposed to the entire enterprise value), then a time-value “benefit” would always accrue to local revenue authorities (a lesser gain, and a write-off spread over-time). In this case, perhaps a recommendable policy TPED – www.tped.eu 130 Page 11 October 19th, 2017 would be to abandon the 50% threshold and tax all indirect transfers of government rights, with a corresponding step-up. In light of the above, we recommend: - To confirm that the evaluation of the 50% threshold should be based on the comparison of the immovable property fair market value with that of all the assets within the transaction scope (including assets and functions of the local Company and its foreign affiliates which, in conjunction, support the enterprise value in question) - To confirm that an economic and financial evaluation should be performed on a case- by-case basis, involving the evaluation of all individual assets in scope (including intangibles, also those that are not within the balance sheet of the companies in scope – both local and parent company involved); - To indicate that PPAs, as performed by accountants, should not exclusively be relied upon but like the OECD has suggested that PPAs of the transaction provide some insights which are not necessarily binding the transfer pricing analysis. In this respect the treatment of goodwill is crucial. - To provide guidance with respect to the suggested techniques to apportion the various blocks of value for the purpose of the 50% evaluation. 3. The taxation of the whole transaction in the state that issues the right, once that the 50% threshold is satisfied, as opposed to pro-rated taxable asset rule. Under current OECD and UN Model MTCs, the full amount of the capital gain is taxed when the 50% is met. The “force of attraction” approach seems to trump the assertion of whether the gain in question is reasonably attributable to the location-specific “asset”, which seems a relatively fair assumption in the context of the existing definition of immovable property. But as the intent is to broaden the definition to a larger scope of industries, firms and activities, we wonder whether applying a “force of attraction” approach for the gain which is not attributable to the immovable property or to the governmental right itself is coherent with the underlying economics that justifies the proposed allocation of taxable rights. The Draft Report presents the taxable asset rule that Kenya has adopted for the extractive sectors, where the source country imposes tax only on a proportionate basis, equal to the share of the revalued immovable property over total value of the assets, when the immovable property accounts for between 20% and 50% of the transaction value. TPED – www.tped.eu 131 Page 12 October 19th, 2017 In light of a broader definition of immovable property, this approach could be in our view be a promising approach, as - It focuses only on the asset of the Company that is tight to the country, i.e.; the government rights, - It avoids a taxation of the other company’s intangibles (not government rights), which all other companies (not operating under government rights) have to similarly use for their business, - It may allow to broaden the scope of the definition of immovable property to all types of government rights (to the extent they fall under the OECD/UN definition of intangibles) - It raises the question of the benefits of maintaining a 50% threshold, which may create adverse effects. In light of the above, we recommend considering the taxable asset rule as an alternative to the source rule, involving the full taxation above a threshold and zero taxation below. Conclusion The Draft Report aims to expand the taxation rights of countries where certain assets are located by an extension of the reach of article 13(4) MTC. This seems to be the intended and economically justifiable extension of the taxing rights of countries where value is created based, for more than 50%, on locally issued government rights. However, it incites certain legal challenges (given varying local-country definitions of “immovable property” in spite of the treaty), and perhaps an entirely new article ought to be created. Government rights within the proposed extended definition of immovable property are “intangible assets”; as such, we recommend an alignment with the OECD and UN intangibles definition. From an economic policy perspective, such category of assets should be further defined so as to avoid creating a “tax incentive for over-regulation” of local markets which could be particularly damaging to workers and consumers in developing economies. As the test for purposes of article 13(4) MTC is whether these rights have a value of more than 50% of the total transaction price, in the first place the identification of the (type of) assets concerned is critical. Secondly, the valuation needs to be done for the right, distinguished from possible other contributing elements to the value creation as a whole. Economic theory on LSRs can provide the foundation for an understanding of the interplay between the various activities, assets and contributing entities. An explicit alignment with the OECD / UN LSA concept is welcomed, as well as guidance with respect to the financial and economic valuations recommended for the purpose of the 50% assessment. TPED – www.tped.eu 132 Page 13 October 19th, 2017 Taxation as suggested under method 1 is based on a deemed disposal of the asset(s) concerned. Taxation under method 2 needs the same analysis of the relations between the different entities concerned. Under both methods however, the source rule applies where the full value is taxed when the 50% threshold is met (and no taxation below). The Kenyan example (on the extractive sector) provides an illustration of an alternative taxable asset rule. Under this method, a pro-rata application of values should be used in certain instances. A taxable asset rule, together with extended immovable property clearly defined as local “government related” intangibles (separate and different from other intangibles) could accompany an even larger definition of immovable property (such as the one proposed by the Draft Report in page 57). This would also prevent the unintended inflation of the “asset step-up” described in the Draft. The choice between methods suggested may well be left to the discretion of the country of location of the asset. It is critical however that this choice is respected and followed by the country of residence of the entity transferred. We finally also mention that the value of the OIT itself may be influenced by the applicability of one method rather than the other. We thank you again for the opportunity of providing comments and remain at your disposal for further comments. Best regards On behalf of TPED8 Sébastien Gonnet, TPED President, NERA Economic Consulting Paris Romero J.S. Tavares, TPED Vice President, WU Vienna University of Economics and Business, Vienna Pim Fris, NERA Economic Consulting, Paris Giammarco Cottani, Ludovici & Partners, Milan 8 The views expressed are those of the authors, not necessarily those of TPED or its other members. TPED – www.tped.eu 133 October 19, 2017 VIA EMAIL The Platform for Collaboration on Tax taxcollaborationplatform@worldbank.org Re: USCIB Comment Letter on the Platform for Collaboration on Tax’s draft toolkit on the taxation of offshore indirect transfers of assets For the attention of the members of the Platform for Collaboration on Tax, USCIB1 is writing to comment on the discussion draft on offshore indirect transfers. In our view, the taxation of offshore indirect transfers should not be considered in the context of ta a “toolkit”. The discussion draft proposes potentially significant shifts in taxing rights for “source” and “residence” countries. Decisions on significant shifts in taxing rights ought to be debated among countries at the appropriate multilateral fora and not resolved by guidance provided by the staff of international organizations without debate among the countries. Therefore, this discussion draft should be withdrawn. This toolkit raises most starkly an issue that has been raised by other toolkits issued by the Platform for Collaboration on Tax. What is the legal status of these “toolkits”? USCIB raised this issue in commenting on other toolkits, but the issue has not been adequately addressed. Because “neither this draft nor the final report should be regarded as officially endorsed views of those organizations or of their member countries”, the status would seem to be no more than an academic article; if well- researched and argued, it may be persuasive, but should not be a source of authority on its own. USCIB is concerned, however, that tax authorities may treat the toolkits as authoritative guidance. Each toolkit should make clear that they are not authoritative and cannot override contrary guidance that is authoritative. Our primary position is that the toolkit should be withdrawn, nevertheless, we address some of the questions raised in the discussion draft below. 1. Does this draft toolkit effectively address the rationale(s) for taxing offshore indirect transfers of assets? No. The discussion draft assumes that the so-called “source” country has the primary right to tax the gain on the underlying property and does not discuss the rationale for residence based taxation of that property. It misstates the current treaty rule. The country of residence has the right to tax capital gains other than those explicitly enumerated by the treaty. The political economy argument focuses on a few high-profile cases that are not representative of 1 USCIB promotes open markets, competitiveness and innovation, sustainable development and corporate responsibility, supported by international engagement and prudent regulation. Its members include top U.S.- based global companies and professional services firms from every sector of our economy, with operations in every region of the world. With a unique global network encompassing leading international business organizations, USCIB provides business views to policy makers and regulatory authorities worldwide, and works to facilitate international trade and investment. 134 the vast majority of asset transfers, whether direct or indirect. Those cases might be more appropriately dealt with narrower targeted rules. 2. Does it lay out a clear principle for taxing offshore indirect transfers of assets? The two proposals are clear in their general outlines, but as noted below in our response to question 9 many difficult issues are ignored or treated cursorily. 3. Is the definition of an offshore indirect transfer of assets satisfactory? No. The example deals with the simplest of cases and the rules would need to adopt and define many thresholds if these rules are to be practical in application. 4. Is the discussion regarding source and residence taxation in this context balanced and robustly argued? No. See response to question 1 above. 5. Is the suggested possible expansion of the definition of immovable property for the purposes of the taxation of offshore indirect transfers reasonable? No. The discussion draft abandons the treaty definition of immovable property and advocates an expansive definition of immovable property, which the draft itself acknowledges would be difficult to capture in legislative language. This is a prescription for uncertainty and double taxation. 6. Is the concept of location-specific rents helpful in addressing these issues? If so, how is it best formulated in practical terms? This is not helpful. As the draft acknowledges, access to a local market could be considered to generate location specific rents. A concept that is intended to be interpreted expansively and that is poorly defined will be interpreted in ways that will reduce certainty and deter investment. 7. Are there other implementation approaches that should be considered? 8. Is the draft toolkit's preference for the 'deemed disposal' method appropriate? 9. Are the complexities in the taxation of these international transactions adequately represented? No. The simplified example that forms the basis of the analysis contained in the discussion draft ignores the complexities involved in determining whether the transaction should be subject to tax. The discussion draft also ignores or skates over: the difficulties dealing with minority shareholders, valuation issues, the treatment of losses, and how economic double-taxation would be avoided. The discussion draft makes no mention of internal reorganizations, which in our view should be exempt from any taxes on indirect transfers, as there is no change in underlying ownership, so tax liabilities would constitute a leakage as there is no genuine economic gain. This is especially the case if the reorganization is driven by regulatory requirements. Sincerely, William J. Sample Chair, Taxation Committee United States Council for International Business (USCIB)