Jill Gatehouse and Susanna Brain (Freshfields Bruckhaus Deringer) review the state of play on the OECD’s project to tackle base erosion and profit shifting (BEPS), and highlight the main points of interest.
Background and scope
1. What is ‘BEPS’?
‘BEPS’ stands for base erosion and profit shifting. The concept is generally used to describe arrangements which aim to minimise taxable profits in high tax jurisdictions, with such profits either disappearing for tax purposes, or arising in lower (or no) tax jurisdictions. However, the term has now become synonymous with an OECD project sponsored by the G20 governments, which aims to formulate a set of international tax rules/principles to combat BEPS and which can then be implemented by individual countries. Fifteen key areas for action to address BEPS are identified in the OECD’s action plan. It would be easy to devote 20 questions to any one of the 15 action points alone. This article seeks to give a snapshot of where the project currently stands and to highlight particular points of interest in relation to some of the actions.
2. What triggered the OECD BEPS project?
Political weight was really thrown at the issue following a joint statement by George Osborne and Wolfgang Schäuble to the G20 in November 2012 calling for concerted international cooperation to strengthen international corporation tax standards. The OECD was asked to report to the G20 on BEPS in 2013. In the run up to this statement, several high profile multinationals had been subject to media (and parliamentary) criticism over their tax affairs. At the time of the statement, the furore over Starbucks’ royalty structure was in full swing and Lin Homer (HMRC chief executive) had just been grilled by the Public Accounts Committee. Amazon, Google and Starbucks faced Margaret Hodge a few days later. International tax planning had become a major political issue, both in the UK and elsewhere.
The G20 have tasked the OECD with designing a policy framework for international corporate tax that eliminates inappropriate double non-taxation, whether that arises from the actions of taxpayers, or indeed from the tax policies of national governments themselves.
3. What is the current timeframe?
The full BEPS package is to be circulated in time for the G20 finance ministers’ meeting in October 2015. The multilateral instrument (see below) is due to be finalised by December 2016.
The first set of seven deliverables were presented to and endorsed by the G20 finance ministers in September 2014. These consisted of three reports (the first two of which are final) assessing:
tax challenges of the digital economy (Action 1);
feasibility of a multilateral instrument to implement BEPS (Action 15); and
harmful tax practices (Action 5).
They also consisted of draft rules in four areas:
hybrid mismatch arrangements (Action 2);
treaty abuse (Action 6);
transfer pricing of intangibles (Action 8); and
transfer pricing documentation and a country-by-country reporting template (Action 13).
However, due to the interaction between the 2014 measures and the 2015 ones, agreed measures will remain in draft form until the 2015 measures are finalised, and even those measures that are described as ‘final’, such as the hybrids paper, are the subject of further work (guidance is expected later this year).
The remaining eight deliverables timetabled to be presented this year relate to:
controlled foreign companies (CFCs) (Action 3);
interest deductions (Action 4);
countering harmful tax practices (Action 5);
artificial avoidance of PE status (Action 7);
transfer pricing (Action 8–10),
methodologies for analysis of BEPS data and actions to address BEPS (Action 11);
disclosure of aggressive tax planning arrangements (Action 12); and
effective dispute resolution (Action 14).
However, work is to continue into 2016 on certain aspects, particularly where the recommendations on one action depend in part on the recommendation on another.
4. Which countries are now involved?
At the time of the February 2015 update to G20 leaders, there were 61 countries directly participating in the OECD BEPS Action Plan. This is a substantial increase from the initial number, due in large part to the launch by the OECD of its strategy for deepening developing country engagement following criticism that they were not adequately represented. In addition, the IMF, UN, World Bank, African Tax Administration Forum (ATAF) and the Inter-American Center of Tax Administrations (CIAT) are ‘observers’.
5. Isn’t BEPS just directed at the digital economy?
The overriding message from the Action 1 report on the tax challenges of the digital economy, which was published last summer, was that it isn’t appropriate to address the digital economy separately. The digital economy is increasingly the economy itself and is so interlinked with the wider economy that it is ‘difficult, if not impossible’ to ring-fence it. The report was clear that any attempt to do so would create arbitrary lines between the digital and non-digital sphere resulting in a highly unsatisfactory system, a conclusion in line with the initial responses from UK business to the early proposals made by the OECD. So, no, this is not a project that only those involved in digital businesses need to take notice of.
Impact to date
6. What has the impact of the OECD BEPS project been to date?
Although the OECD BEPS project is not yet finished, we are already seeing domestic law changes resulting from it. Enabling legislation has been introduced for regulations providing for country by country reporting (FA 2015 s 122), and a consultation paper was launched last year in respect of hybrid mismatches. Other states, such as Germany, have already introduced legislation aimed at hybrid structures.
The UK’s patent box regime has been challenged on the basis that it is a harmful preferential tax regime (Action 5) and, on 2 December 2014, the UK government published a summary of likely changes to the regime. This follows joint proposals made by the UK and German governments on 11 November 2014 for a new approach to patent box regimes which now form the basis of continuing work in this area. Countries with non-compliant regimes must either make the regime compliant, or close the regime to new IP by 30 June 2016 and abolish it by 30 June 2021. The UK Treasury has confirmed its commitment to retaining a patent box and to having a compliant regime in place when the existing regime is closed to new IP, although no further details have been provided.
Other actions are likely to have less impact on the UK’s tax system. For instance, the discussion on CFCs (Action 3) so far seems largely compatible with the UK’s current CFC regime. However, it remains to be seen how the partial exemption for finance companies within that regime will be treated by other jurisdictions for the purposes of rules introduced to combat hybrid structures.
In the UK, we have also had the introduction of the diverted profits tax, although such unilateral action pre-empts the outcome of the OECD BEPS project and has therefore angered some other participators.
Country by country reporting
7. When will country by country reporting start?
Model legislation requiring the ultimate parent of a multinational group to file the country by country report in its jurisdiction of residence was published on 8 June 2015 as part of an implementation package. Preparation for reporting by multinationals with a turnover above €750m is expected to start in 2016 with the first information obtained by tax administrations in 2017.
8. What will be disclosed and to whom?
Information is to be provided on revenues, profits, taxes accrued and paid and some other indicators of the location of economic activity, as well as information about which entities conduct business in a particular jurisdiction and the business activities each entity engages in. Once provided, the information can then be shared with tax authorities in the other jurisdictions in which the group operates.
Even with revisions made to original proposals, there will be significant compliance burdens. There are also concerns about whether the proposals will give unnecessary access to sensitive information. An underlying concern is that making comprehensive information available so widely may, notwithstanding guidance on how such information should be used, give rise to an unprecedented level of (perhaps unsubstantiated) challenges from tax authorities under pressure to secure their share, leading to additional compliance issues for the multinational.
The multilateral instrument
9. What is the ‘multilateral instrument’?
As mentioned above, there are 15 actions in the OECD BEPS project, many of which cannot be addressed without amending bilateral tax treaties (of which there are over 3,000 worldwide). Action 15 analysed the possibility of developing a multilateral instrument to enable countries to update this network of treaties efficiently and to make implementation rapid and consistent. The idea would be that states could simply sign up to the multilateral instrument and this would automatically modify that state’s double tax treaties with other signatory states.
The OECD’s report on Action 15, published on 6 February 2015, concluded that such an instrument was both feasible and desirable, and that negotiations for the instrument should be convened quickly.
10. What actions would it cover?
This will be the subject of debate, but the OECD would like it to cover treaty changes to implement recommendations on the following actions:
treaty abuse involving hybrid entities (part of Action 2);
treaty abuse (Action 6);
permanent establishments (Action 7); and
dispute resolution (Action 14).
To the extent that treaty changes are required in relation to country by country reporting (Action 13) or the actions in relation to transfer pricing (Actions 8–10), these might also be included.
11. How would it work?
Whilst tax practitioners are generally more familiar with bilateral treaties, there are existing examples of multilateral instruments. In the tax context, the OECD Convention on Mutual Administrative Assistance in Tax Matters governing information exchange and assistance in recovery of taxes has 66 signatory states.
This co-exists with and essentially supplements the bilateral treaties of those states, so the concept is not new. However, in the case of BEPS, one particular challenge for the working group will be agreeing a set of provisions that all states are willing to adopt. How many of the actions will there be sufficient consensus on to include within the instrument? Moreover, certain of the actions, for instance preventing treaty abuse (see below), currently contemplate a choice being made between alternative acceptable approaches. It will be interesting to see how this is addressed. The OECD considers that it is possible for a multilateral instrument to allow ‘opt-outs’ regarding certain measures, or even choices to be made between provisions. That may be true as an academic matter, but it would detract from the aim of having a swift introduction of consistent measures, and might involve bilateral negotiations that the multilateral instrument is intended to avoid.
12. Who is participating in the multilateral instrument discussions?
Work began in Paris on 27 May 2015 with the establishment of a large ad hoc group in which over 80 countries are participating. The OECD aims to conclude drafting of the instrument by 31 December 2016. There are certain states with very wide treaty networks whose participation will be crucial in the success of the multilateral instrument. The UK is one of those and it is notable that Mike Williams of HM Treasury is the chair of the group. A notable absentee is the US, apparently on the basis that its treaties already contain effective anti-avoidance provisions.
Treaty abuse
13. What are the latest proposals on countering treaty abuse?
The OECD released a third discussion draft on Action 6 in May 2015. The draft provided ‘conclusions and proposals’ on 20 issues identified in earlier discussion documents. The work focuses on two areas: the ‘limitations on benefits’ (LOB) rule; and the proposed introduction of a ‘principal purpose test’ (PPT), both designed to address treaty shopping.
The OECD and G20 countries have agreed on a common minimum standard requiring the adoption of: (1) a combination of a simplified LOB rule and the PPT rule; (2) the PPT rule; or (3) a more complex LOB rule supplemented by a mechanism to deal with conduit financing arrangements.
The revised draft also proposes a new rule under which entities benefiting from a ‘special tax regime’ resulting in a low effective tax rate would be denied treaty relief. States would be able to exclude specific types of entity from this rule.
14. Is this going to be a significant change, or is it similar to existing treaty shopping provisions and Indofood?
The potential impact of a LOB rule, and of the PPT enabling tax authorities to test subjective intentions of the parties, should not be underestimated. Anyone who has had cause to apply the LOB clause in the UK/US treaty will appreciate the complexity of these provisions in practice. Both types of clause have clear implications for, for example, a Luxembourg holding company structure. It is fair to say that some tax authorities are already, based on existing treaty shopping clauses, or the application of domestic general anti-avoidance principles, seeking to challenge the availability of treaty relief where there is a lack of substance, and in these jurisdictions the provisions may simply reinforce the approach. However, the standard proposed goes well beyond the sort of analysis that might currently fall to be considered in the UK, for instance in relation to whether treaty relief in respect of UK withholding on interest is available.
HMRC’s guidance on Indofood (Indofood International Finance Ltd v JP Morgan Chase Bank [2006] EWCA Civ 158) requires consideration of whether the recipient of the interest is beneficially entitled to it, essentially by reference to whether it might be obliged to pay the interest on (see HMRC’s International Tax Manual at ITM332040). A LOB clause would (very broadly) consider where the shareholders of the entity are based and, if such shareholders are in a third state, whether the entity is receiving the income as part of an active trade (not as a investment business). A PPT clause might require identification of non-tax business reasons for the location of the lender in the treaty state.
Final proposals are expected in September and will be included in the next update to the Model Tax Convention. However, discussions in relation to so-called ‘non-CIVs’ (including private equity funds, sovereign wealth funds and pension funds) will continue into 2016. Whilst representations have been made that special treatment should be afforded to such funds, governments are concerned that non-CIVs may be used to provide treaty benefits to investors that are not themselves entitled to treaty benefits and that investors may defer recognition of income on which treaty benefits have been granted. In the meantime, uncertainty remains for such funds as to the implications of this Action and the implications of the proposals on hybrid mismatches. In relation to the latter, there is concern that the concept of income not being taxed might be broad enough to capture such funds. Guidance on this is expected soon.
Permanent establishments
15. Is the treaty definition of ‘permanent establishment’ going to change?
A revised discussion draft was released on 15 May 2015. The main proposals address commissionaire arrangements, anti-fragmentation and splitting of contracts.
A key point is on commissionaire arrangements and the amendment of art 5(5) of the Model Treaty to cover a person who not only ‘concludes contracts’ (a relatively bright line) but also who ‘negotiates the material elements of contracts’. The draft accepts that what this means depends on the contract but it is clear that it would usually include negotiations of price, nature and quantity of goods or services and that the basic premise is that it is intended to catch a person who acts as the salesforce of the entity. The proposed changes would also clarify that the commonly listed exclusions from PE status (storage and delivery) would apply only where such activities are of a preparatory or auxiliary nature.
The draft acknowledges concerns raised that the proposals may create significant numbers of PEs (with a greater compliance and administrative burden) but with little or no additional tax. Arguably, the changes here go further than preventing BEPS – there is no requirement that the arrangements be tax motivated. But these concerns have not led to a change in approach and it does seem that the question of whether a PE has been created is set to become less clear cut. Follow-up work regarding profit attribution will be carried on after September, with a view to providing guidance by the end of 2016 (in line with the timetable for the Multilateral Instrument that would implement this Action and taking into account the conclusions on the transfer pricing actions (Actions 8 to 10).
Transfer pricing
16. Isn’t BEPS about transfer pricing? What’s happening on that?
Actions 8 to 10 concern transfer pricing principles and are 2015 deliverables. The proposals involve revisions to the OECD Transfer Pricing Guidelines and consider issues such as the circumstances in which transactions might be recharacterised, the use of hindsight when valuing hard to value intangibles and cases where ‘special measures’ might be necessary. Special measures essentially involve a departure from the arm’s length principle, the fundamental basis for transfer pricing principles to date. The proposals being discussed are therefore very material. There is also recognition that certain of the areas of discussion will need to be aligned with other actions; for instance, work on capitalisation will need to take into account proposals on restrictions on interest deductibility more generally.
Interest deductibility
17. What are the proposals on interest deductibility?
The Action 4 discussion draft was published last December and provoked strong responses from business and the adviser community. The principal approaches outlined are:
a group wide rule: limiting a company’s net interest deductions by reference to a proportion of the group’s net third party interest expense;
a fixed ratio rule: limiting a company’s interest deductions to an amount determined by applying a fixed benchmark ratio to an entity’s earnings, assets or equity (similar to the German model); or
a combination of the first two.
The group-wide rule (more stringent than the UK’s worldwide debt cap (WWDC) regime) has support within the OECD, but it has met widespread resistance from business and advisers as being academically attractive but impractical. It is thought that such a rule would be certain to result in a group not getting deductions for all its external financing costs.
18. Is this really necessary in the UK?
The UK already has a large number of anti-avoidance rules which can target excessive debt deductions (transfer pricing/thin capitalisation, unallowable purpose tests, deemed distribution rules, anti-arbitrage rules, WWDC). Moreover, as a basic matter, the UK’s declining headline rate of corporation tax means that, in a global context, it is not necessarily going to be the first jurisdiction a group looks at for ‘debt dumping’. That was a specific policy point at the time of the 2009 review of the taxation of foreign profits, when more aggressive restrictions on interest deductibility were considered (and rejected). Taking into account the other BEPS Actions, there is a question as to whether further rules restricting interest deductions are really necessary. There is also an argument that this is another example of the OECD project overreaching itself. In most cases there is nothing ‘artificial’ about decisions regarding where to locate interest deductions, even where the relative tax rates of jurisdictions are taken into account.
19. What has been the outcome of industry-specific lobbying in the area of interest deductibility?
A number of industry specific representations have been made to HM Treasury, notably in respect of infrastructure/project financings, property and upstream oil and gas (areas that the discussion draft indicated might require special considerations).
Reports of those discussions suggest that an exemption for specific industries is unlikely, not least because it would be difficult to get agreement from other OECD countries as to what those industries should be. Instead, the focus is on excluding types of lending arrangement that pose a low BEPS risk.
However, in recognition of the fact that the proposals have the potential to have a major impact on the infrastructure industry, the OECD is considering a targeted ‘project finance’ based exemption to exclude interest deductions for third party debt linked to this type of financing (which is typically heavily leveraged and non-recourse) from a group’s fixed ratio rule calculation. That proposal is understood to have support across the participating countries.
And finally...
20. Will it work? Is it likely that the OECD project will successfully tackle BEPS?
It is too early to say, but the amount of work and apparent consensus that has been achieved in a relatively short time is impressive.
It is likely that the multilateral instrument will hold the key to the implementation of the revisions to treaties, but implementation will also require domestic law changes. A lack of consistency between states could result in double taxation; this is an important concern. The OECD (and many others) would say that existing rules and principles are skewed too far towards preventing double taxation and therefore permit double non-taxation. This may well be correct, but care must be taken to ensure that the reverse does not become reality. This is perhaps more likely because of the fact that some of the Actions involve proposals that arguably go beyond preventing BEPS in the sense of tax focused planning to shift profits in an artificial way.
However, leaving aside the risks, perhaps the greatest measure of the success of the BEPS project would be if the UK soon felt able to repeal the DPT.
Jill Gatehouse and Susanna Brain (Freshfields Bruckhaus Deringer) review the state of play on the OECD’s project to tackle base erosion and profit shifting (BEPS), and highlight the main points of interest.
Background and scope
1. What is ‘BEPS’?
‘BEPS’ stands for base erosion and profit shifting. The concept is generally used to describe arrangements which aim to minimise taxable profits in high tax jurisdictions, with such profits either disappearing for tax purposes, or arising in lower (or no) tax jurisdictions. However, the term has now become synonymous with an OECD project sponsored by the G20 governments, which aims to formulate a set of international tax rules/principles to combat BEPS and which can then be implemented by individual countries. Fifteen key areas for action to address BEPS are identified in the OECD’s action plan. It would be easy to devote 20 questions to any one of the 15 action points alone. This article seeks to give a snapshot of where the project currently stands and to highlight particular points of interest in relation to some of the actions.
2. What triggered the OECD BEPS project?
Political weight was really thrown at the issue following a joint statement by George Osborne and Wolfgang Schäuble to the G20 in November 2012 calling for concerted international cooperation to strengthen international corporation tax standards. The OECD was asked to report to the G20 on BEPS in 2013. In the run up to this statement, several high profile multinationals had been subject to media (and parliamentary) criticism over their tax affairs. At the time of the statement, the furore over Starbucks’ royalty structure was in full swing and Lin Homer (HMRC chief executive) had just been grilled by the Public Accounts Committee. Amazon, Google and Starbucks faced Margaret Hodge a few days later. International tax planning had become a major political issue, both in the UK and elsewhere.
The G20 have tasked the OECD with designing a policy framework for international corporate tax that eliminates inappropriate double non-taxation, whether that arises from the actions of taxpayers, or indeed from the tax policies of national governments themselves.
3. What is the current timeframe?
The full BEPS package is to be circulated in time for the G20 finance ministers’ meeting in October 2015. The multilateral instrument (see below) is due to be finalised by December 2016.
The first set of seven deliverables were presented to and endorsed by the G20 finance ministers in September 2014. These consisted of three reports (the first two of which are final) assessing:
tax challenges of the digital economy (Action 1);
feasibility of a multilateral instrument to implement BEPS (Action 15); and
harmful tax practices (Action 5).
They also consisted of draft rules in four areas:
hybrid mismatch arrangements (Action 2);
treaty abuse (Action 6);
transfer pricing of intangibles (Action 8); and
transfer pricing documentation and a country-by-country reporting template (Action 13).
However, due to the interaction between the 2014 measures and the 2015 ones, agreed measures will remain in draft form until the 2015 measures are finalised, and even those measures that are described as ‘final’, such as the hybrids paper, are the subject of further work (guidance is expected later this year).
The remaining eight deliverables timetabled to be presented this year relate to:
controlled foreign companies (CFCs) (Action 3);
interest deductions (Action 4);
countering harmful tax practices (Action 5);
artificial avoidance of PE status (Action 7);
transfer pricing (Action 8–10),
methodologies for analysis of BEPS data and actions to address BEPS (Action 11);
disclosure of aggressive tax planning arrangements (Action 12); and
effective dispute resolution (Action 14).
However, work is to continue into 2016 on certain aspects, particularly where the recommendations on one action depend in part on the recommendation on another.
4. Which countries are now involved?
At the time of the February 2015 update to G20 leaders, there were 61 countries directly participating in the OECD BEPS Action Plan. This is a substantial increase from the initial number, due in large part to the launch by the OECD of its strategy for deepening developing country engagement following criticism that they were not adequately represented. In addition, the IMF, UN, World Bank, African Tax Administration Forum (ATAF) and the Inter-American Center of Tax Administrations (CIAT) are ‘observers’.
5. Isn’t BEPS just directed at the digital economy?
The overriding message from the Action 1 report on the tax challenges of the digital economy, which was published last summer, was that it isn’t appropriate to address the digital economy separately. The digital economy is increasingly the economy itself and is so interlinked with the wider economy that it is ‘difficult, if not impossible’ to ring-fence it. The report was clear that any attempt to do so would create arbitrary lines between the digital and non-digital sphere resulting in a highly unsatisfactory system, a conclusion in line with the initial responses from UK business to the early proposals made by the OECD. So, no, this is not a project that only those involved in digital businesses need to take notice of.
Impact to date
6. What has the impact of the OECD BEPS project been to date?
Although the OECD BEPS project is not yet finished, we are already seeing domestic law changes resulting from it. Enabling legislation has been introduced for regulations providing for country by country reporting (FA 2015 s 122), and a consultation paper was launched last year in respect of hybrid mismatches. Other states, such as Germany, have already introduced legislation aimed at hybrid structures.
The UK’s patent box regime has been challenged on the basis that it is a harmful preferential tax regime (Action 5) and, on 2 December 2014, the UK government published a summary of likely changes to the regime. This follows joint proposals made by the UK and German governments on 11 November 2014 for a new approach to patent box regimes which now form the basis of continuing work in this area. Countries with non-compliant regimes must either make the regime compliant, or close the regime to new IP by 30 June 2016 and abolish it by 30 June 2021. The UK Treasury has confirmed its commitment to retaining a patent box and to having a compliant regime in place when the existing regime is closed to new IP, although no further details have been provided.
Other actions are likely to have less impact on the UK’s tax system. For instance, the discussion on CFCs (Action 3) so far seems largely compatible with the UK’s current CFC regime. However, it remains to be seen how the partial exemption for finance companies within that regime will be treated by other jurisdictions for the purposes of rules introduced to combat hybrid structures.
In the UK, we have also had the introduction of the diverted profits tax, although such unilateral action pre-empts the outcome of the OECD BEPS project and has therefore angered some other participators.
Country by country reporting
7. When will country by country reporting start?
Model legislation requiring the ultimate parent of a multinational group to file the country by country report in its jurisdiction of residence was published on 8 June 2015 as part of an implementation package. Preparation for reporting by multinationals with a turnover above €750m is expected to start in 2016 with the first information obtained by tax administrations in 2017.
8. What will be disclosed and to whom?
Information is to be provided on revenues, profits, taxes accrued and paid and some other indicators of the location of economic activity, as well as information about which entities conduct business in a particular jurisdiction and the business activities each entity engages in. Once provided, the information can then be shared with tax authorities in the other jurisdictions in which the group operates.
Even with revisions made to original proposals, there will be significant compliance burdens. There are also concerns about whether the proposals will give unnecessary access to sensitive information. An underlying concern is that making comprehensive information available so widely may, notwithstanding guidance on how such information should be used, give rise to an unprecedented level of (perhaps unsubstantiated) challenges from tax authorities under pressure to secure their share, leading to additional compliance issues for the multinational.
The multilateral instrument
9. What is the ‘multilateral instrument’?
As mentioned above, there are 15 actions in the OECD BEPS project, many of which cannot be addressed without amending bilateral tax treaties (of which there are over 3,000 worldwide). Action 15 analysed the possibility of developing a multilateral instrument to enable countries to update this network of treaties efficiently and to make implementation rapid and consistent. The idea would be that states could simply sign up to the multilateral instrument and this would automatically modify that state’s double tax treaties with other signatory states.
The OECD’s report on Action 15, published on 6 February 2015, concluded that such an instrument was both feasible and desirable, and that negotiations for the instrument should be convened quickly.
10. What actions would it cover?
This will be the subject of debate, but the OECD would like it to cover treaty changes to implement recommendations on the following actions:
treaty abuse involving hybrid entities (part of Action 2);
treaty abuse (Action 6);
permanent establishments (Action 7); and
dispute resolution (Action 14).
To the extent that treaty changes are required in relation to country by country reporting (Action 13) or the actions in relation to transfer pricing (Actions 8–10), these might also be included.
11. How would it work?
Whilst tax practitioners are generally more familiar with bilateral treaties, there are existing examples of multilateral instruments. In the tax context, the OECD Convention on Mutual Administrative Assistance in Tax Matters governing information exchange and assistance in recovery of taxes has 66 signatory states.
This co-exists with and essentially supplements the bilateral treaties of those states, so the concept is not new. However, in the case of BEPS, one particular challenge for the working group will be agreeing a set of provisions that all states are willing to adopt. How many of the actions will there be sufficient consensus on to include within the instrument? Moreover, certain of the actions, for instance preventing treaty abuse (see below), currently contemplate a choice being made between alternative acceptable approaches. It will be interesting to see how this is addressed. The OECD considers that it is possible for a multilateral instrument to allow ‘opt-outs’ regarding certain measures, or even choices to be made between provisions. That may be true as an academic matter, but it would detract from the aim of having a swift introduction of consistent measures, and might involve bilateral negotiations that the multilateral instrument is intended to avoid.
12. Who is participating in the multilateral instrument discussions?
Work began in Paris on 27 May 2015 with the establishment of a large ad hoc group in which over 80 countries are participating. The OECD aims to conclude drafting of the instrument by 31 December 2016. There are certain states with very wide treaty networks whose participation will be crucial in the success of the multilateral instrument. The UK is one of those and it is notable that Mike Williams of HM Treasury is the chair of the group. A notable absentee is the US, apparently on the basis that its treaties already contain effective anti-avoidance provisions.
Treaty abuse
13. What are the latest proposals on countering treaty abuse?
The OECD released a third discussion draft on Action 6 in May 2015. The draft provided ‘conclusions and proposals’ on 20 issues identified in earlier discussion documents. The work focuses on two areas: the ‘limitations on benefits’ (LOB) rule; and the proposed introduction of a ‘principal purpose test’ (PPT), both designed to address treaty shopping.
The OECD and G20 countries have agreed on a common minimum standard requiring the adoption of: (1) a combination of a simplified LOB rule and the PPT rule; (2) the PPT rule; or (3) a more complex LOB rule supplemented by a mechanism to deal with conduit financing arrangements.
The revised draft also proposes a new rule under which entities benefiting from a ‘special tax regime’ resulting in a low effective tax rate would be denied treaty relief. States would be able to exclude specific types of entity from this rule.
14. Is this going to be a significant change, or is it similar to existing treaty shopping provisions and Indofood?
The potential impact of a LOB rule, and of the PPT enabling tax authorities to test subjective intentions of the parties, should not be underestimated. Anyone who has had cause to apply the LOB clause in the UK/US treaty will appreciate the complexity of these provisions in practice. Both types of clause have clear implications for, for example, a Luxembourg holding company structure. It is fair to say that some tax authorities are already, based on existing treaty shopping clauses, or the application of domestic general anti-avoidance principles, seeking to challenge the availability of treaty relief where there is a lack of substance, and in these jurisdictions the provisions may simply reinforce the approach. However, the standard proposed goes well beyond the sort of analysis that might currently fall to be considered in the UK, for instance in relation to whether treaty relief in respect of UK withholding on interest is available.
HMRC’s guidance on Indofood (Indofood International Finance Ltd v JP Morgan Chase Bank [2006] EWCA Civ 158) requires consideration of whether the recipient of the interest is beneficially entitled to it, essentially by reference to whether it might be obliged to pay the interest on (see HMRC’s International Tax Manual at ITM332040). A LOB clause would (very broadly) consider where the shareholders of the entity are based and, if such shareholders are in a third state, whether the entity is receiving the income as part of an active trade (not as a investment business). A PPT clause might require identification of non-tax business reasons for the location of the lender in the treaty state.
Final proposals are expected in September and will be included in the next update to the Model Tax Convention. However, discussions in relation to so-called ‘non-CIVs’ (including private equity funds, sovereign wealth funds and pension funds) will continue into 2016. Whilst representations have been made that special treatment should be afforded to such funds, governments are concerned that non-CIVs may be used to provide treaty benefits to investors that are not themselves entitled to treaty benefits and that investors may defer recognition of income on which treaty benefits have been granted. In the meantime, uncertainty remains for such funds as to the implications of this Action and the implications of the proposals on hybrid mismatches. In relation to the latter, there is concern that the concept of income not being taxed might be broad enough to capture such funds. Guidance on this is expected soon.
Permanent establishments
15. Is the treaty definition of ‘permanent establishment’ going to change?
A revised discussion draft was released on 15 May 2015. The main proposals address commissionaire arrangements, anti-fragmentation and splitting of contracts.
A key point is on commissionaire arrangements and the amendment of art 5(5) of the Model Treaty to cover a person who not only ‘concludes contracts’ (a relatively bright line) but also who ‘negotiates the material elements of contracts’. The draft accepts that what this means depends on the contract but it is clear that it would usually include negotiations of price, nature and quantity of goods or services and that the basic premise is that it is intended to catch a person who acts as the salesforce of the entity. The proposed changes would also clarify that the commonly listed exclusions from PE status (storage and delivery) would apply only where such activities are of a preparatory or auxiliary nature.
The draft acknowledges concerns raised that the proposals may create significant numbers of PEs (with a greater compliance and administrative burden) but with little or no additional tax. Arguably, the changes here go further than preventing BEPS – there is no requirement that the arrangements be tax motivated. But these concerns have not led to a change in approach and it does seem that the question of whether a PE has been created is set to become less clear cut. Follow-up work regarding profit attribution will be carried on after September, with a view to providing guidance by the end of 2016 (in line with the timetable for the Multilateral Instrument that would implement this Action and taking into account the conclusions on the transfer pricing actions (Actions 8 to 10).
Transfer pricing
16. Isn’t BEPS about transfer pricing? What’s happening on that?
Actions 8 to 10 concern transfer pricing principles and are 2015 deliverables. The proposals involve revisions to the OECD Transfer Pricing Guidelines and consider issues such as the circumstances in which transactions might be recharacterised, the use of hindsight when valuing hard to value intangibles and cases where ‘special measures’ might be necessary. Special measures essentially involve a departure from the arm’s length principle, the fundamental basis for transfer pricing principles to date. The proposals being discussed are therefore very material. There is also recognition that certain of the areas of discussion will need to be aligned with other actions; for instance, work on capitalisation will need to take into account proposals on restrictions on interest deductibility more generally.
Interest deductibility
17. What are the proposals on interest deductibility?
The Action 4 discussion draft was published last December and provoked strong responses from business and the adviser community. The principal approaches outlined are:
a group wide rule: limiting a company’s net interest deductions by reference to a proportion of the group’s net third party interest expense;
a fixed ratio rule: limiting a company’s interest deductions to an amount determined by applying a fixed benchmark ratio to an entity’s earnings, assets or equity (similar to the German model); or
a combination of the first two.
The group-wide rule (more stringent than the UK’s worldwide debt cap (WWDC) regime) has support within the OECD, but it has met widespread resistance from business and advisers as being academically attractive but impractical. It is thought that such a rule would be certain to result in a group not getting deductions for all its external financing costs.
18. Is this really necessary in the UK?
The UK already has a large number of anti-avoidance rules which can target excessive debt deductions (transfer pricing/thin capitalisation, unallowable purpose tests, deemed distribution rules, anti-arbitrage rules, WWDC). Moreover, as a basic matter, the UK’s declining headline rate of corporation tax means that, in a global context, it is not necessarily going to be the first jurisdiction a group looks at for ‘debt dumping’. That was a specific policy point at the time of the 2009 review of the taxation of foreign profits, when more aggressive restrictions on interest deductibility were considered (and rejected). Taking into account the other BEPS Actions, there is a question as to whether further rules restricting interest deductions are really necessary. There is also an argument that this is another example of the OECD project overreaching itself. In most cases there is nothing ‘artificial’ about decisions regarding where to locate interest deductions, even where the relative tax rates of jurisdictions are taken into account.
19. What has been the outcome of industry-specific lobbying in the area of interest deductibility?
A number of industry specific representations have been made to HM Treasury, notably in respect of infrastructure/project financings, property and upstream oil and gas (areas that the discussion draft indicated might require special considerations).
Reports of those discussions suggest that an exemption for specific industries is unlikely, not least because it would be difficult to get agreement from other OECD countries as to what those industries should be. Instead, the focus is on excluding types of lending arrangement that pose a low BEPS risk.
However, in recognition of the fact that the proposals have the potential to have a major impact on the infrastructure industry, the OECD is considering a targeted ‘project finance’ based exemption to exclude interest deductions for third party debt linked to this type of financing (which is typically heavily leveraged and non-recourse) from a group’s fixed ratio rule calculation. That proposal is understood to have support across the participating countries.
And finally...
20. Will it work? Is it likely that the OECD project will successfully tackle BEPS?
It is too early to say, but the amount of work and apparent consensus that has been achieved in a relatively short time is impressive.
It is likely that the multilateral instrument will hold the key to the implementation of the revisions to treaties, but implementation will also require domestic law changes. A lack of consistency between states could result in double taxation; this is an important concern. The OECD (and many others) would say that existing rules and principles are skewed too far towards preventing double taxation and therefore permit double non-taxation. This may well be correct, but care must be taken to ensure that the reverse does not become reality. This is perhaps more likely because of the fact that some of the Actions involve proposals that arguably go beyond preventing BEPS in the sense of tax focused planning to shift profits in an artificial way.
However, leaving aside the risks, perhaps the greatest measure of the success of the BEPS project would be if the UK soon felt able to repeal the DPT.
