This article will deal with the 2017 and 2021updates of the text of the UN Model Double Taxation Convention between Developed and Developing Countries[1], and will more specifically try to say something about their possible impact on the tax treaty practice of countries. After a short overview of the most relevant changes to the text of the UN Models 2017 and 2021, I will each time provide some reflections regarding their possible impact in practice. Special attention will also be paid to the pos-sible interaction of some of the most striking new articles introduced in the UN Models in relation to the so-called OECD/G20/Inclusive Framework Two Pillar Solution[2]. However, before doing that, I will briefly refer to previously done extensive impact research regarding the UN Model in Practice, discussing the differences between the various versions of the UN Model and of the OECD Model[3] preceding the UN Model 2017 and 2021 updates. Finally, some short conclusions and concluding remarks regarding the possible influence of the 2017 and 2021 UN Model updates on tax treaty practice will be given.
In the past, the IBFD staff carried out several large impact research projects regarding the UN Model in Practice. These research projects were focussed on the occurrence in practice in tax treaties of the distinctive provisions in the various versions of the UN Model deviating from the various versions of the OECD Model, covering all treaties and protocols concluded in the period from 1980–2013. The latest of these studies[4], which covers the period from 1st April, 1997, to 1st January, 2013, included an analysis of the occurrence of 30 such distinctive UN Model provisions in 1811 tax treaties and amending protocols concluded in that period. The results were published in an extensive article, which also referred to the results of the previous impact research undertaken in 1997, covering the occurrence of 26 such distinctive provisions in 811 tax treaties and amending protocols concluded in the period from 1st January, 1980, to 1st April, 1997.
Reference is made to the previously mentioned research with its elaborate tables and conclusions. In the context of this article, it is interesting to note that the number of deviations between the UN and the OECD Models has remained stable at around 30 over this very long period of time from 1980–2013, during which there were various versions of the UN and OECD Models. Furthermore, also the occurrence of the distinct-ive UN Model provisions was surprisingly stable over these years, with some provisions generally not being included in many tax treaties, whereas others were very frequently included. The highest level of occurrence was, as to be expected found in treaties concluded between States which are both not a member of the OECD, a lower level in treaties between a State member of the OECD and a State not-member of the OECD, whereas in treaties between OECD member States there was generally a lower but rather stable percentage of inclusion of distinctive UN Model provisions, with a remarkable note that with respect to 8 such distinctive UN Model provisions the inclusion of these in tax treaties between OECD member States was about the same as in treaties between the other two categories of treaty partners as mentioned above.
In the context of the treatment of services in the OECD and UN Models, it is in view of the Articles 8(3)(b), 12A and 12B which were amended or introduced in the UN Model updates 2017 and 2021, interesting to note what the research performed in 2013 has shown with respect to the treatment of services, either on a net profits basis (like Art. 5(3)(b) UN Model) or on a gross income basis (a withholding tax in either a self-standing provision, or as included services in Art. 12 UN Model, a kind of provision which did not occur in either the OECD or UN Models).
As regards Article 5(3)(b) UN Model (see the 2013 research[5]), it was respectively included in 58% of the tax treaties between two non-OECD countries, 35% in tax treaties between a non-OECD and an OECD country and even in 17% of the tax treaties between two OECD countries. Striking is also that compared with the 1997 research, the figures 2013 are about 50% higher as regards the first two mentioned categories combined (46% versus 31%) and as regards treaties between two OECD countries the provision was even included relatively much more (17% versus 2% in the 1997 research).
As regards taxation of services on a gross basis, reference is made to the the 2011 impact research[6], comprising 1586 treaties and protocols concluded in the period from 1st April, 1997, to 1st January, 2011. A provision on included services was included in respectively 5% in treaties between non-OECD countries, 5% in treaties between OECD and non-OECD countries, and 8% in treaties between OECD countries. As regards self-standing provisions, they were included in respectively in 13%, 6%, and 0% in the same three categories of countries. If added together, these percentages were 18%, 11%, and 8% of the tax treaties concluded, which seems remarkable, as no provision of that kind was at that time included in any of the Models and the research was done already many years ago.
When considering the 2017 UN Model update and its possible impact, it seems useful to distinguish between the so-called BEPS-related changes to the Model and those which are not related to BEPS.
However, before discussing these changes, it can be interesting to get a very rough impression of the size of the 2017 UN Model update and also already of the subsequently discussed 2021 UN Model update, simply by looking at the following number of pages of the hardcover UN Models (including both Model texts and commentaries) 2011, 2017 and 2021, as follows:
– 2011 UN Model: 483 pages;
– 2017 UN Model: 804 pages;
– 2021 UN Model: 911 pages!
The following UN Model provisions were added or amended in 2017 as a result of the inclusion of the BEPS related tax treaty provisions:
Title and preamble – aim to avoid tax avoidance and treaty shopping
– Art. 1 (2 and 3) – transparent entities and saving clause;
– Art. 4 (3) – dual residence of entities;
– Art. 5 (4) – exception auxiliary and preparatory activities;
– Art. 5 (4.1) – anti-fragmentation;
– Art. 5 (5) – extended dependent agent PE;
– Art. 5 (7) – limited exception for independent agent;
– Art. 5 (9) – closely related enterprise;
– Art. 10 (2) (a) – anti-dividend stripping provision;
– Art. 13 (4 and 5) – anti-dilution provisions;
– Art. 23A (1) and 23B (1) – amendment relief of double taxation methods;
– Art. 29 – entitlement to treaty benefits.
I note that almost all these BEPS-related changes in the 2017 UN Model are identical to those in the 2017 OECD Model, with the exception of Art. 29(1–7) where the UN Model follows the text of the 2016 US Limita-tion of Benefits (hereafter) LOB provision, whereas the 2017 OECD Model includes a framework LOB provision, which leaves more room for bilateral negotiations of the text. Furthermore, it can be noted that the BEPS-related proposed amendment of Art. 25(1) of the Models, dealing with the possibility for taxpayers to be able to file a request for MAP to the competent authorities of both Contracting States concerned, was not adopted in the UN Model.
As regards the possible impact of these BEPS-related changes on country’s treaty practice, I would like to observe the following. Unfortunately, to my knowledge, there is no recent comprehensive impact research available on all the tax-treaty-related BEPS measures. Thus, I inevitably can only provide some more qualitative and personal impressions. In this context, I think one needs to make a distinction between the BEPS-related UN and OECD Model amendments which are part of the so-called Minimum Standards[7], and those which are not. As more than 140 countries have joined the so-called BEPS Inclusive Framework and have thus accepted the commitment to implement the Minimum Standards, whereas more than 90 countries have signed the Convention regarding the so-called Multilateral Instrument[8], which enables the inclusion of the BEPS-related tax treaty provisions in the country’s tax treaties without the bilateral renegotiation of the covered treaties, a quick and very large impact of the BEPS provisions which are Minimum Standards is to be expected and actually also occurs in practice according to the regular reports published by the OECD on the implementation of such Minimum Standards[9]. As regards the BEPS-related Model amendments not being part of the Minimum Standards, I can only point to the fact that many signatories to this previously-mentioned Convention regarding the Multilateral Instrument made reservations regarding such provisions (e.g. those regarding the anti-abuse provisions against avoidance of having a permanent establishment) and thus the uptake of those provisions in tax treaty practice will probably still be substantial, but much less than those related to these Minimum Standards. It should, however, also be observed that the fact that a country made a reservation regarding BEPS-related provisions in the before-mentioned Convention does not necessarily mean that such country is not prepared to accept such provision in a bilateral tax treaty context.
The following UN Model provisions were added or amended in 2017, but not as part of the the BEPS project:
– Art. 3 (1)(d) – Definition international traffic;
– Art. 5 (3)(b) – Same/connected project requirement;
– Art.’s 8A, 8B – Allocation profits from international traffic;
– Art. 10 (2)(a) – Threshold participation dividends;
– Art. 12A – Fees for technical services;
– Art. 13 (3) – International traffic;
– Art. 13 (4) – Exception use in active business;
– Art. 13 (5) – Comparable interests, 365 days;
– Art. 15 (3) – Allocation profits international traffic;
– Art. 22 (3) – Allocation capital International traffic;
– Art. 23A (4) – Double non-taxation;
– Art. 24 (4) – Fees for technical services;
– Art. 26 (2) – Use of information for other purposes.
Unfortunately, also here there has been to date no recent comprehensive impact research available on these. Thus, I can inevitably also here only provide some more qualitative and personal impressions. The provisions relating to international traffic, to the threshold for participation dividends (Art. 10(2)(a)), to capital gains on immovable property shares or participations in other entities (Art. 13(4)), to double non-taxation (Art. 23A(4)), and to information for other purposes (Art. 26(2)) are all identical to the similar provisions in the OECD Model 2017, so in due time a large impact of these provisions on treaty practice can be expected. As these provisions are not included in a Multilateral Instrument like the one mentioned in section 3.1. above, it seems justified to expect a much slower uptake due to the need for bilateral negotiation of the relevant tax treaties than for the BEPS-related provisions which have been included in such Multilateral Instrument. For the sake of completeness, it is pointed out that a kind of UN Multilateral Instrument, called Fast Track Instrument, enabling to faster include specific UN Model provisions in existing tax treaties, is being considered to be introduced by the UN[10].
The provisions included in Art.’s 5(3)(b) and 13(5) are specific for the UN Model and such distinctive provisions were already included in previous versions of the UN Model but have now been amended to provide more taxing rights to source States and are thus, generally speaking, more attractive for developing countries, but less acceptable for the OECD countries.
Especially the newly introduced Art. 12A, allowing taxation on a gross income basis of payments of fees for technical services are arising in the source State and paid to a resident of the other State, is contentious for the OECD countries. The contentious nature is clearly expressed in the Commentaries to that provision in the UN Model. It thus seems justifiable to expect a substantial uptake of Article 12A in tax treaties between developing countries but to a much lesser degree in tax treaties between non-OECD and OECD countries. I do, however, also refer back to the last paragraph of section 2 above, and note that the impact research over the period 1997–2011 shows that already self-standing similar provisions such as Art. 12A were included in 13% of the tax treaties concluded between two non-OECD countries and 6% between a non-OECD and a OECD country (plus an additional 5% for each before-mentioned category of treaties for included technical services in the royalty article). So, now that the taxation at source of such fees is expressed in a standard UN Model provision, further uptake of such provision in tax treaties can perhaps be expected. OECD countries might, however, also be even more reluctant to accept such provision, as it also covers fees for digitally-provided services without any in person presence in the source State, and thus may be considered to have an overlap with the Pillar One[11] solution regarding the digitalised economy.
In this respect, it is interesting to note the alternative text for Art. 12A[12]. In that text in principle no source taxing right is granted for services provided by a non-resident in a digital form without any physical presence in the country of the recipient of the service. However, a source taxing right is provided with respect to payments for any type of services which are either performed in the source country, or even if not provided in person in the source State if the fees are paid to a closely-related enterprise or person. Thus, in the alternative text, the problem of the definition of what constitutes technical services is avoided, whereas a source taxing right is granted if digital services have been provided by a closely-related company (to avoid base erosion through the payment of such fees between closely-related companies). Under this alternative text, fees may thus be taxable on the gross amount in the source State which would not be allowed to be taxed (on a net profits basis) under the extended scope of Art. 5(3) (b) mentioned above. However, fees for digital services provided purely digitally by non-resident third parties would not be covered by such alternative Art. 12A, nor by the extended Art. 5(3)(b).
The most interesting new articles included in the text of the 2021 UN Model are in my view:
– Art. 12B: Automated Digital Services
– Art. 13(6): Capital gains on the direct transfer of certain rights granted under the law of a Contracting State for the use of resources naturally present in that State, and
– Art. 13(7): Offshore indirect transfers of shares in companies or of comparable interest in an entity, if the alienator at any time during the 365 days preceding the alienation held (in)directly at least X% of the company or entity and at any time during that period the shares or interests derived more than 50% of their value (in)directly from property taxable under the preceding provisions of Art. 13.
Besides these, the following further amendments to existing articles or new articles have been included in the 2021 UN Model:
– Art. 1 (3 and 4): resp. saving clause and placeholder collective investment vehicles)
– Art. 3 (1) (g)): definition recognised pension funds)
– Art. 4(1), Art. 29(2)(e) and (g): respectively dealing with recognised pension funds
and collective investment vehicles
– Art. 7 (note on purchase removed)
– Art.’s 10, 11 and 12 (paragraphs 2: recipient intermediary in the 3rd State)
– Art. 10(2)(a): exclusion of partnerships as parent deleted
– Art.’s 23A (2 and 4), Art 24 (4) and Art. 29(2)(B)(1): consequential to inclusion of Art. 12B.
As regards the possible impact of these provisions, I limit myself to the in my view most interesting articles mentioned above and observe again that there is no (and in fact cannot yet be due to the very recent publication of these 2021 UN Model Articles) impact research available, and thus I can only provide some more qualitative and personal impressions regarding their possible impact.
Art. 12B UN Model is very contentious and politically-sensitive for OECD countries and for other countries that are members of the Inclusive Framework which want to stick to the so-called Two Pillar approach[13], and these countries most likely do not accept it in their tax treaties, as there is a clear tension and overlap with that Two Pillar approach, and especially with Pillar One. Furthermore, this Article 12B has an even broader scope than Art. 12A as the services covered are not limited to services tailored to the specific customers and even also as payments by individuals for services for own use are covered, which is not the case with Article 12A. Thus, I do not expect an uptake of Art. 12B in tax treaties with countries part of the Inclusive Framework if in the end Pillar One (as still to be further elaborated) is accepted by them. Matters may be different if Pillar One fails to be successfully agreed upon and implemented, as Art. 12B might then perhaps be considered as a reasonable alternative.
Art. 13(6). The underlying approach in this article is already known from specific tax provisions included in tax treaties relating to capital gains made on licences granted in the context of the so-called extractive industries. However, in this newly introduced Art. 13(6), the scope of government licences is much broader and as most countries grant licences of the types covered and many times large amounts of money are involved when they are alienated, this provisions may be far less contentious for OECD countries than 12B, and, in fact, some of these OECD countries may even want to include this provision in their own tax treaty policy.
Art. 13(7). This article has been formulated in a similar way as Art. 13(4) UN Model and is probably less contentious for OECD countries than the politically very sensitive 12B, but still contentious as its scope goes far beyond the indirect sale of immovable property and of government rights granted, and thus could provide an extension of taxing rights to source States not in line with OECD country policies, whereas also a proper relief of double taxation seems not assured. Thus, for OECD countries, an uptake may perhaps only take place with respect to an amended version of Art13(7) limited to the capital gains on offshore indirect sale of the type of government licences covered under Art. 13(6) of the 2021 UN Model.
Having been asked to discuss the possible impact of the 2017 and 2021 updates of the UN Model on tax treaty practice which I have done in the previous sections, but also realising the special technical but more importantly possible political sensitivity of the relationship between the inclusion of Art.’s 12A and 12B and the digitalised economy approach as expressed in the OECD/G20/Inclusive Framework two Pillar solution, I would like to deal in a bit more detail with the relationship between the two. If Art.’s 12A and 12B would be incompatible with that two Pillar approach an uptake of these Articles in tax treaties would seem less likely in particular for countries which in the Inclusive Framework reached agreement on the two Pillar approach, at least as long as that approach is indeed going to be agreed upon and implemented.
For an impression of the relationship between these articles and Pillar One (which briefly stated grants additional taxing rights on the so-called Amount A to market countries with respect to a part of the excess profits, determined on an amended commercial accounting profits basis, realised by the around 100 largest companies in the world), it is important to realise that the additional granting of taxing rights under Pillar One is made conditional on countries not introducing or withdrawing any type of domestic digital service taxes (hereafter DST). To get a better impression of this possible tension between Art.’s 12A and 12B, and Pillar One, it seems necessary to look at the draft Multilateral Convention on Digital Service Taxes and similar measures[14].
Article 37(1)[15] of that draft Multilateral Convention contains the obligation on Parties to the Convention to remove any measure listed in Annex A ( List of Existing Measures Subject to Removal) as from the date on which the convention enters into effect with respect to that Party. However, such (draft) Annex A has to date not yet been published, so the impact on the domestic tax legislation providing the legal basis to levy tax in the source State in accordance with Art.’s 12A and 12B, cannot yet be determined on the basis of this Article 37(1).
Article 38[16] of that draft Convention, Provision Eliminating Amount A Allocations for Parties Imposing DST’s and Relevant Similar Measures, contains the criteria to qualify domestic tax legislation as a DST or Similar Measure. Article 38(2) defines the term “digital services tax or relevant similar measure”, whereas Article 38(3) clarifies that the term shall not include certain types of mentioned legislation (so carves out certain types of domestic tax legislation from the term).
As regards Art. 12A UN Model, I conclude when reading Art. 38(2) that the criteria listed in that provision to qualify an underlying tax as a DST or similar measure do not seem to be met and accordingly a tax levied on the gross amount of fees for technical services does not seem to be qualified as DST or similar measure. Thus, the inclusion of Art. 12A and the existence of domestic legislation enabling to exercise the taxing right granted under that article, would seemingly not negatively impact an entitlement to receive a taxing right under Pillar One. I also observe that an obligatory removal of such in practise already longstanding source taxing rights which have already been regularly included in tax treaties as mentioned in the last paragraph of section 2 above, would also seem unacceptable for developing countries which already included an Art. 12A like provision in their tax treaties and have the enabling domestic legislation in their tax law.
As regards Art. 12B UN Model, I note that underlying domestic legislations enabling to exercise the taxing right allocated under Art. 12B, are not uniform and thus one or more of the criteria to qualify domestic legislation as DST or similar measure under Art. 38(2) of the draft Convention may depending on the type of legislation be considered met. If that would be the case, an Amount A allocation would not be granted to a country having such legislation unless such legislation is withdrawn. In other words, this may cause a true obstacle for countries to agree upon and include an Article 12B in their tax treaties, assuming that Pillar One is finally agreed to and the Convention signed by a country.
In view of the importance of this determination for yes or no including Art.’s 12A and 12B UN Model, more clarity on these matters is of course desirable via either the before-mentioned Annex A or further analysis of specific domestic legislations.
Finally, although I am not aware of any official publication in this respect, in literature[17] reference is made to an ongoing discussion in the Inclusive Framework where some participating countries take the view that also taxes levied in accordance with a tax treaty on the gross amount of passive income arising in that source State, as customary already for many years in many tax treaties, should reduce the amount of the entitlement to an amount A. If that would be the case, that might drastically reduce the amount A allocation to in particular developing countries and thus make agreeing to Pillar One even more doubtful for these. If they would, however, accept such elaboration of Pillar One, including Articles 12A or 12B in tax treaties would probably not make much sense anymore. However, it should be observed that such reduction has not been part of the Statement on a Two-Pillar solution as agreed to and released in October 2021[18] and thus it would seem doubtful to me whether such reduction can still be included in Pillar One at this stage.
For an impression of the relationship between the Articles 12A and 12B, and Pillar Two (which briefly stated introduces a global minimum corporate effective income tax rate of 15% calculated on an amended commercial accounting profits tax base of jurisdictional constituent entities of a covered multinational company), it seems necessary to look at the Global Anti-Base Erosion Model Rules (Pillar Two) of 14[19]. The further Administrative Guidance on the Global Anti-Base Erosion Model Rules (Pillar Two) of 1 February 2023[20] did not, in my impression, shed any further light on this relationship.
In particular, relevant for the relationship between Art.’s 12A and 12B seem to be Art. 4(2)[21] and (3)[22] of the Model Rules (Pillar Two) dealing respectively with the Definition of Covered Taxes, and Allocation of Covered Taxes from one Constituent Entity to another Constituent Entity.
When reading Art. 4(2)(1) Model Rules (Pillar Two), both types of domestic taxes enabling States to exercise a taxing right granted under Art.’s 12A and 12B would seem to qualify as covered taxes on income or profits, like other withholding taxes on the gross amount of parts of the profits of a recipient company which is arising in the source State. If in case of Art. 12B(3) a taxpayer opted for taxation on a qualified net profits basis, also such taxation on deemed net income would seem to qualify as a covered tax. Moreover, the types of domestic taxes enabling to exercise the taxing right allocated to a source State under Art.’s 12A and 12B seem to me not to be excluded as covered tax under Art. 4(2)(2) Model Rules (Pillar Two). This is relevant, because if the domestic taxes enabling to exercise the taxing right allocated to a source State under Arti’s 12A and 12B would not be considered taxes covered, they would not count as taxes for determining the effective tax rate (ETR) under Pillar Two and thus inclusion of Art.’s 12A and 12B in tax treaties might, depending on the circumstances, be considered less attractive.
Finally, both taxes levied under Art.’s 12A and 12B would seem to have to be allocated to the recipient Constituent Entity for the purposes of determining the latter’s Effective Tax Rate (ETR), like withholding taxes on passive income, and Art. 4(3) Model Rules (Pillar Two) does not seem applicable to the taxes underlying Art.’s 12A and 12B.
The 2017 and 2021 updates of the UN Model have been large and also introduced some new provisions substantially increasing source country taxing rights. As far as I am aware no comprehensive research has yet been done regarding their impact on tax treaty practise and as it generally takes quite some time before new or amended provisions in the UN and OECD Models are effectively included in tax treaties, it is perhaps also still too early for such comprehensive impact research to be undertaken. Thus, I inevitably had to resort to a more qualitative and personal assessment of the possible impact of these updates on treaty practise.
In section 2 of this article, I dealt with the comprehensive impact research done with respect to distinct UN Model provisions predating the 2017 and 2021 updates, as this may give a feeling about the acceptability of some approaches and thus can be indirectly of interest when trying to assess the possible impact of some of the distinct provisions of the 2017 and 2021 UN Model updates.
Subsequently, I dealt in section 3 with the possible impact of the 2017 UN Model, making a distinction between the BEPS-related and the non-BEPS-related changes to the text of the UN Model. Almost all BEPS-related amendments (dealt with in section 3.1.) are identical in both the UN and OECD Models, thus substantially increasing the chance of these having an uptake in tax treaty practise.
However, within this category of BEPS-related changes a further distinction should in my view be made between provisions which constitute so-called BEPS Minimum Standards the inclusion of which is peer reviewed. Reporting of OECD on the implementation of the tax treaty related Minimum Standards shows that their uptake is impressive. The uptake of the other BEPS-related amendments will be smaller and slower.
In all BEPS-related tax treaty provisions the Convention implementing the so-called Multilateral Instrument is very helpful in realising a relatively fast uptake, but especially as regards the amendments which are not Minimum Standards a lot of reservations have been made, thus not leading to a very large and fast uptake of these in actual tax treaty practise. It should, however, not be forgotten that countries making such reservation may still be willing to include such provisions in the context of bilateral tax treaty negotiations, which will, however, only show over the longer period time, which it takes to bilaterally renegotiate tax treaties.
The possible impact of the non-BEPS-related amendments, is dealt with in section 3.2. As a number of these amendments have been included also in the OECD Model, their uptake may be expected to be substantial in the longer run.
However, in the 2017 UN Model update, also some very distinct provisions have been amended or newly introduced (especially Art.’s 5(3)(b), 12A), which will substantially strengthen source taxing rights if included in tax treaties. Thus, it may be expected that OECD countries will generally be reluctant or even not willing to accept such provisions (or against a high price to be paid for these with respect to other treaty provisions).
In particular, Art. 12A is contentious as can also be clearly seen from the commentaries to that provision in the UN Model 2017. Yet, it is pointed out that a more longstanding, albeit relatively limited, tax treaty practice of including a source taxing right with respect to fees for technical services already existed. Now that such provision in the form of Art. 12A is included in the UN Model, more countries may try to get it included in their tax treaties and an uptake, especially in tax treaties between developing countries, can be expected. OECD countries may, however, be even be more reluctant to accept Art. 12A than they were in the past in view of the overlap with the digitalised economy for which they feel the solution should be found in the Two Pillar solution. It is interesting to note that in the Commentaries to Art. 12A an alternative text is mentioned which would substantially strengthen source taxing rights with respect to any type of services provided by a resident of a contracting State in the other contracting State either in person or digitally if in the latter case the recipient of the service is closely related to the provider of such digital services.
In section 4, the amendments included in the 2021 UN Model are discussed and an attempt is made to assess the possible impact of these very recently introduced distinct UN Model provisions on tax treaty practise. Most attention has been devoted to the in the 2021 Model update newly introduced Art.’s 12B, 13(6) and 13(7).
Of these, Art. 12B seems most contentious, both from a technical and a political perspective, to OECD countries and other countries that joined the Inclusive Framework as this Article is perceived to be in conflict with the approach agreed to in the Two Pillar solution and especially with Pillar One. Thus, as long as Pillar One is still to be finalised and agreed to, not much uptake on Art. 12B may in my view be expected, especially not in tax treaties between developing and OECD / Inclusive Framework countries. This may, however, substantially change if Pillar One would not be finalised successfully, or the Multilateral Convention dealing with Pillar One would not be signed by many countries.
I do, however, expect a much better uptake of Art. 13(6), as being able to tax capital gains on the alienation of government licences (a policy aim recognised and realised in specific tax treaty provisions already for a long time in the extractive industries by developed and increasingly also by developing countries) can be substantial and seemingly increasingly important due to many other valuable licenses being granted by States.
Art. 13(7) may also enjoy a gradual uptake especially in treaties between developing countries but its scope may perhaps be considered too wide for OECD countries which generally accepted such source taxing rights for offshore indirect transfers only for immovable property (see Art. 13(4) of the OECD and UN Models) and thus may perhaps want to limit its scope to the licenses referred to in Art. 13(6).
In section 5, I have dealt with the compatibility of Art.’s 12A and 12B with Pillar One and Pillar Two, which assessment is of course based on the (draft) rules currently known regarding these Pillars. I looked at that as such compatibility may of course have an impact on the possible uptake of these provisions in tax treaty practise.
It seems to me that the domestic taxes enabling the exercise of the taxing rights expressed in Art. 12A are compatible with Pillar One, whereas depending on the various domestic legislations incompatibility of taxes aimed to exercise the taxing rights allocated by Art. 12B is more likely and thus a further obstacle to including Art. 12B in tax treaties. The latter also depends on the successful conclusion of Pillar One and its uptake in practise.
Finally, I have also dealt with the question whether the taxes underlying Art.’s 12A and 12B would be covered taxes under Pillar Two and have the impression they would be and that this aspect should thus not create an obstacle to including these in tax treaties from the perspective of Pillar Two.
Autor przedstawia przegląd najważniejszych zmian w tekście Konwencji Modelowej ONZ w spawie unikania podwójnego opodatkowania między państwami rozwiniętymi a rozwijającymi się dokonanych w latach 2017 i 2021 oraz możliwe interakcje pomiędzy dwoma nowymi najważniejszymi postanowieniami Konwencji Modelowej ONZ: art. 12A dotyczącym opłat za usługi techniczne oraz art. 12B dotyczącym dochodów z zautomatyzowanych usług cyfrowych oraz Filaru Pierwszego i Filaru Drugiego.
Organisation for Economic Development and Co-operation, Model Double Tax Convention on Income and on Capital (latest version: Paris: OECD 2017).
Public Consultation Document Pillar One- Amount A: Draft Multilateral Convention Provisions on Digital Services Taxes and other Relevant Measures, as published by the OECD on 20.12.2022.
Resolution of the UN General Assembly 22 December 2022 on a more inclusive tax co-operation, especially CRP1 on the Fast Track instrument “UN MLI” to implement Art.’s 12B and https://www.un.org/development/desa/financing/events/26th-session-committee-experts-international-cooperation-tax-matters, (access: 21.03.2023).
United Nations, Model Double Taxation Convention between Developed and Developing Countries (New York: UN, 2017 and 2021).
Wijnen W., de Goede J., The UN Model in Practice 1997–2013, IBFD, “Bulletin for International Taxation”, March 2014, no. 3.
Wijnen W., de Goede J., Alessi A., The Treatment of Services in Tax Treaties, IBFD, “Bulletin for International Taxation”, January 2012, no. 1.
Witholding Tax Emerges as Wedge in OECD Deal, Daily Tax Report 26.08.2022, Bloomberg.
https://www.oecd.org/publications/prevention-of-tax-treaty-abuse-fifth-peer-review-report-on-treaty-shopping-9afac47c-en.htm (access: 21.03.2023).
https://www.oecd.org/tax/beps/agreed-administrative-guidance-for-the-pillar-two-globe-rules.pdf (access: 31.03.2023).
https://www.oecd.org/tax/beps/beps-actions/ (access: 21.03.2023).
https://www.oecd.org/tax/beps/statement-on-a-two-pillar-solution-to-address-the-tax-challenges-arising-from-the-digitalisation-of-the-economy-october-2021.htm (access: 31.03.2023).
https://www.oecd.org/tax/beps/tax-challenges-arising-from-the-digitalisation-of-the-economy-global-anti-base-erosion-model-rules-pillar-two.htm (access: 31.03.2023).
https://www.oecd.org/tax/treaties/multilateral-convention-to-implement-tax-treaty-related-measures-to-prevent-beps.htm (access: 21.03.2023).
