to article 2(2) MLl
Adv LLM thesis
submitted by
Gaetano Manzi
in fulfilment of the requirements of the
'Advanced Master of Laws in International Tax Law'
degree at the University of Amsterdam
supervised by
PERSONAL STATEMENT
Regarding the Adv LLM Thesis submitted to satisfy the requirements of the 'Advanced Master of Laws in International Tax Law' degree:
1. I hereby certify (a) that this is an original work that has been entirely prepared and written by myself without any assistance, (b) that this thesis does not contain any materials from other sources unless these sources have been clearly identified in footnotes, and (c) that all quotations and paraphrases have been properly marked as such while full attribution has been made to the authors thereof. I accept that any violation of this certification will result in my expulsion from the Adv LLM Program or in a revocation of my Adv LLM degree. I also accept that in case of such a violation professional organizations in my home country and in countries where I may work as a tax professional, are informed of this violation.
2. I hereby authorize the University of Amsterdam and IBFD to place my thesis, of which I retain the copyright, in its library or other repository for the use of visitors to and/or staff of said library or other repository. Access shall include, but not be limited to, the hard copy of the thesis and its digital format. 3. In articles that I may publish on the basis of my Adv LLM Thesis, I will include the following statement in a footnote to the article’s title or to the author’s name:
“This article is based on the Adv LLM thesis the author submitted in fulfilment of the requirements of the 'Advanced Master of Laws in International Tax Law' degree at the University of Amsterdam.”
4. I hereby certify that any material in this thesis which has been accepted for a degree or diploma by any other university or institution is identified in the text. I accept that any violation of this certification will result in my expulsion from the Adv LLM Program or in a revocation of my Adv LLM degree.
signature:
name: Gaetano Manzi
Table of Contents
Table of Contents ... III
List of Abbreviations and Definitions used ... IV
Executive Summary ... V
Main Findings ... VI
1.
Introduction ... 1
2.
General and special interpretation’s rules ... 2
3.
Main outcomes from the discussions around Art. 3(2) OECD MC ... 7
3.1. Definition of “context” ... 7
3.1.1. In general ... 7
3.1.2. First Approach ... 8
3.1.3. Second approach ... 9
3.2. When and whether the “context otherwise requires” ... 9
3.2.1. First Approach ... 9
3.2.2. Second Approach ... 10
3.2.3. Third Approach ... 11
3.3. The relevance of Art. 3(2) OECD MC for interpretation purposes ... 11
3.3.1. The legal status of the OECD MC Commentaries ... 15
4.
Interpretation of the terms used but not defined in the MLI ... 16
4.1. Introduction ... 16
4.2. Why and whether Art. 2(2) MLI is the starting point for interpretation purposes ... 18
4.2.1. First approach: the MLI directly amends/implements the relevant CTAs ... 18
4.2.1.1. Compatibility issues between Art. 2(2) MLI and Art. 3(2) CTA ... 19
4.2.2. Second approach: the MLI has to be read together with the relevant CTAs ... 20
4.2.2.1. Compatibility issues between Art. 2(2) MLI and Art. 3(2) CTA ... 22
4.3. The rule provided by Art. 2(2) MLI ... 25
4.4. Possible extension of the considerations made on Art. 3(2) OECD MC ... 26
4.5. The applicable principles and means of interpretation ... 28
4.5.1. The interpretation principles ... 28
4.5.2. The most relevant means of interpretation ... 29
4.5.2.1. The Explanatory Statement to the MLI ... 30
4.5.2.2. The BEPS projects ... 33
4.5.2.3. The 2017 OECD MC Commentary ... 35
List of Abbreviations and Definitions used
(alphabetical list of abbreviations used in the Thesis)
Art., Arts. Article, Articles
Art. 3(2) CTA Provision similar to Art. 3(2) OECD MC included in a CTA
BEPS OECD’s Action Plan on base erosion and profit shifting
C-, T- Court Case Number
Contracting Jurisdiction(s) A State or territory which is part of the MLI Contracting State(s) A State or territory which is part of a DTT
Convention MLI
CTA, CTAs Covered Tax Agreement, Covered Tax Agreements
DTT, DTTs Double Tax Treaty, Double Tax Treaties
OECD Organisation for Economic Cooperation and Development
OECD MC OECD Model Tax Convention (2017)
MLI Multilateral Convention to Implement Tax Treaty Related Measures to Prevent Base Erosion and Profit Shifting
PE, PEs Permanent Establishment, Permanent Establishments
UN United Nations
UN MC UN Model Tax Convention (2017)
Executive Summary
This Paper aims to assess how the terms used but not defined in the MLI should be interpreted.
The entire interpretation process is unclear; this derives, first, from the doubted nature of the MLI itself. Some authors qualify the MLI as an autonomous treaty, while others believe that it is a sort of Protocol that directly amends the relevant CTAs and then loses its effectiveness. Furthermore, the special interpretative rule provided in the MLI (Art. 2(2) MLI) is vague and general and it has to be interpreted in its turn. In the light of the above, two main research questions are addressed through this Paper. The first one aims to assess which is the appropriate interpretative process to be followed for the interpretation of the MLI “undefined terms”. This question entails the assessment of the following sub-questions: (a) which is the nature of the MLI and (b) how it interacts with the relevant CTAs for interpretation purposes. The second main question aims to assess which are the applicable rules for the interpretation of the terms under investigation.
In relation to the first question, this Paper shows that the interpreter should start the analysis from Art. 2(2) MLI. According to the MLI drafters and most authors, the MLI is an autonomous and independent agreement which has to be read together with the relevant CTAs. The MLI falls within the definition of “treaty” provided by Art. 2 VCLT and can be considered “an international agreement concluded between States in written form and governed by international law”1. If the MLI is considered an autonomous treaty, Art. 2(2) MLI is
not directly integrated in the relevant CTAs. Besides, it maintains a specific scope which does not overlap with the one of Art. 3(2) CTA2. Since these rules have different scopes no conflicts should derive from their
application and no compatibility rules seem needed in this respect3. In light of the above, among the different
interpretative rules provided in the MLI and in the relevant CTAs, only those expressed in Art. 2(2) MLI should be applied by the Contracting Jurisdictions for the interpretation of the MLI “undefined terms”. Concerning the second question, instead, this Paper shows that Art. 2(2) MLI, assessed on a stand-alone basis, is only the starting point for the interpretation of the MLI “undefined terms”. In fact, it has limited relevance to their interpretation. Art. 2(2) MLI is not self-explanatory. It admits different interpretations of the terms at hand (in contrast with the MLI object and purpose), and it has to be interpreted in its turn. This interpretation cannot be carried out under the principles of interpretation provided by the different domestic system (since they will analogously lead to different outcomes). An autonomous interpretation under the VCLT should be done to comply with the MLI object and purpose and to determine the international law meaning of the terms under investigation. According to the VCLT, the interpretation has to be carried out in good faith, in accordance with the ordinary meaning these terms have in the context of the MLI and in accordance with its the object and purpose. The worth of these rules is that they “oblige” the interpreters to determine the meaning of the terms at hand taking into account all the elements provided in Art. 31 VCLT. Of course, this process will be exposed to the subjective interpretation of the judges or tax authorities; the interpretation is something subjective in nature, and objective criteria to carry out this process can be hardly established. However, this approach will at least introduce a sort of “minimum standards” of interpretation, “forcing” the interpreters to at least consider all the mentioned elements, leading to a more uniform interpretation of the MLI. In this process, the interpreter should rely on the Explanatory Statement, the BEPS project and – to a limited extent – on the 2017 OECD MC Commentary. For the reasons described in this Paper, under certain circumstances, these means of interpretation could be included in Art. 31 VCLT.
1 Art. 2 VCLT.
2 I.e. only Art. 2(2) MLI has the scope to interpret the terms used but not defined in the MLI. 3 No compatibility clauses have been provided by the MLI drafters with respect to Art. 2(2) MLI.
Main Findings
Art. 2(2) MLI is the starting point for the interpretation of the MLI “undefined terms”. The MLI has to be considered as an autonomous treaty to be read alongside the relevant CTAs. Art. 2(2) MLI is not directly incorporated in the CTA, and it does not share its scope with Art. 3(2) CTA. These provisions have different objects and scopes. As a consequence, no compatibility issue should arise among them for the purpose of Art. 30(3) VCLT. Both the provisions will apply simultaneously to interpret the CTA “undefined terms” (Art. 3(2) CTA) and the MLI “undefined terms” (Art. 2(2) MLI) respectively. Even in the lack of Art. 2(2) MLI there is no reason to think that such an interpretation would be made through Art. 3(2) CTA, since the latter rule has a different scope which the Contracting States did not agree to implement. In addition, leaving this assignment to Art. 3(2) CTA would be contrary to the object and purpose of the MLI and would lead to different interpretations of the MLI “undefined terms”; Art. 3(2) CTA, in fact, could rely on the domestic laws of the different Contracting States for interpretation purposes. Most likely, in the absence of Art. 2(2) MLI the MLI “undefined terms” would be interpreted according to the VCLT.
Notwithstanding the above, Art. 2(2) MLI is of limited relevance for the interpretation of the MLI “undefined terms”. From one side, it is true that this provision is the theoretical starting point for interpretation purposes; but from the other side, this provision does not help the interpreter to determine the meaning of the terms at hand. Art. 2(2) MLI, in fact, is vague and general and can lead to different approaches (which would be contrary to the object and purpose of the MLI itself). This provision has to be interpreted in its turn. This interpretation cannot be done through the general rules and principles provided in the legal system of the different Contracting Jurisdictions since it would analogously lead to different interpretations of the MLI terms. For this purpose, an autonomous interpretation under the VCLT rules and principles should be performed. Even if also these rules are exposed to the subjective views of the interpreters, they are able to set at least a “minimum standard” of interpretation, uniform for all the Contracting Jurisdictions. The application of these rules by the interpreters (e.g. judges and tax authorities) would better achieve the object and purpose of the MLI, interpreting the terms at hand in a more consistent way in all the Contracting Jurisdictions. For this purpose, under certain circumstances, the interpreter should be allowed to use the Explanatory Statement, the BEPS project and - to a more limited extent - the 2017 OECD MC Commentary.
1. Introduction
In order to counteract the artificial shift of profits to low tax jurisdictions, the OECD Committee on Fiscal Affairs - endorsed by the G20 leaders - released the BEPS project in November 2015. It consists of 15 Actions which deal with different measures aimed to tackle tax avoidance arrangements and ensure that profits are taxed where the relating business activities are carried out and the value is created4. These
measures need to be implemented, inter alia, in the bilateral tax treaties. For this purpose, Action fifteen provides with a Multilateral Instrument to modify bilateral tax treaties (MLI). This instrument has been already signed by 94 countries and aims to implement the BEPS measures in a fast, consistent and synchronised way. In fact, the implementation of this project on a treaty by treaty basis would have been extremely time-consuming, inefficient and ineffective. The MLI is accompanied by an Explanatory Statement aimed to reflect the agreed understanding of the negotiators with respect to the MLI itself. The object of this study is related to the interpretation of the MLI with particular reference to the “undefined terms” of the MLI. Many terms are used in the MLI but not defined therein (e.g. permanent establishment, business profits, residence, dividend, etc.). Their meaning, therefore, is not expressed in the MLI itself and should be determined through the interpretation process.
The interpretation process, however, is not clear for different reasons: first of all, not uniform consent has been registered on the nature of the MLI. Some authors have qualified the MLI as an autonomous treaty to be read alongside the relevant CTAs; others instead have considered the MLI as an amending treaty (a sort of Protocol) which directly modifies the relevant provisions of the CTAs and then loses its efficacy. The different approaches upheld in relation to the nature of the MLI entail different considerations on the compatibility and interaction between the MLI and the CTAs provisions and, therefore, the application of different interpretative processes for the purpose of this Paper. The above issues represent the first step of the analysis carried out in this Paper.
The second step of the present analysis focuses on the special and general interpretation rules which apply for the interpretation of the terms under investigation.
Art. 2(2) MLI provides with a special rule for the interpretation of the “undefined terms”. In particular, according to this provision: “2. As regards the application of this Convention at any time by a Party, any term not defined herein shall, unless the context otherwise requires, have the meaning that it has at that time under the relevant Covered Tax Agreement”.
This Paper evaluates the functioning and the relevance of this provision. At first sight, this rule seems of a simple application. In order to interpret an “undefined term”, the interpreter should look at the meaning it has in the relevant CTA unless the context of the MLI requires a different interpretation (or the use of a different mean of interpretation). However, the more this provision is analysed in detail, the more it is possible to understand its complexity. The rule is extremely vague and lacks any guidance for its interpretation and application. First, it does not provide with a definition of the term “context” 5.
Secondly, it does not clarify when (and whether) the “context otherwise requires”6.
4 OECD, Explanatory Statement to the Multilateral Convention to Implement Tax Treaty Related Measures to
Prevent Base Erosion and Profit Shifting [2017] in OECD Multilateral Convention ( MLI ), §1.
5 E.g. it does not specify which means of interpretation are comprised in its concept and can be used for
interpretation purposes; in particular, the MLI does not clarify the legal status in this respect of the Explanatory Statement of the MLI, the OECD MC and Commentaries and the BEPS project.
6 I.e. when the MLI’s context prevents the reference to the relevant CTA for interpretation purposes and requires
These two aspects are pivotal for the correct and uniform interpretation and application of the MLI’s “undefined terms” and, as a consequence, of the MLI itself and of the relevant CTAs. The vagueness of this rule can lead to non-uniform interpretations of the terms at hand. However, the purpose of the MLI is to implement the BEPS measures in the treaty network in a uniform and consistent way. Uncoherent interpretations of the terms at hand would lead to different interpretations of the MLI and of the relevant CTAs and, therefore would conflict with the MLI’s purpose.
The analysis of the above aspects has been structured in the following two main parts7.
The first part of this Paper goes through the main outcomes released with respect to Art. 3(2) OECD MC. In particular, the main approaches adopted to define the “context” of the OECD MC and to understand when/whether it requires an autonomous treaty interpretation are highlighted in the present work. In addition, the main considerations on the relevance of the article at hand for interpretation purposes and to what extent the principles of the VCLT can be used to interpret the “undefined terms” is within the scope of this Paper8.
The above part is a significant starting point for the analysis carried out in this work. Art. 2(2) MLI in fact almost reproduces the wording of Art. 3(2) OECD MC and has been drafted in a similar environment, triggering similar issues. Although the former article adds an extra layer of complexity to the interpretation of the “undefined terms”, the discussions and considerations made in the past with respect to Art. 3(2) OECD MC are of great relevance for the present analysis and to support the final conclusions. The second part of this Paper focuses on the specific issues relating to the interpretation of the MLI “undefined terms”.
The first step analysis aims to understand whether Art. 2(2) MLI is actually the starting point for the interpretation of these terms. In this regard, the nature of the MLI is assessed and consequently the compatibility of its interpretative provisions with those included in the relevant CTAs.
Thereafter, Art. 2(2) MLI is assessed in detail as well as the other applicable interpretation rules and principles (i.e. the VCLT).
In addition, it is investigated the possibility to extend the main outcomes and considerations released in relation to Art. 3(2) OECD MC to the MLI context. In particular, in this respect it is assessed the relevance of Art. 2(2) MLI for interpretation purposes. Lastly, established the applicable principles/rules, the analysis focuses on the legal status of the most relevant means of interpretation which could be used for the interpretation of the MLI “undefined terms” (i.e. the Explanatory Statement of the MLI, the BEPS project and the 2017 OECD MC and Commentary).
2. General and special interpretation’s rules
When terms used in the MLI or in the relevant CTAs are not defined therein they should be interpreted according to the relevant general and special interpretation rules.
7 The structure of this Paper shares few similarities with the one provided in: Svetlana Wakounig, Interpretation
of Terms Used in the Multilateral Instrument (M Lang et al eds, Kluwer Law International 2018). However, these Papers analyse the interpretation of the MLI “undefined terms” under different perspectives and with different approaches.
8 The discussions related to the dynamic or static approach to be upheld with respect to Art. 3(2) OECD MC are
For the purpose of this Paper, the relevant general rules are those codified in the VCLT and aim to promote a uniform interpretation of international treaties9.
At the time of the negotiation of the VCLT, there was no uniform consensus about the actual existence of common general interpretation rules and to the weight of the different principles and rules of interpretation10. Some Scholars accorded pre-eminence (or exclusive importance) to the literal/textual
interpretation of the terms, while others gave preference to the teleological approach, giving more relevance to the intention of the parties and/or to the object and purpose of the provisions to be interpreted11. The different interpretation approaches let to different applications of the international
treaties in each contracting jurisdiction. This circumstance was considered unacceptable, since each international treaty should be applied in a uniform and consistent way by all the signatory parties. In this context, the international community understood that that was the right time to negotiate some common rules for the interpretation of the international treaties. The result was the codification of a set of rules/principles which obliges the interpreter to take into consideration different elements and approaches during the interpretation process; a sort of pre-eminence was given to the text of the provisions to be interpreted, respecting the importance of the other (subjective) elements (e.g. intention of the parties, object and purpose of the provisions under investigation)12. The interpretative process
was conceived as a single combined operation able to encompass the analysis of the text, the object and purpose of the provision under investigation and the relating intention of the parties13.
These rules and principles apply to all tax treaties unless the parties have provided otherwise14; the
VCLT in fact is considered customary law and, therefore, it applies to all countries regardless whether or not they are signatory parties of the VCLT itself15.
For the purpose of this work, the most relevant provisions included in the VCLT are Arts. 31 and 32 VCLT.
Art. 31 VCLT has been headed “general rule of interpretation”; the singular form has been chosen to underline that it comprises different approaches (i.e. literal, systematic and teleological approach)16,
which should be adopted and combined by the interpreter to understand the meaning of a treaty term. Although there is no hierarchy among these elements17, Art. 31 expresses the concept that the “text
must be presumed to be the authentic expression of the intentions of the parties”18; therefore, the
interpreter should first analyse the wording of the provision under investigation and cannot start his/her
9 Official Commentary on a Preliminary Draft of The Vienna Convention [2018] in Materials on International, TP
and EU Tax Law (K. van Raad ed., International Tax Center Leiden 2018-2019), p. 2268.
10 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2268. 11 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2268. 12 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2269. 13 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2269.
14 Frank Engelen, Interpretation of Tax Treaties under International Law (Books IBFD 2004), chapter 10.1, p. 1;
Jörg Manfred Mössner, Klaus Vogel Lecture 2009 – Comments (2010) 64 Bulletin for International Taxation, p. 18; OECD MC Commentary (2017), Introduction, §16.2.
15 John F Avery Jones, Treaty Interpretation - Global Tax Treaty Commentaries - Global Topics IBFD, p. 8; Brian
J Arnold, The Interpretation of Tax Treaties: Myth and Reality (2010) 64 Bulletin for International Taxation, p. 5.
16 Andrés González Becerra, The Interpretational Approaches to the Vienna Convention - Application to (Tax)
Treaty Analysis (2011) 65 Bulletin for International Taxation, § 2.2; Michael Lang, Chapter 4: The Interpretation of Double Taxation Conventions, Introduction to the Law of Double Taxation Conventions (Second Revised Edition) (Books IBFD 2013), p.1.
17 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2269.
18 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2270; in this respect according
to note no. 6 at p. 2270 of the same document, the International Court in many occasions underlined that the function of the interpretation is not to read into treaties “what they do not, expressly or by implication, contain”.
analysis from the “intention of the parties”19. However, the text of the provision is only the starting point
since thereafter the interpreter could/should take into account also all the other mentioned elements20.
In particular, according to the first paragraph of Art. 31 VCLT a treaty shall be interpreted:
(i) in good faith. This means that the interpretation should be carried out loyally and honestly21 in
accordance with the agreements reached by the parties and the pacta sunt servanda rule provided by Art. 26 VCLT22. This element is considered essential to protect the legitimate
expectations of the parties and avoid abusive application of the relevant treaty23;
(ii) in accordance with the ordinary meaning to be given to the terms of the treaty. This means that the interpreter should literally assess the term under investigation, understanding what it actually says before what the parties intended to say with that term24. The ordinary meaning is not
necessarily the dictionary meaning but rather it relates to its common understanding and usage25,
i.e. its “daily-life” meaning26. In this respect it has been pointed out that “the more complicated or
artificial a term is, the less does an ordinary or a daily-life meaning exist”27. On the other end,
when an ordinary meaning exists and it is sufficiently clear, this does not prevent the interpreter to test it according to the other interpretation approaches (e.g. teleological approach)28. The
possible lack of an ordinary meaning or the existence of different ordinary meanings for a single term requires a further investigation of the term in its context, according to a systematic approach29 (see infra) in order to find and adopt the more consistent and reasonable one30;
(iii) in its context. The ordinary meaning of a term should not be evaluated in abstract or merely upon particular phrases but in the context of the treaty as a whole and in accordance with its environment31 (see infra for the definition of “context”);
(iv) in the light of its object and purpose (teleological and historical interpretation approach). The treaties are agreed by the parties for a specific purpose, in a specific social, historical and legal environment32; therefore, they should be interpreted in accordance with a meaning that promote
the intention of the parties rather than frustrates it33.
Art. 31(2), (3) and (4) VCLT further elaborate the general rule provided in Art. 31(1) above. In particular, the second paragraph clarifies the meaning of the term “context”. It comprises the text, the preamble and annexes of the relevant treaty34. In addition it encompasses any agreement relating to the treaty
which was made between all the parties in connection with its conclusion, and any instrument which
19 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2269. 20 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2269.
21 Edwin Van der Bruggen, Unless the Vienna Convention Otherwise Requires: Notes on the Relationship between
Article 3(2) of the OECD Model Tax Convention and Articles 31 and 32 of the Vienna Convention on the Law of Treaties (2003) 43 European taxation, note 32, p. 145.
22 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2270. 23 Van der Bruggen (n 21), p. 145.
24 Engelen (n 14), chapter 10.2, p.1.
25 González Becerra (n 16), § 4.2-4.3; Ekkehart Reimer, Interpretation of Tax Treaties (1999) 39 European
taxation, p. 459.
26 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2270; Reimer (n 25), p. 459. 27 Reimer (n 25), p. 459.
28 Reimer (n 25), p. 459. 29 Reimer (n 25), p. 459. 30 Reimer (n 25), p. 459.
31 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2270; González Becerra (n 16),
§ 4.3.
32 González Becerra (n 16), § 4.5.
33 Arnold (n 15), p. 5; González Becerra (n 16), § 4.5. 34 González Becerra (n 16), § 4.3.
was made by one or more parties in connection with the conclusion of the treaty and accepted by the other parties as an instrument related to the treaty35.
According to the third paragraph, few subsequent agreements or practices and international law rules form part of the “context”, i.e. (i) any subsequent agreement or practice regarding the interpretation of the treaty or the application of its provisions agreed by the parties and (ii) any relevant rules of international law applicable in the relations between the parties36.
Finally, according to the fourth paragraph, prevalence should be given to the special meaning of a term if it is established that the parties so intended37. This rule gives preeminence to the intentions of the
parties over the ordinary meaning of a term to the extent that can be proven an agreement of the parties in this respect3839.
On the other hand, the role of Art. 32 VCLT is to support the general interpretation rules codified in Art. 31 VCLT40. It has a more limited scope and can be used by the interpreter only under certain
circumstances41. In particular, Art. 32 VCLT allows the use of the so called “supplementary means of
interpretation” (e.g. parallel treaties42 or preparatory works which can be seen as treaty’s historical
interpretation43). They can help the interpreter to confirm the meaning of a term resulting from the
application of Art. 31 VCLT or to determine the meaning of that term when the interpretation according to Art. 31 VCLT leaves it ambiguous or obscure or leads to a result which is manifestly absurd or unreasonable44. The materials used according to Art. 32 VCLT in fact could be considered less authentic
or less representative of the agreement reached by the parties45. However, neither Art. 32 VCLT nor the
related VCLT Commentary specify what should be regarded as “supplementary means of interpretation”; it, therefore, appears left open to the evaluation of the interpreters.
35 Van der Bruggen (n 21), note 72, p.147. The Author noted that with reference to the definition of “context”
provided by Art. 31(2) VCLT, some commentators pointed out that it is not an exhaustive list of what the context comprises. Contra: Avery Jones (n 15), § 3.4.5., p. 17.
36 Engelen (n 14), Chapter 10.4, p. 2-3. According to the Author in fact DTTs cannot be assessed without taking
into account the other international rules.
37 Avery Jones (n 15), § 3.4.11. 38 Arnold (n 15), p. 5.
39 Van der Bruggen (n 21), p. 149. The Contracting State which wants to apply the treaty and to give a special
meaning to a given term have to prove that that was agreed among the parties.
40 Avery Jones (n 15), § 3.5.
41 Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2272. 42 Van der Bruggen (n 21), p. 150.
43 Nathalie Bravo, Introduction, A Multilateral Instrument for Updating the Tax Treaty Network (Books IBFD 2020),
p. 2.
44 Avery Jones (n 15), § 3.5.
45 Arnold (n 15), p. 7. The Author elaborate the process to be followed by the interpreter who wants to adhere to
the provisions of Arts. 31 and 32 VCLT: “(1) consider both the authentic material covered by Art. 31 and supplementary materials covered by Art. 32; (2) determine the meaning of the words in question exclusively on the basis of the Art. 31 material; (3) determine if that meaning is ambiguous, obscure, manifestly absurd, or unreasonable; (4) if the result of step (3) is that the meaning is not ambiguous, obscure, manifestly absurd, or unreasonable, determine if the Art. 32 materials confirm the meaning reached in step (2); (5) if the Art. 32 materials do confirm the meaning reached in step(2),that is the end of the matter; (6) if the Art. 32 materials do not confirm the meaning reached in step (2),that is also the end of the matter because in this situation the Art. 32 materials must be disregarded (in sum, the supplementary materials are irrelevant in any case where the meaning reached in step (2) is not ambiguous, obscure, manifestly absurd, or unreasonable, because they cannot be used to alter the result reached in step (2)); and (7) if the result of step (3) is that the meaning is ambiguous, obscure, manifestly absurd, or unreasonable, determine the meaning on the basis of the material covered by both Arts. 31 and 32”.
Notwithstanding the above, also special interpretation rules can/should be used by the interpreter to ascertain the meaning of a treaty term46. In general, they qualify as “special rules” when they are
applicable only in respect to certain treaties47 and/or have limited purposes.
In this respect, the MLI provides with a special interpretation rule in paragraph 2 of Art. 2, headed Interpretation of Terms. This provision literally states that: “2. As regards the application of this Convention at any time by a Party, any term not defined herein shall, unless the context otherwise requires, have the meaning that it has at that time under the relevant Covered Tax Agreement”. This rule requires the interpreter to investigate the meaning of the given “undefined term” in accordance with the meaning that it has under the relevant CTA, unless the “context” of the MLI requires a different interpretation or approach. As we will see infra in further details, Art. 2(2) MLI has a limited scope. In fact, it applies only to interpret the terms that are used in the MLI but are not defined therein (the terms that are not used in the MLI fall outside the scope of this rule)48. Besides, this rule is applicable only to
the CTAs included by a given country or territory in the MLI framework.
A similar special provision is provided in Art. 3(2) OECD MC49, generally included in Art. 3(2) of the
CTAs. This rule is very similar to Art. 2(2) MLI. Actually, it can be said that the latter provision has been built on the basis of Art. 3(2) OCED MC since it has been drafted in a similar context and almost reproduces its wording (see infra for further details).
Art. 3(2) OECD MC only applies whether it is included in a given DTT and concerns only the interpretation of the terms that are used in the DTT but are not defined therein50 (therefore this rule
should not be used to interpret unclear general treaty concepts51). In particular, it provides that the
“undefined terms”, “unless the context otherwise requires or the competent authorities agree to a different meaning pursuant to the provisions of Article 25, have the meaning that it has [they have] at that time under the law of that State for the purposes of the taxes to which the Convention applies, any meaning under the applicable tax laws of that State prevailing over a meaning given to the term under other laws of that State”. According to this rule, when the meaning of a term in the CTA is not defined, the interpreter should look at the meaning that it has under the relevant domestic law; but this is admissible only whether the “context” of the DTT does not require otherwise, e.g. it does not require an autonomous treaty interpretation.
In general, the special rules take precedence over the general ones52; consequently, in theory Art. 2(2)
MLI and/or Art. 3(2) OECD MC should lead the interpretation process of the “undefined terms” over the general principles and rules laid down in the VCLT.
However, most scholars believe that the VCLT applies to tax treaties in any case and that its rules and principles can/should be used by the interpreters to interpret Art. 2(2) MLI and Art. 3(2) OECD MC and the respective “undefined terms”53. In particular, the VCLT rules should be used in this respect since the 46 E.g. if any special rule is provided in the treaty under investigation.
47 E.g. those in which they are included.
48 Klaus Vogel and Rainer Prokisch, Interpretation of Double Conventions, Generalbericht (IFA 1993 Florenz)
[1993] CDFI (cahiers de droit fiscal international), Vol. LXXVIIIa, p. 79.
49 Engelen (n 14), chapter 10.10.2; Johannes Becker, Ekkehart Reimer and A Rust, Klaus Vogel on Double
Taxation Conventions (Kluwer Law International 2015), p. 207. The authors point out that Art. 3(2) OECD MC can be seen also as a general rule of interpretation as compared to other special rules of interpretation provided for specific provisions in the OECD MC (e.g. definitions of terms provided by Artt. 3(1), 4, 5, 10(3), 11(3) OECD MC etc. - although to the extent that undefined terms are used in those definition a fall back to Art. 3(2) OECD Mc is still possible); but the same article is a special rule as compared to the VCLT.
50 Becker, Reimer and Rust (n 49), p. 207. 51 Vogel and Prokisch (n 48), p. 79. 52 Engelen (n 14), § 10.10.2. 53 Engelen (n 14), § 10.1, p. 1.
meaning of the mentioned provisions is not self-explanatory; such provisions need further clarification or interpretation5455.
3. Main outcomes from the discussions around Art. 3(2) OECD MC
As anticipated in the Introductory chapter, the present analysis departs from the assessment of the main outcomes released from the discussions carried out around Art. 3(2) OECD MC56. The reason behind
this approach is that Art. 2(2) MLI - which is the core of this work - is very similar to Art. 3(2) OECD MC. The wording of the provisions, the environment in which they were drafted and also the issues which arise from their application appear similar. Therefore, the most reliable contributions issued with respect to Art. 3(2) OECD MC are of great importance for the analysis of Art. 2(2) MLI and to support the conclusions of this Paper.
Many authors analysed in detail Art. 3(2) OECD MC. In particular, they focused on two main aspects: the meaning of the term “context” and the understanding of when/whether it “otherwise requires”. The ascertainment of these issues is critical for the correct interpretation of the treaty’s “undefined terms”; the “context” in fact could prevent the interpreter from assessing the terms at hand according to their own domestic law. In the international community, a uniform consensus on the interpretation of this provision has not been reached so far. In the following paragraphs are highlighted the main approaches upheld in this respect.
3.1. Definition of “context” 3.1.1. In general
Apparently, the origins of the use of the concept of “context” comes from the British law57. According to
this statute, the reference to the “context” of the relevant law for interpretation purposes appears to be a common practice58. In this respect, the concept of “context” has been traditionally broadly conceived
to include the “historical and developmental background of a provision” 59 and/or “the surrounding text
of a word or passage used to determine the meaning of that word or passage; setting or environment”60
and/or “the statute as a whole”61.
For what concern the DTTs, the concept of “context” has been embedded for the first time in the UK-USA DTT (1945) in a provision similar to Art. 3(2) OECD MC62; thereafter it has been included in the
OECD MC and in the vast majority of DTTs. However, the reason for this “success” is still not very clear in literature63, considering the many issues triggered by this provision.
54 Engelen (n 14), § 10.10.2; Avery Jones (n 15), p. 54. See also Frank van Brunschot, The Judiciary and the
OECD Model Tax Convention and Its Commentaries (2005) 59 Bulletin for International Taxation, p. 8. The author believes that often the wording of the OECD MC and Commentary is intentionally not completely clear, since these documents have to observe a “double agenda”: (i) avoid double (non) taxation and (ii) prevent as much as possible the erosion of the states’ taxing rights. Consistent: Maarten J Ellis, The Influence of the OECD Commentaries on Treaty Interpretation - Response to Prof. Dr Klaus Vogel (2000) 54 Bulletin for International Taxation, p. 618.
55 Van der Bruggen (n 21), p.155.
56 A comprehensive and exhaustive analysis of the literature released with respect to Art. 3(2) OECD MC is outside
the scope of the present Paper. Only the main outcomes/conclusions are taken into consideration as starting point for the analysis on Art. 2(2) MLI.
57 Avery Jones (n 15), § 5.1.1.
58 Van der Bruggen (n 21), note 24, p. 144. 59 Vogel and Prokisch (n 48), p. 82. 60 Van der Bruggen (n 21), p. 144.
61 Van der Bruggen (n 21), note 24, p. 144. 62 Avery Jones (n 15), § 4.3.1.
It has been pointed out that the “context” of Art. 3(2) OECD MC is not the same provided in Art. 31(2) VCLT64. In general, the former is intended broader than the latter. This can be due to the fact that, as
we have already seen in the previous § 2, Art. 31(2) VCLT defines the meaning of “context”. Differently, Art. 3(2) OECD MC remains vaguer, without defining the concept of “context”. No further explanations are given by the OECD MC Commentary in this respect. Therefore, the interpreter has apparently no limitations for the determination of its meaning.
The OECD MC Commentary, in fact, gives to the interpreter only some further clarifications to determine the concept at hand. In particular, it clarifies that the title and the preamble of the treaty form part of the DTT’s context and “constitute a general statement of the object and purpose of the Convention”65. It also
specifies that the general rules of interpretation provided by the VCLT apply to the DTTs. Besides, it states that the OECD MC Commentary is not legally binding for the contracting state but “can nevertheless be of great assistance in the application and interpretation of the conventions”66. Lastly, it
clarifies that the “context” “is determined in particular by the intention of the Contracting States when signing the Convention as well as the meaning given to the term in question in the legislation of the other Contracting State”67. The last paragraph seems to point out that the DTT’s context does not
comprise only the title and the preamble of the treaty. It should be rather defined on a case by case basis, looking at the intention of the parties (therefore using a teleological approach) and also the relevant domestic laws.
Given the vagueness of the above “clarifications”, important discussions have been carried out in literature around the interpretation of Art. 3(2) OECD MC. Two main approaches have been followed by the interpreters in this respect. They are generally linked to a different understanding of the entire interpretation process and of the weight attributed to the concept of “context”. They are described below and should be read alongside with the considerations made in the subsequent § 3.2.
3.1.2. First Approach
According to the First Approach the term “context” should be intended in a broad sense, including all possible means of interpretation.
This view is generally shared by Vogel68, Avery Jones69 and other important Authors70. According to
them, the meaning of “context” is broader than the one provided by Art. 31(2) VCLT and comprises the text of the treaty, any supplementary means of interpretation, the object and purpose of the provision, the relevant domestic provisions of both Contracting States71 and the OECD MC and Commentary72.
This concept has the scope to compensate the negative consequences deriving from the necessary
64 Avery Jones (n 15), § 5.1.1.
65 See §16.2 of the Introduction to the OECD MC Commentary. 66 See §29 of the Introduction to the OECD MC Commentary. 67 See §12 to the Commentary on Art. 3(2) OECD MC.
68 Becker, Reimer and Rust (n 49), p. 212; consistent: Avery Jones (n 15), p. 72. 69 Avery Jones (n 15), p. 70.
70 Van der Bruggen (n 21), p. 153-155. This Author added another point to the mentioned consideration: in
particular, the fact that Art. 3(2) OECD MC makes a specific reference to domestic law does not mean that the latter law is the only mean of interpretation can be used by the interpreter to ascertain the meaning of the treaty “undefined terms”. This view in fact would lead to the qualification of the reference to domestic law made by Art. 3(2) OECD MC as lex specialis, able to displace the application of the VCLT; this would be inconsistent with international law and the treaty itself. The term “context” should include and refer to all the means of interpretation provided by the VCLT (therefore the context of Art. 3(2) OECD MC could be wider than the context of the VCLT).
71 This seems in accordance with § 12 of the Commentary to Art. 3 OECD MC.
72 Consistent on the possible inclusion of the OECD MC Commentaries in the concept of “context”: Mössner (n
reference to the relevant domestic law to interpret the “undefined terms”73; in fact, if the concept of
“context” is understood too narrowly the parties too frequently would rely on the relevant domestic law leading to a frequent not-uniform application/interpretation of the DTT (see infra)74. The same view is
shared by Avery Jones, even if in the Author’s opinion the definition of the “context” has another purpose, namely to understand the rare circumstances in which the domestic law cannot be used for the interpretation of the “undefined terms” (see infra).
3.1.3. Second approach
According to the Second Approach the term “context” should be interpreted in a narrow sense and should not include external means of interpretation.
In this respect, Engelen75 believes that the “context” should be defined as the environment in which the
terms under investigation are used. Therefore, it should comprise the entire treaty, its preamble, annexes, object and purpose and the mutual agreements reached by the parties on interpretation issues, but – in general – it should not include other external means of interpretation. In any case, however, the term “context” (as the entire sentence “unless the context otherwise requires”) should be interpreted in accordance with the VCLT principles, laid down in Arts. 31 and 32. In this respect, if external means of interpretation are agreed by the parties (in an express or tacit way) they would be included in the “context” pursuant to paragraph 2 of Art. 31 VCLT.
A similar view is shared by van Brunschot76, according to whom the “context” comprises the purpose of
the DTT, its structure, chapters, paragraphs and sentences. It does not include the OECD MC and Commentaries since they are often vague and contradictory. They can be of important help during the interpretation process, but they are not much more than that.
3.2. When and whether the “context otherwise requires”
The second critical issue with respect to Art. 3(2) OECD MC relates to the understanding of when/whether the “context” of the DTT requires an autonomous treaty interpretation of the terms under investigation and prevents the reference to the relevant domestic law. In this regard, three main approaches have been found in literature and depend on the weight that the different interpreters give to the treaty “context” for interpretation purposes.
3.2.1. First Approach
The First Approach is based on a textual analysis of Art. 3(2) OECD MC77. The use of “unless” from a
grammatical standpoint shows the less weight that the “context” has in the interpretation process. It should be used by the interpreter in exceptional circumstances; in this respect, preference should be given to the relevant domestic law (i.e. the domestic law of the interpreter who applies the DTT). In particular, Vogel78 believes that the interpretation according to the “context” is admissible only under 73 As it will be described infra in more detail, Vogel believes that according to the sentence “unless the context
otherwise requires” included in Art. 3(2) OECD MC when a term is used in the treaty but is not defined therein reference to the relevant domestic law should be preferred to ascertain its meaning; an autonomous treaty interpretation should be preferred only in rare critical circumstances.
74 Vogel and Prokisch (n 48), p. 81; Nathalie Bravo, Chapter 2: The Concept behind the Multilateral Instrument, A
Multilateral Instrument for Updating the Tax Treaty Network (Books IBFD 2020), p. 15.
75 Engelen (n 14), chapter 10.10.3. the view of the Author is consistent with the VCLT principles and opens to the
inclusion of e.g. the OECD Commentary in the concept of “context” as a tacit agreement between the parties (e.g. whether the treaty is signed between OECD countries, is identical to the OECD MC, the parties did not make any reservations to the OECD Commentary).
76 van Brunschot (n 54), p. 8. 77 Engelen (n 14), chapter 10.10.1.
rare circumstances, for important reasons79. This interpretation seems supported not only by the
grammatical interpretation of the word “unless” but also by the use of the verb “to require” (in French: “exige”). This verb has a strong meaning and seems to demand a treaty autonomous interpretation only in those critical cases in which it is necessary to move away from the ordinary interpretation process80.
Furthermore, according to this Author the text of Art. 3(2) OECD MC does not say that the interpretation according to the treaty “context” should be preferred.
In order to partially offset the negative consequences deriving from the close relation between the DTT and the domestic law81, as already anticipated, Vogel believes the term “context” should be interpreted
as broadly as possible in order to reduce the cases in which a reference to the domestic law is necessary. Furthermore, although originally an autonomous treaty interpretation used to be preferred in order to avoid double taxation/non-taxation issues, the introduction of the new approach82 in Art. 23
OCED MC should have solved this issue.
A similar view is shared by Engelen83, who agrees on the pre-eminence of domestic law in the
interpretation process. According to this Author the wording of Art. 3(2) OECD MC (i.e. “unless the context otherwise requires”) recalls the functioning of Art. 31(4) in the VCLT; this means that unless the parties have attributed a special (treaty) meaning to a given term (or it is possible to demonstrate that they so intended) the ordinary (domestic law) meaning of that term should be applied. In this respect, it does not matter whether this leads to different interpretation by the parties. In other words, when some terms have a specific meaning under the relevant domestic law, they are used in the DTT and the Contracting States did not have agreed to a different common (treaty) interpretation, the domestic law meaning of these terms should be applied. In this regard, it does not matter if Art. 3(2) OECD MC is included in the DTT, since the same result could be achieved through the VCLT.
3.2.2. Second Approach
The Second Approach is based on a teleological interpretation of the DTT, focusing on the purpose of the treaty84. According to this interpretation, the purpose of the DTT can be achieved only if the parties
have a common understanding of the meaning of the treaty’s terms; the opposite behaviour would lead both to the application of the DTT in two different ways and to a possible distortion of the parties’ obligations85.
This approach seems supported also by the fact that a mere textual approach would be contrary to the international law principles86. In fact, pursuant to Art. 31 VCLT any treaty should always be interpreted
also in accordance with its object and purpose. The OECD MC Commentary seems consistent with this view; in fact, §12 of the OECD MC Commentary to Art. 3(2) OECD MC clarifies that this paragraph “applies only if the context does not require an alternative interpretation”. For clarification purposes, the mentioned Commentary uses the words “only if” instead of the word “unless”, suggesting a sort of pre-eminence of the “context” in the interpretation process87; in other words, according to this clarification 79 E.g. when the interpretation according to the relevant domestic law renders the DTT not applicable or leads to
double (non) taxation; in case of changes in the domestic law of one Contracting State aimed to alter the application of the DTT’s distributive rules and increase that Contracting State’s taxing rights.
80 I.e. the interpretation according to the Contracting States’ relevant domestic laws. 81 I.e. a possible not uniform application of the DTT by the Contracting States. 82 Becker, Reimer and Rust (n 49), p. 209.
83 Engelen (n 14), chapter 10.10.2.
84 I.e. avoiding double (non) taxation and encouraging cross-border trades and investments. See: Van der Bruggen
(n 21), p. 149; Engelen (n 14), chapter 10.10.1.
85 Van der Bruggen (n 21), note 92, p. 149. 86 Van der Bruggen (n 21), p. 148.
87 Avery Jones (n 15), p. 70. According to the Author the clarification given by the OECD MC Commentary is not
the reference to the domestic law would be allowed only to the extent that an autonomous treaty interpretation cannot be carried out or does not lead to reasonable conclusions.
This view is shared by Lang88, according to whom the treaty meaning should prevail over the domestic
law meaning. The interpretation according to the treaty “context” entails all possible means of interpretation, e.g. a systematic, historical, grammatical and teleological interpretation89. For this reason,
the reference to the relevant domestic law should be made in exceptional cases, where the autonomous treaty interpretation – carried out using all the possible means of interpretation – does not lead to any (reasonable) results. This approach prevents that the Contracting States from developing different positions concerning the treaty’s obligations and, as a consequence, from applying the same DTT in a non-uniform way. As anticipated above, in fact, only if both the Contracting States have a common understanding of the DTT’s terms double (non) taxation, tax avoidance and evasion, can be avoided. Furthermore, according to Lang the new approach90 of Art. 23 OECD MC (supported by Vogel) has no
legal basis and can increase situations in which double taxation occurs.
A similar view is shared by Van Brunschot91 (Judge of the Dutch Supreme Court). According to this
Author during the interpretation process the main focus should be on the purpose of the treaty92, so that
among different possible interpretations - according to domestic law or treaty “context” - the latter should be preferred if the interpretation according to domestic law gives fewer remedies to double taxation. 3.2.3. Third Approach
The last Approach is apparently shared only by the Tax Authorities of some countries93. According to
this interpretation the sentence “unless the context otherwise requires” has almost no relevance and, therefore, the interpreter should always ascertain the meaning of the “undefined terms” relying on the relevant domestic law. The reference to the latter law would be avoided only in very exceptional cases. 3.3. The relevance of Art. 3(2) OECD MC for interpretation purposes
The terms that are used in a DTT but are not defined therein should be interpreted in accordance with Art. 3(2) OECD MC. However, this article in its turn contains terms that are general and not defined94.
Therefore, before giving the correct interpretation to the “undefined terms” under investigation is necessary to interpret and determine the meaning of the terms used in Art. 3(2) OECD MC.
As already anticipated above, the problem is that no guidance is provided to the interpreters in this respect95.
Considering the many different approaches discussed in § 3.1., it seems that the interpreters are substantially free to establish the meaning of “context” and, in particular, which means of interpretation this concept comprises. Similarly, they seem free to uphold the pre-eminence of the domestic law or of
88 Michael Lang, 2008 OECD Model - Conflicts of Qualification and Double Non-Taxation (2009) 63 Bulletin for
International Taxation, p. 206; Michael Lang, Conflicts of Income Allocation in Tax Treaty Law : The Differing Opinions of the Austrian Federal Ministry of Finance and the OECD (2013) 67 Bulletin for International Taxation, p. 105.
89 Michael Lang, Die Auslegung Des Multilateralen Instruments (2017) 1 SWIWI, p. 14. 90 Becker, Reimer and Rust (n 49), p. 209.
91 van Brunschot (n 54), p. 8.
92 I.e. avoidance of double (non) taxation.
93 Michael Lang and others, The OECD Multilateral Instrument for Tax Treaties: Analysis and Effects (Kluwer Law
International BV 2016), §3.1; Lang, ‘Die Auslegung Des Multilateralen Instruments’ (n 89), note no. 5 p. 14.
94 E.g. “context” and “unless otherwise requires”.
95 Art. 3(2) OECD MC neither clarifies what “context” means and which are the means of interpretation included
in its concept that can be used by the interpreter (i.e. which means of interpretation the tax courts and administrations shall consult to interpret the “undefined terms” and/or that can be considered binding tools for interpretation purposes), nor when it requires an autonomous treaty interpretation.
the treaty “context” for interpretation purposes9697. If one looks at the wording of Art. 3(2) OECD MC,
the concept of “context” does not have any limitation in this respect, it is not circumscribed; the words used are vague and general. In this situation, in the lack of any interpretation guidance, the interpreter could determine the meaning of the provision at hand as preferred. Thus, the definition of the “context” is exposed to many possible subjective interpretations and, therefore, to many different variables and approaches9899.
Concerning the above, it seems that on the basis of Art. 3(2) OECD MC, theoretically any of the mentioned approaches could be upheld and applied by the interpreters. If this is true, few further questions should be assessed: does it mean that, for example, potentially all the possible means of interpretation which have been included in the concept of “context”100 can be binding for the interpreters
and the taxpayers? Could all the possible interpretations (derived from any of the mentioned possible approaches) be binding for the taxpayers? In this situation, in other words, it should be ascertained whether the tax courts or tax administrations (i.e. the interpreters) could decide a DTT case in favour or not of a taxpayer/country or territory based on all the possible (means of) interpretation or if they have any restrictions in this respect.
It has been submitted somehow that Art. 3(2) OECD MC would be limited by the general principles of interpretation, established in any given country or territory101; therefore e.g. only the means of
interpretation allowed by these principles should be used by the interpreters. However, could these “domestic” general principles be applied by the tax court and administrations for the interpretation of a DTT even if general interpretation rules for the interpretation of international laws are provided in the VCLT? In this respect, the answer should be negative since the general “domestic” principles could vary in the different jurisdictions102; this would lead to different interpretations of Art. 3(2) OECD MC and
would be inconsistent with the scope of the VCLT.
The VCLT codifies the most important international rules and principles of interpretation; as anticipated in the previous § 2, it has a binding character for all countries103. According to the Commentary to the
preliminary draft of the VCLT104, the scope of the VCLT, inter alia, was (as it is) to establish some
common interpretation rules and principles (a sort of “minimum standards”) to be followed in order to interpret international law in a more consistent and uniform way. Its purpose was (as it is) in fact to encourage international law common interpretations and to avoid different countries practices105.
96 See previous § 3.2. of this Paper.
97 Needless to say, this is true unless the parties have listed in the relevant DTT or Protocol the applicable means
of interpretation.
98 E.g. the interpretation would be subject to the interpreters’ national culture, legal system, professional education,
personal and political opinions, etc.
99 We already saw in § 3.1 and 3.2. some of the different approaches that can be upheld with respect to these
issues (e.g. broad or narrow concept of “context”, external means of interpretation allowed or not, pre-eminence accorded to the domestic law or to the autonomous treaty “context” for interpretation purposes).
100 I.e. in accordance with the relevant approach adopted. 101 Arnold (n 15), p. 14; Van der Bruggen (n 21), p. 151-152.
102 E.g. in certain countries the textual interpretation is allowed only whilst in certain others only the teleological one
is followed. See: Van der Bruggen (n 21), p. 151. The Author pointed out that “domestic courts should, as a point of international law, resist the temptation to interpret treaties guided by nationalistic concerns or corresponding exclusively to legal concepts of its legal system”; Vogel and Prokisch (n 48), § 3, p. 60.
103 Included the countries that did not sign the VCLT, since it is considered customary law.
104 E.g. the VCLT wanted to avoid that one state considers only the intention of the parties, and that another
consider only the wording of the terms under investigation, for interpretation purposes). See: Official Commentary on a Preliminary Draft of The Vienna Convention (n 9), p. 2268.
105 Reimer (n 25), p. 458. According to the Author interpretation of international law has the scope to reconcile the
The above is a critical issue also for the international tax treaties. National tax law is generally applied and interpreted in accordance with the relevant State’s own interpretation rules and principles; in a purely domestic situation, this process does not trigger any negative consequences for the purpose of the best interpretation of the laws/agreements under review. However, in the international context this approach seems not to work properly. It would inevitably lead to a different understanding of the parties’ obligations and, therefore, to a different application of the relevant DTTs in the relevant parties’ jurisdictions106. According to certain Authors, in the context of the DTTs such approach would not be
consistent with the purpose of the DTT itself and could lead to double (non) taxation107.
Thus, in order to apply the DTTs in a uniform and consistent way in the relevant jurisdictions they should be interpreted under the VCLT. The VCLT should apply to the DTT even if the latter provides with special interpretation rules (e.g. Art. 3(2) OECD MC)108109. This could happen, for example, when the special
interpretation rules provided in the DTT are not self-explanatory or sufficiently clear110 (which seems
exactly the case with respect to Art. 3(2) OECD MC). In this respect, the VCLT establishes which are the usable means of interpretation, i.e. those which can be considered sources of legal obligation for the parties involved and therefore able to bound the interpretation of the interpreters.
The above explains why certain commentators believe that the sentence “unless the context otherwise requires” should be read as “unless the VCLT otherwise requires”111: the interpretation of Art. 3(2) OECD
MC and of the terms used but not defined in the OECD MC should be carried out in accordance with the relevant domestic law only if the principles of (the means of interpretation allowed by) the VCLT do not require a different (more appropriate) interpretation.
With this regard, the inclusion of Art. 3(2) in the DTTs seems of scarce relevance112. Imagining a DTT
without this provision, the interpretation process - most likely - would not change and would be consistent with the customary law principles laid down in the VCLT. The interpreter would be “forced”113 to assess
the treaty’s “undefined term” under both a textual and a teleological approach. This is probably the “binding” outcome that results from the VCLT; if the VCLT principles are binding for all the Contracting States, it can be said that the relevant tax courts and authorities should be “obliged” to consider all the elements included in the VCLT, to justify the outcomes reached, to explain why they have given preference to one approach rather than to the other.
On the contrary, if we assume the inclusion of such provision in a DTT and the non-existence of the VCLT, we would not probably reach the same conclusions. Most likely, in fact the interpreter would not be conditioned by the same interpretation rules. As discussed in the previous paragraphs, in the absence of the VCLT the interpreter is freer to adopt its preferred (interpretation) approach. The interpreter would be only limited by the “domestic” principles of interpretation provided by its own country. However, if these principles do not demand the interpreter to take into consideration all the
106 Van der Bruggen (n 21), p. 151.
107 Van der Bruggen (n 21), p. 151 and p. 154. 108 Van der Bruggen (n 21), p. 154.
109 Unless the Contracting States specify in the DTT which are the usable means of interpretation. 110 Engelen (n 14), § 10.10.2.; González Becerra (n 16), § 5.
111 Van der Bruggen (n 21), p. 155; Engelen (n 14), § 10.10.2. 112 Van der Bruggen (n 21), p. 155.
113 Arnold (n 15), p. 6. According to the Author, Art. 31 VCLT does not prevent the interpreter of a treaty to assess
only the wording of a treaty avoiding the analysis of the term under investigation according to the object and purpose of the treaty or the intention of the parties; this to the extent that the textual interpretation is able to lead the interpreter to a satisfying interpretation. Contra: Reimer (n 25), p. 459: by reason of principle of good faith the self-clarity of a term does not prevent its test by the interpreter according to the other interpretation approaches. This view seems shared by Arnold (n 15), p. 14. According to this Author, the VCLT has limited role in the interpretation process. Treaty terms can be interpreted with or without Arts. 31 and 32 of VCLT. The interpretation is an ordinary human subjective process and cannot be ruled by a priori criteria and conditions.
