Maastricht University
International and European Tax Law
2021/2022
Grade: 9
1
,Week 1 – Interpretation of treaties
Lecture:
Welcome / Interpretation of Tax Treaties
Article 3, par. 2 OESO Model prescribes ‘As regards the application of the Convention at any time by a
Contracting State, any term not defined therein shall, unless the context otherwise requires, have the
meaning that is has 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.’
The two main questions are:
- Who can tax? (national legislation)
- Who may tax? (Tax treaties)
Different ways to interpret a legislation:
1) Grammatical
2) Teleological (aim and purpose)
3) Contextual/ systematic
4) Historical.
The wording of the treaty is the boundary in which the interpretation must take place. As the
wording is what is agreed upon by the participatory state.
Three ways to define concepts:
1) Country A follows source state (this is the UN Approach)
2) lex fori (both countries would apply their domestic law, and use their own interpretation)
3) Autonomous interpretation (interpreting the treaty) (OECD model)
Tutorials:
1. Please indicate the importance of Article 31 VCLT for the interpretation of tax treaties. What are
the main differences in respect of interpretation of national tax law and international tax law? Do
you think that a substance over form approach as used for instance in par. 33 of the OECD
Commentary on Article 5 of the OECD Model will be covered by the VCLT?
What are the main differences in respect of interpretation of national tax law and international tax
law? International: ordinary meaning, grammatical interpretation. Object and purpose (historical
perspective). There is a system here for interpreting. It’s similar to domestic law. But, you see in the
tax treaties that it is much more focussed on the wording, it is favouring the wording. Strong
emphasis on the wording.
Substance over form states that you should look at the goal of the legislation, but this is not possible
and especially if it cannot be united within the text of the treaty.
But sometimes there can be a more practical approach used which can deviate from this strict
approach.
31-2: what is part of the context.
2
,31-3: what is to be taken into account, together with the context.
So, article 31 tells us the methods of interpretation. These are the same to domestic law. But, in the
treaties, there is a big emphasis on the wording. You cannot check the purpose at the date of
signing, very easily. So we look at the wording. Focus on the wording.
In par. 33 of the Commentary on Article 5, there is said: substance over form approach. ‘What is
done, is not reflected in reality’. Can we use this? No, because it is not in line with article 31-1 VC.
The substance over form approach goes over/behind the wording.
OECD Model Convention and the Vienna Convention is not law. These are just models, to help the
states. They are not competing, there is no different hierarchy. The OECD is more in detail. First
reference is the model you based the treaty on. Have in mind: the Vienna Convention is from 1969,
that doesn’t mean that all states signed and ratified the convention. You have to sign and ratify, to be
bound.
2. A Belgian resident is the owner of a company located in the NL. He is also the only director of
the company but does not receive any salary but only dividends. In such case, Dutch tax law
determines that a part of the dividends paid is fictitiously treated as salary. How will the “salary
payment” be treated under the tax treaty between Belgium and the NL? Please consider that
Belgium will only see a dividend but not a salary payment.
The issue here is that one state uses a fiction that creates a ‘salary payment’. Article 3-2 OECD-Model
says: if you can’t interpret the treaty, you have to fall back on own domestic law. That’s what
happened here. The Dutch see a payment of dividend and a payment of salary. The Belgian says: I
only see a dividend.
Article 3, par. 2 OECD Model. Article 3 refers to the state of the source state. Imagine: you close a
treaty, and the wording is leading. Then one state creates a fiction. That is not in line with the treaty.
The fiction is going against in the Vienna Convention. That is not in good faith. If states were allowed
to make a lot of fictions, that would not work with signing treaties.
The Hoge Raad decided that since there is no definition, but a fiction, Belgium cannot be expected to
be known of the fiction. So the fiction is not allowed here. Domestic fictions cannot be brought in
into treaties. [I think the case concerned was mentioned in the lecture]
So you look at the tax treaty between state. The tax treaty is based on the OECD Model Convention.
States don’t sign models, they sign treaties.
3. A-Corp, a company resident of State A with the place of effective management in State A,
operates a hovercraft-line between a city in State A and a city in State B. A-Corp owns a hotel, a
restaurant and a duty-free shop in State B, located close to the landing stage of the hovercrafts.
Those facilities are mainly used by passengers who are using the hovercrafts for crossing the sea.
The tax authorities of State B claim taxing rights in respect of profits derived from the
hotel/restaurant/shop activities. They argue that Article 8 OECD MA cannot be applied for those
activities as they are a separate business of A-Corp so that Article 7 should be applied. The term
3
,“operation of ships” does not include such separate activities what had been confirmed by a tax court
of State B.
A-Corp refers to the law of State A. Under this law such activities would be considered as being closely
related to the operation of ships. Accordingly, State A claims the taxation right in regard of the
hotel/restaurant/shop activities maintaining that the place of effective management was located in
State A. The result would be therefore double taxation.
Show how the term “operation of ships” can be interpreted. In doing this, explain the three
different approaches which are used in this respect and indicate which approach you find most
suitable. Which method is used by the OECD?
We have state A, resident company, having the hovercraft line (A-corp company) from my state to
state B. Apart from that, the company A-corp has a hotel, restaurant and shop activities in state B.
Some activities are not an operation of ships literally, but can be directly related to the operation of
ships. These activities then fall under the scope of article 8 par. 1 OECD instead of art. 7 par. 1 OECD,
meaning that they are taxed in the Contracting State of residence, instead of the Contracting State of
permanent establishment.
Article 3-2 comes into play. How to classify the ‘operation of ships’. Article 3-2 says: if there is no
definition here, we look at the domestic law of the country that is applying the treaty. Who is
applying the treaty here? State A is looking to domestic law, and to the definition of the treaty. If
they are applying an exemption, they are applying the treaty, one could say. 3-2 says: the state that
applies the treaty.
[One approach used is whether the activities are almost exclusively used by the customers of the
operation of ships. If this is the case, the activities fall under the scope of art. 8 par 1 OECD. If not,
they fall under the scope of art 7 par 1 OECD.
A second approach is the check whether they are ancillary to the operation of ships. If this is the
case, then art 8 par 1 OECD applies. If not, art 7 par 1 OECD applies.
The third approach is whether the activities are only carried out because the operation of ships is
carried out. If this is the case, art 8 par 1 OECD applies. If not art 7 par 1 OECD applies.]
1) Country A follows source state (this is the UN Approach; also called Lex Causae), par. 33.2 of
Commentary on article 23. If we don’t allow the source state, then we go to the residence state. But
source state already followed their approach and taxed.
2) lex fori (both countries would apply their domestic law, and use their own interpretation). State B
would say: part of it is article 7, state A would say: part of it is article 8, so double taxation. No
providing of credits. Tax payer is worse off. Tax authorities then have to get together and look for a
common ground.
3) Autonomous interpretation (interpreting the treaty) (OECD model). Look at international law and
cases. This interpretation tries to avoid the domestic law as much as possible.
If, in this case, both states would use domestic law (lex fori), then double taxation will occur. Author
Lang favours approach 3): the autonomous interpretation. This idea comes with the issue of 3.2: go
4
,to the domestic rules, unless the context asks otherwise. In case of autonomous interpretation: look
in international documents relating to the treaty. If there is a international practice that explains
‘operation of ships’, then use these rules. Lang says: ‘if you cannot find the context requires
otherwise’, you are not a lawyer. Author Jones is more in favour of following the source state.
Tutor would always favour the autonomous interpretation.
OECD seems to be pushing the autonomous view. Thus, let’s look for international concepts. You can
use these approaches:
- One approach used is whether the activities are almost exclusively used by the customers of
the operation of ships. If this is the case, the activities fall under the scope of art 8 par 1
OECD. If not, they fall under the scope of art 7 par 1 OECD.
- A second approach is the check whether they are ancillary to the operation of ships. If this is
the case, then art 8 par 1 OECD applies. If not, art 7 par 1 OECD applies.
- The third approach is whether the activities are only carried out because the operation of
ships is carried out. If this is the case, art 8 par 1 OECD applies. If not art 7 par 1 OECD
applies.
It depends on which one you follow. Treaties and commentaries use a lot of vague language. That
does not help.
4. The status of the OECD Model Commentary is heavily discussed in literature. What is your
opinion on the following statements?
Three ways to define concepts:
1) Country A follows source state (this is the UN Approach)
2) lex fori (both countries would apply their domestic law, and use their own interpretation)
3) Autonomous interpretation (interpreting the treaty) (OECD model)
a. The commentary is part of the context as defined in Article 31(1) VCLT
The commentary on the model cannot be the context, unless it has specifically been agreed upon by
the states that signed the treaty, therefore it does not fall under art 31 VCLT.
However, there are some countries that, when they sign the tax treaty, they accept the commentary.
In that case, the outcome is different. This doesn’t happen often.
b. Does the commentary indicate the ordinary meaning of a term?
The ordinary meaning is more than the just up-to-date version of the word but can also encompasses
the more common understanding of the term. The commentary can thus be included to determine
the ordinary meaning of the terms used in the tax treaty especially for OECD members.
For OECD members you can, thus, argue that the commentary (which was there when the treaty was
signed) indicate the ordinary meaning of a term. See this in the light of the OECD members drafting
the Commentary.
For non-OECD members this works out different.
5
,c. Does the commentary indicate a special meaning of a term?
Question b and c are linked. If you say it is the ordinary link, this cannot be the special meaning. In
both cases, however, the outcome is the same: treaties partners that rely on the model, also rely on
the commentaries.
Some authors argue that the commentary, or the terms in treaties, have a special meaning (instead
of the ordinary meaning). Outcome is the same.
d. Is the commentary an agreement that relates to the treaty (Article 31(2)(a) VCLT)?
No. Not directly related to the treaty.
e. Is the commentary an instrument that was made in connection with the conclusion of the treaty
and accepted by the parties involved (Article 31(2)(b) VCLT)?
No. Not directly related to the treaty.
f. Is the commentary a subsequent agreement in terms of Article 31(3)(a) VCLT?
If you look at 31-3-a VCLT, think about ‘subsequent’. This follows after the treaty. The commentaries
are, however, already there. So, not a subsequent agreement.
g. Is the commentary a subsequent practice in terms of Article 31(3)(b) VCLT?
If you look at 31-3-a VCLT, think about ‘subsequent’. This follows after the treaty. The commentaries
are, however, already there.
h. Is the commentary a relevant rule of international law in the application of the treaty (Article
31(3)(c) VCLT?
The commentary is not a rule, but a recommendation. Therefore it cannot be a relevant rule. It is a
recommendation on what should be done, but states are not bounded.
i. Is the commentary part of the preparatory work of the treaty?
The Commentary cannot be seen as preparatory work, as it is not the actual preparatory work for the
treaty, but rather it is not part of the negotiation process.
There is no link between the treaty and the commentary. Important here, is that we focus in the
Commentary, people reference the Commentary. It is a guidance. But states are not bound.
5. A Corp, incorporated under the laws of state A, has its place of effective management in State B.
A Corp holds all the shares in C Corp, incorporated and established in state C. C Corp distributes a
dividend to A Corp. A domestic court in state C must decide on the question whether the tie-
breaker rule in the tax treaty between states A and B can be used to determine whether A Corp is a
treaty resident of the country (see Article 4(1), 2nd sentence OECD Model and par. 8.2, 2nd
sentence OECD Commentary on Article 4). Suppose state C would have a tax treaty with state A,
but not with state B. The treaty between states A and C was concluded in 1975. If you were the
judge, how would you decide? Please assume that the treaties in question follow the Model of
2014.1
1
For next week the focus is: look into the Commentaries, look into the wording and try to understand: what
does it mean to have a permanent establishment.
6
,When we look in treaties, the first thing we look at is the scope. Article 4(1) tells us who is the
resident. Residence comes from domestic law. The moment we know that residence comes from
domestic law, we might have multiple resident states. If you look at the structure of treaties, it is very
important to know: where is someone resident. When more than one state sees themselves as a
resident state, we need rules.
4(2): Tie-breaker for individuals.
4(3): Tie-breaker rules for things other than individuals.
Difference between model 2014 and model 2017 about tiebreaker rule: 2014: place of effective
management. Now we say: 2017: mutual agreement that takes into account: effective management.
2014 model was more straightforward. Why did this change? States would not agree upon the place
of effective management.
In our case, the place of effective management is in state B. What about the second sentence of
article 4 (1)? ‘This term, however, does not include any person who is liable to tax in that State in
respect only of income from sources in that State or capital situated therein.’ They are telling us, who
is a resident. Go to domestic law. Second sentence says. Par. 8.2 Commentary says: when there is a
conflict and you use the tie breaker rule… A corp is a resident in state A, A corp is also a resident in
state B. We said, effective management in state B. If we follow 8.2, it says A corp can never be a
resident in state A anymore.
To conclude: if we follow the Commentary, A corp can never be a resident of state A
anymore. But, for this subject? Or forever? 8.2 is one of the most discussed Commentary paragraphs.
Treaties are bilateral. 8.2 says: if you lose tax right between A-B, you cannot be seen as a resident of
state A anymore. Between A-C you also lose your tax rights. This is discussed, since article 4-1 and
thus, eventually domestic law, still says state A is a resident.
The second sentence of 4-1 was at first included to exclude diplomats.
A remark in the commentary does not provide a substantial legal basis. Therefore the winner and
loser scenario cannot have impact on the treaty between the loser and the third state. However, this
is the academic perspective. From the OECD perspective the treaty between the third country and
the loser has been void, as the loser has lost its taxing right. (Discussion 8.1 and 8.2 commentary on
the convention on the model tax treaty).
The tax payer was worse off now. There is no treaty between B-C. Most people say: the tax payer
being worse off, is wrong.
States and courts are following the Commentary.
We have another problem here. We make use of 8.2 Commentary. 8.2 is replaced on 17 July 2008.
The treaty in our case was concluded in 1975. Do we apply the Commentary on this case, on this
treaty? OECD favours the ambulatory approach, instead of the static approach. Follows from the
introduction in the Commentary. So apply the Commentary of 2008 to the treaty of 1975.
India: in 2012 they lost a big case, millions of dollars. They came up with the amendment to the tax
treaty that would say: our approach is the right one, we have tax rights. The taxable person is
protected by the Court rule, but all the others aren’t anymore, because of the amendment of the
legislation by the legislator.
7
,OECD always ways: the newest things I wrote, applies to older cases. There are court decisions that
say: that goes too far.
Always keep in mind: the OECD is formed by Ministries of Finance and Tax authorities. They want to
tax. So that’s why many Commentary paragraphs are quite sketchy. The commentaries are
guidelines, they are not signed. A lot of issues with the commentary. Don’t always take the
commentaries as the correct position. Understand what is going on, what is the idea underneath.
To conclude this question: know the static and ambulatory approach, know the win-or-lose scenario.
8
,IETL
Week 2 –Taxation of Business Income
Lecture:
International Taxation of Business Income
General:
- For a PE you need to have:
(a) Fixed
- Temporal (time wise)
- Geographical (in one place)
(b) Place of business, through which
(c) business of the enterprise is wholly or partially carried on
1. It is rather clear what at place of business is and under which circumstances it can be seen as
fixed. There are only some questions left:
a) Under which circumstances can machinery or equipment (e.g. a server) be a place of business?
The commentary assumes that the equipment is moveable. It has to be at the place, that makes it a
place of business. The equipment itself is not a place of business. Test whether the PE criteria are
met. For place of business: aim to obtain profit, par. 28 Commentary: six months. Par. 125:
equipment. Par. 126. These paragraphs are relevant here. Par. 22 as well. All on article 5 OECD
Model.
The machine itself is not a fixed place of business, but it is the fact that it is the located in a building
that could make it a fixed place as the building itself is fixed and the building which contains the
equipment fulfils the criteria of article 5. The place where the equipment or machine is, will then be
able to tax the business.
WHEN YOU COME WITH AN ARGUMENT AND YOU HAVE SUPPORT IN THE COMMENTARY, ALWAYS
CITE THE EXACT PARAGRAPH.
Par. 22: ‘[…] As recognised in paragraphs 51 and 57 below a single place of business will generally be
considered to exist where, in light of the nature of the business, a particular location within which the
activities are moved may be identified as constituting a coherent whole commercially and
geographically with respect to that business’
Par. 28: ‘[…] Whilst the practices followed by member countries have not been consistent in so far as
time requirements are concerned, experience has shown that permanent establishments normally
have not been considered to exist in situations where a business had been carried on in a country
through a place of business that was maintained for less than six months (conversely, practice shows
that there were many cases where a permanent establishment has been considered to exist where
the place of business was maintained for a period longer than six months) […]’.
Par. 125 regarding computer equipment: ‘Computer equipment at a given location may only
constitute a permanent establishment if it meets the requirement of being fixed. In the case of a
server, what is relevant is not the possibility of the server being moved, but whether it is in fact
moved. In order to constitute a fixed place of business, a server will need to be located at a certain
place for a sufficient period of time so as to become fixed within the meaning of paragraph 1.’
9
, Par. 126: ‘Another issue is whether the business of an enterprise may be said to be wholly or partly
carried on at a location where the enterprise has equipment such as a server at its disposal. The
question of whether the business of an enterprise is wholly or partly carried on through such
equipment needs to be examined on a case-by-case basis, having regard to whether it can be said
that, because of such equipment, the enterprise has facilities at its disposal where business functions
of the enterprise are performed.’
b) Why can a construction site as such not constitute a PE (cf. par. 49)?
[There is a conflict on scholar side whether the construction site can constitute a PE. Prokisch thinks
it is not possible, by nature, and other scholars think it is possible.]
A construction site can only be considered to be a PE if it located longer than 12 months. The
construction site is temporary, by nature.
However, this is only an exception as according to par. 1 a permanent establishment means a fixed
place of business through which the business of an enterprise is wholly or partly carried on. Whilst a
construction site is normally not fixed, as they can moved, or be removed from a state once the
construction is done. Therefore, it is not meeting the criteria of it being fixed. However, this is where
paragraph 3 creates a special fiction from which a construction site can be considered a permanent
establishment.
The construction company can use the land. But they don’t have anything else at the land, so the
land is not at the disposal of the construction company, therefore it is not a PE as such. (par. 16 MC).
The construction company can have a PE per art 5(1) if it owns the land or has a port de cabin on the
construction site.
Some people say: a construction site is by nature not a fixed place of business. Temporary. So it can
never meet the criteria 5-1. These people say: article 5-3 is a special rule.
On the other hand, people say that a construction site can be a PE if it meets the criteria of 5-1 and
the 12-month rule.
If you see 5-3 as a special condition, then this is the only condition. You don’t have to meet 5-1
criteria.
If you don’t think 5-3 is a special condition, then you have to meet both 5-1 and 5-3 criteria. For
Arnold B.J., this is the case. Recommend you read this. He is a Canadian author.
The site cannot as such be a 5-1 PE, since the construction is just a service, it is not a fixed place of
business. If there are buildings of the workers, where they make decisions, then it can be a fixed
place of business.
The core is the following. The discussion is: do you only need the 12 months, or also the 5-1 criteria?
Teachers at our faculty say: you only need the 12 months of 5-3, since the site construction by nature
cannot be a fixed place of business as in 5-1 – you just render service. You might have a fixed place of
business when you make decisions regarding the construction activity, in that place, because the
activity as such can never be a fixed place of business.
c) Will an Italian circus, that travels through the Netherlands for longer than 6 months and moves
every second or third day to another location, have a PE in the NL (see par. 22 “a particular location
within which the activities are moved”)? What is the difference between this example and the
10
