EU Tax Arbitration Convention
Dr Jean-Philippe Chetcuti. All Rights Reserved.
In this paper, the Convention is
explained from various points of view. First of all, the troublesome process of
its adoption is traced from a historical perspective, starting from the earliest
debates on the need for such a measure. Then
follow detailed explanations of the scope and the subject matter of the
Convention and its workings.
The process of the adoption of the Convention
up its Communication to the Council of November 1973 on multinational
undertakings, the Commission stressed in its 1975
action programme that it would continue its work with a view to presenting a
proposal on the elimination of double taxation which can result from adjustments
made to profits by a MS.
A fair complement to the Mutual Assistance Directive
This proposal for a directive was made on 29 November 1976
when the Commission presented a draft Council Directive on an
arbitration scheme to settle double taxation disputes
to complement its proposal for a directive on mutual assistance in taxation
In its explanatory memorandum,
the Commission stressed the close link which this
proposal maintained with the Mutual Assistance Directive.
The Commission acknowledged that the introduction of a system for the
exchange of information between the tax authorities could potentially increase
the number of instances of double taxation, particularly in the field of
transfer pricing. Thus, the
Commission felt that the Arbitration Directive should be adopted simultaneously
with the Mutual Assistance Directive to ensure for community taxpayers a remedy
for instances of double taxation ensuing from overlapping assessments of the
same income in more than one jurisdiction.
In a session held in Brussels on 26 and 27 October 1977,
the opinion submitted by the Economic and Social Committee of the European
Community was generally favourable and the (nine) MSs were in agreement on the
principle underlying this draft directive.
On 14 June 1977, the European Parliament submitted its position
and the EC Council commenced its deliberations in 1978.
Causes of delay
The binding nature of awards and the jurisdiction of
The adoption of the proposed directive was held-up, inter
alia, because the Financial Committee of the Council disagreed on whether
the findings of the Arbitration Commission should be binding and on the
competence of the ECJ. The
application or interpretation of a convention cannot be submitted to the
jurisdiction of the ECJ and therefore, the legal protection of companies would
be limited. At the most they would
be able to summon their own tax authorities before the national court if they
think that the Convention has not been correctly applied.
Neither can the European Commission take MSs to the ECJ where MSs are
considered to be acting contrary to the provisions of the Convention, as in the
case of a directive.
Legal form & legal basis
Moreover, the Council was of the opinion that the issue
should be tackled by a multilateral convention based on Article 293 of the
EC Treaty rather than by a directive. Under
Article 293, MSs’ desire to eliminate double taxation must be met through
negotiations between the independent MSs. These
negotiations should eventually culminate in a multilateral convention.
At the initiative of the Netherlands, the proposed
directive of 1976 was transformed into a convention.
The draft convention prepared by the Dutch government was the final
product then submitted by the Commission for adoption by all MSs.
Unwillingness of MSs to surrender fiscal sovereignty
From that point onwards, it took fourteen years to have
the measure adopted. A
possible difficulty which could have served to hold back the introduction of a
mechanism for the elimination of double taxation resulting from transfer pricing
adjustments was the typical unwillingness of the MSs to surrender such a
fiscally sensitive area to the authority of a supranational entity.
Evidence pointing to the probable reason for the dragging of feet on the
issue is the careful terminology employed in the wording to the official title
of the Convention in that it fails to include the term “arbitration” or
“arbitration procedure” so as not to emphasise the supranational elements of
an arbitral commission.
Going back to the 1976 Arbitration Directive, the Explanatory Memorandum
included in the Commission’s proposal assured MSs that the suggested arbitral
commissions “by no means constitute supranational judicial bodies.”
The package approach
The fact that the Mutual Assistance Directive was adopted
on 19 December 1977 and amended on 6 December 1979, while the
arbitration directive was delayed until 1990, was condemned by advisers claiming
that, at least in the field of direct taxation, the European Community didn’t
protect the interests of taxpayers as much as the interests of the tax
It has been argued that this demonstrates that measures originally
intended to pass through Council as a package should be dealt with, and enacted
by, the Council simultaneously.
On 17 January 1984, Mr Tugendhat, Commissioner for
taxation, addressed a Communication to the Council treating fiscal measures
aimed at the encouragement of co-operation between undertakings of different MSs.
This Communication dealt with four proposals: EEIG Regulation, Merger
Directive, Parent-subsidiary Directive and the Convention.
At the time, the Commissioner stressed that it was ‘only the outcome
which mattered.’ Thus the
Commission was ready to accommodate any requests to amend the said proposals as
long as they resulted in the elimination of double taxation.
The Commission addressed
another Communication to the Council on fiscal
measures aimed at encouraging co-operation between undertakings of different MSs,
and aspiring for the quick approval of the Merger and Parent-subsidiary
Directives and the Arbitration Convention.
However, agreement on these measures was only reached after another five
years. Arguably, tying the Convention to the Merger and Parent-subsidiary
Directives could have held up the process of its adoption; however, as the
result demonstrates, the Commission has thus succeeded in seeing the adoption of
all three measures together.
On 23 July 1990, the Ministers of Finance of the twelve EC
MSs agreed on a multilateral Convention on the Elimination of Double Taxation in
Connection with the Adjustment of Profits of Associated Enterprises
(hereinafter, Convention). France
was the first State to deposit the instrument of ratification on 21 February
1992 while Portugal was the last State to deposit the instrument of ratification
on 28 October 1994. Article 18 of
the Convention states that entry into force would be on the first day of the
third month following the depositing of the instrument of ratification by the
last signatory State. Accordingly,
the Convention came into effect on 1 January 1995, between Belgium, Denmark,
Germany, Greece, Spain, France, Ireland, Italy, Luxembourg, the Netherlands,
Portugal and the UK.
Accession, life span and possible termination
In 1995, the three new MSs, Finland, Austria and Sweden
concluded a separate Treaty of accession to the Convention
which would enter into force -between
the ratifying States three months after at least one of the three new MSs and at
least one of the old twelve MSs ratifies it.
For MSs which would ratify later, it would enter into force three months
after their ratification. The
Accession Convention has resulted in minor adjustments to the terms of the main
Convention (the taxes affected by the Convention, the definition of competent
authorities and the notion of ‘serious penalties’ in the acceding States).
The initial application period of five years expired as on
1 January, 2000. However, the
Convention could be expressly extended by the Contracting States before 1
In fact, the life span of the Convention was extended with an additional
period of five years by Protocol on 25 May 1999.
The Protocol is subject to ratification by all (now) fifteen MSs and it
will not enter into force until the beginning of the third month following
ratification by the last MS.
On entry into force, the Protocol will give retrospective effect to the
Convention as from 1 January, 2000.
The Convention is not in force at this point in time and,
until the fifteenth MS finally ratifies the Protocol, MNGs of companies
operating within the EU face a vacuum in their protection against double
taxation. There also remains the
possibility that not
all MSs will ratify in which case the Protocol will never come into force.
The Convention will in fact have been terminated as per 1 January 2000.
As Luk Hinnekens suggests,
the time limit may result in timing problems.
The Convention does not provide for procedures which will still be
pending at the end of an unrenewed five-year term.
It seems therefore, that such cases, if any, will have to be continued on
the basis of domestic law or mutual agreement procedures.
It is suggested that, for the time being, taxpayers can partially escape
these obstacles by formally initiating domestic legal remedies and then
withdrawing under the strict condition that the procedures of the Convention
While under the original agreement from 1990, any
extension required the unanimous consent from all MSs, it has now been agreed
that the extension will occur automatically unless the MSs decide otherwise.
This avoids the need for MSs actively agree on further extensions.
The other option was that MSs agree that the Convention be extended
Transfer pricing practices, price-adjustment of cross-border
dealings and resulting double taxation
The scenario which the Convention sets out to address is
one of transfer pricing. This
heading explains the practice of transfer pricing and how international double
taxation arises therefrom.
Intercompany and intracompany dealings
The transfer of intermediate products and services
among different companies forming part of the same group of companies is not
uncommon and, given the group relationship existing between such companies, one
can understand why such transfers do not occur at prices which would apply
between two unrelated companies (“arms’ length prices”).
Technically, the financial position of the group as a whole remains
unaltered by, and is indifferent to, the pricing of intra-group transactions
since with the flow of such goods or services within the group, the loss which
one member company suffers as a result of such transfer-pricing is compensated
by a corresponding profit ensuing in the hands of the other party to the
transaction. In fact, this is so
until the goods or services leave the group, in which event the profit or loss
generated by a member of a group transacting with an unassociated company
becomes the profit or loss of the whole group.
Reasons for transfer pricing
Fiscally, this principle of the neutrality of groups to
intra-group transfer pricing does not apply to MNGs of companies.
Due to differences in fiscal law between MSs, the pricing of intra-group
transactions and resulting profits out of one national tax jurisdiction into
another may have very significant consequences on the overall tax position of
the group. There is a considerable
potential for multinational companies to shift (whether artificially or
otherwise) significant amounts of taxable profits to subsidiaries in
jurisdictions with low or zero corporate tax rates, in order to avoid payment of
taxation on profits or to attract lower tax rates.
In practice, such tax saving structures of affiliates in tax haven
countries are relatively easy to set up. For
tax authorities, profit transfers translate into a decreased inflow of fiscal
revenue into public coffers. For
this reason, multinational companies with subsidiaries in lower tax
jurisdictions are particularly susceptible to transfer price adjustments by
national tax authorities.
Tax incentives used by such tax havens can take the form
of zero or low taxation, the absence of withholding taxes
as well as differences in the availability of loss carry-forwards or group
A simplified illustration of a transfer pricing situation
will elucidate the central matter of this chapter.
E1 and E2 are two distinct legal entities forming part of the same MNG of
companies. The former is resident
in MS MS1 and the latter in MS2.
The two enterprises are engaged in commercial and financial transactions
between each other and, considering the largely artificial separation between
the two enterprises, the conditions of such dealings can easily be manipulated
to obtain fiscal benefits for the MNG. A
common tax planning opportunity has become evident in that the corporate tax
rate obtaining in MS1 is higher than that in MS2.
Alternatively, E1 has carryover tax losses which risk not being
compensated. Thus the invoice value
of a particular transaction or transactions (e.g. the sale of a stock of raw
material) is set lower than that which would be charged by independent
businesses in similar circumstances. The lower prices paid by E2 should reduce
the profits generated by E1 for the purposes of MS1 corporate taxation and
therefore a smaller portion of the MNG’s profits should be taxed at the higher
tax rate applicable in MS1. The
correspondingly higher profits generated by E2 should be taxed at the lower tax
rate chargeable in MS2, resulting in tax saving for the MNG of companies.
The competent authorities of MS1 are likely to be aware of
this transfer pricing opportunity being availed of in this example.
The tax administration of MS1 may therefore increase its assessment of
the company’s taxable profits according to the arm’s length principle
(‘primary adjustment’) and therefore neutralise the tax avoidance scheme for
the purposes of E1. However,
the tax authorities of MS2 might be in disagreement with MS1 over the adjusted
and may be unwilling to reduce the associated company’s declared profits
accordingly (‘corresponding downward adjustment’) or they might only be
willing to do so following transfer price fixing rules which differ from those
applied in MS1. This unilateral
adjustment of profits of associated enterprises by national tax authorities
causes overlapping assessments of the same income within the Community, which
instance of double taxation is precisely what the arbitral procedure contained
in the Convention seeks to eliminate.
I. Case study illustrated.
= Associated Enterprise / Undertaking
= Member State
= Selling price
= Transfer Price
Authority for transfer pricing adjustments
MSs’ tax authorities counter transfers of profits or losses by
adjusting taxable profits as if transfer transactions were carried out at
arm’s length, thereby re-allocating to the accounts of a taxpaying company the
profit which it would unduly have transferred to its associate company using the
transfer pricing mechanism.
Typically, the competent authorities derive the power to conduct
this exercise from the relative provisions under domestic law.
Furthermore, such authority is also found in the bilateral tax treaty, if
any, existing between the two countries in question, generally styled in line
with the OECD Model:
“Where (…) conditions are made or imposed between the
two enterprises in their commercial or financial relations which differ from
those which would be made between independent enterprises, then any profits
which would, but for those conditions, have accrued to one of the enterprises,
but by reason of those conditions, have not so accrued, may be included in the
profits of that enterprise and taxed accordingly.”
When the relationship between the two enterprises involved
in a transfer pricing case is that between an enterprise and its PE, the
competent authority derives its mandate from the following provision:
where an enterprise of a Contracting State carries on business in the other
Contracting State through a permanent establishment the situated therein, there
shall in each Contracting State be attributed to that permanent establishment
the profits which it might be expected to make if it were a distinct and
separate enterprise engaged in the same or similar activities under the same or
similar conditions and dealing wholly independently with the enterprise of which
it is a permanent establishment.”
Of course, the effectiveness of articles 7(2) and
9(1) arises on a specific provision of the domestic tax law of the MS as they
are not self executing. Besides, as
with most of the countries negotiating entry in the EC, the Contracting States
may not have entered any bilateral tax treaty at all.
In other jurisdictions, the domestic legal basis for the profit
re-instatement is simply derived from the national tax definition of profit.
Scope of the Convention
Personal scope of application
“Enterprise of a Contracting State” and
The application of the
Parent-Subsidiary and the Merger Directives is limited to enterprises assuming
the legal forms listed in the Annexes to the Directives.
On the other hand, the personal scope of application of the Convention
includes not only the corporate form but also individual persons and
In the English version
of the Convention, the terms ‘enterprise’ and ‘undertaking’ have been
used interchangeably. It has been
that, since the French version uses the term ‘enterprise’ throughout, it can
reasonably be assumed that the two terms should be attributed the same meaning.
Nonetheless, the terms ‘enterprise of a Contracting State’ and
‘enterprise’ are not defined in the Convention and should assume the meaning
given to them by the DTC applicable between the MSs.
Where such treaty follows the OECD Model, the definitions are,
enterprise carried on by a resident of a Contracting State”
and “the carrying on of any business”.
The term ‘business’ is in turn taken to include “the
performance of professional services and of other activities of an independent
The connecting factor
The Convention is
silent regarding the
connecting factor implied between an enterprise and
the Contracting State concerned. Referring
to the bilateral treaty in terms of Article 3(2) of the Convention, the
applicable criterion is residence, which is only defined in terms of the
domestic laws of the Contracting Parties. A
“resident of a Contracting State” means
“any person who, under the laws of that State, is liable to tax therein by
reason of his domicile, residence, place of management or any other criterion of
a similar nature…”.
In the case of dual residence, also by virtue of Article 3(2) of the
Convention, Article 4(2) and (3) of the OECD Model apply and uphold the
doctrine of effective management and control as opposed to the incorporation
absence of a DTC,
domestic law applies (the lex fori).
This is not an ideal solution since the incongruous definitions
applicable in the Contracting States concerned may result in the inapplicability
of the Convention to their transfer pricing cases which were meant to be
Similar problems may arise as regards the definition of ‘profits’
which is nowhere found in the Convention and which is alien to countries such as
the UK and Malta (a potential MS).
All issues regarding these and other conditions of application of the
Convention fall within the jurisdiction of the national court to which the
taxpayer would refer his transfer pricing case;
they are neither arbitrable under the Convention, nor reviewable by the
It is clear that the
two (or more) enterprises contemplated by the Convention must be
“associated” as suggested by the heading of Chapter II Section II. This relationship is defined in Article 4(1) (a) and (b)
which apply the same criteria used by the ESC to define ‘intra-group’
transfer pricing, by the 1976 proposed Arbitration Directive to define ‘group
of enterprises’ and by the OECD Commentary to define ‘parent and subsidiary
companies and companies under common control’.
An enterprise and its PE are, by a legal fiction,
separate legal entities and this satisfies the prerequisite of association that
the two enterprises have to be distinct enterprises for the Convention to apply
Transfer pricing relationships subject to the
The following table
shows the number of possible transfer pricing relationships which could
potentially be covered by the Convention.
II. Personal-territorial scenarios covered by the Convention
author for figure)
One of the basic
conditions for the application of the Convention is the involvement of two
enterprises of two different Contracting States.
This immediately excludes transfer pricing disputes involving two
associated enterprises based in the same MS.
Article 1(1) brings
within the scope of the Convention only those enterprises of any Contracting
State the profits of which, besides being subject to tax in that MS, also form
part, or are likely to form part, of the tax base of other enterprises of other
Contracting States (scenario 1 in Figure
). Similarly, by specific
provision, “the permanent establishment of an enterprise of a Contracting
State … shall be deemed to be an undertaking of the State in which it is
Therefore, the Convention also applies to (a) an enterprise of a MS
and its PE in another MS and also to (b) an enterprise of a MS and a PE
belonging to an associated enterprise based in the same MS, which PE is situated
in another MS (scenarios 3 and 5 respectively in Figure
basing oneself on Articles 1 and 4, the Convention does not cover the
transfer pricing relationship existing between an enterprise of a Contracting
State and a PE of an enterprise of a Contracting State which is situated in a
third Contracting State. On the
other hand, the Joint Declaration to Article 4(1) seems to contradict this
“The provisions of
Article 4(1) shall cover both cases where a transaction is carried out directly
between two legally distinct enterprises as well as cases where a transaction is
carried out between one of the enterprises and the PE of the other enterprise
situated in a third country.”
The rather poor
drafting and the vagueness of the wording of the Joint Declaration to Article 4(1)
as to what is meant by “a third country” should be so interpreted that it
reconciles with Article 1(2) and therefore that the Convention applies where
there are two legally distinct enterprises in different Contracting States, with
the PE of one of these situated in a third Contracting State (scenarios 6 to 9
in Figure II
It is generally accepted that the Joint Declaration to Article 4(1)
should be taken as an expansion or clarification of Article 4(1) rather
than a restriction of Article 1(2).
1(2) precludes the application of the Convention to a PE situated within a MS
unless the enterprise of which it forms a part is itself an enterprise of a MS.
Therefore transfer pricing relationships contemplated in scenarios 10
to 13 and those between PEs based in different MSs, both of which belong to an
enterprise based in a non-MS (scenario 14) are not covered by the Convention.
Rather than the Convention, the transfer pricing relationship between,
say, a US and a German PE of a Swiss corporation would be
arbitrable under the US-German DTC.
The MS-based PE of a
non-MS-based enterprise would, however, be covered by the OECD Model (Article 9(1))
and secondary adjustments should still be applicable vis-à-vis
the said enterprise as well as the imposition of a withholding tax on a
‘Serious penalty’ exclusion
Article 8, in the case of an enterprise liable to a “serious penalty”,
the competent authority “shall not be obliged to initiate the mutual agreement
procedure” in terms of Article 6, nor to take part in the formation of
the advisory commission in terms of Article 7.
It can even stop these proceedings.
penalties include not only criminal sanctions but also administrative sanctions.
Time will show us whether the concept of serious penalty will be interpreted as
widely as to restrict the Convention in serving its preordained purpose.
It is however clear that the Convention will definitely apply to
associated enterprises which fulfil all their accounting and tax obligations and
which are not involved in fraudulent transactions or transactions falling under
‘abuse of law’ rules.
The Convention applies
to income taxes generally,
in particular, to the number of taxes existing in the Contracting States and
and “to any identical or similar taxes which are imposed after the date of
signature thereof in addition to, or in place of, existing taxes.”
Therefore the Convention covers income taxes on physical persons and on
companies on residents and on non-residents, principal taxes and surtaxes,
present income taxes and identical or similar future income taxes.
It does not cover stamp duties, taxes imposed on a taxpayer’s wealth
and other non-profit based taxes.
The wording of the
Convention differs from that of the OECD Model in that it remains silent as to
income taxes imposed on behalf of political subdivisions or local authorities.
The generality of the term “taxes on income” but then the absence of
a mention of these taxes in the list of presently covered taxes are
contradictory and ambiguous.
Fines and penalties
Penalties are often
calculated as a percentage of the adjusted amounts and make up an integral
portion of the adjustment. It is
submitted, however, that excessive penalties are prohibited.
Article 9 of the OECD Model, running parallel with Article 4,
“only permits adjustments of transfer prices up to that level independent
parties would have remunerated”, in accordance with the arm’s length
Ultimately, penalties remain exclusively an internal tax matter of the
Types of double taxation
The Convention deals with the following types of double
Economic double taxation:
which arises when one MS taxes the profits of a subsidiary and these same
profits are taxed again by a second MS on payment of dividends to the parent
company. The Convention has a
specific mandate to eliminate such double taxation suffered by the MNG of
Juridical double taxation:
which arises when profits, payable to one person, fall within the taxing
jurisdiction of two MSs and consequently are taxed twice.
A typical instance of juridical double taxation is the scenario of a PE.
The profits generated by a branch in one MS form part of the profits of
the enterprise located elsewhere to which the branch belongs and are taxed in
both MSs. Despite the unitary
character of the entities concerned, by specific provision,
the Convention considers these as separate and distinct undertakings.
of double taxation may be the result of transfer
pricing adjustments made by multinational enterprises.
Causes of EC double taxation
In the general considerations of its proposed Convention,
the Commission noted that:
“(…) When one country’s tax authority increases the
profits of an enterprise but the profits of the associated concern that is its
partner in the transaction are not correspondingly reduced in the other country,
the group as a whole suffers double taxation.
Such double taxation may well give rise to distortions,
both in the conditions of competition and in capital movements, of a kind that
would otherwise not exist.
Such consequences are not acceptable within the Community,
because they directly affect the operation of the Common Market.”
The Convention owes its inception to the above realisation
by the Commission. However, how
does double taxation arise? First of all, if a network of DTCs links almost all MSs and
Article 9 of the OECD Model has been provided, how can double taxation due to
transfer pricing arise as in the above case?
Lack of a DTC or of Article 9(2)
The lack of a DTC linking a number of MSs is the first
cause of such double taxation. There results a number of such cases
and the situation in this respect is likely to be aggravated by the prospective
In the case where such bilateral relations have been
undertaken, double taxation arises out of the failure by such treaty parties to
adopt Article 9(2) of the OECD Model. Whereas Article 9(1) authorises the first treaty party
to make the adjustment and neutralise the tax saving scheme, Article 9(2)
binds the second State to make corresponding downward adjustments.
The absence of this article in a DTC between two States means that such
corresponding adjustment is not likely to be made, especially if there are no
unilateral domestic provisions to that effect, and the profits in question will
form part of the tax base of both taxpayers.
Conflicting interpretations of the arm’s length
purpose of the primary adjustment of transfer prices, there exist three main
criteria for determining the arm’s length transfer price applicable between
associated enterprises: the ‘Comparable Uncontrolled Price Method’, the
‘Resale Price Method’ and the ‘Cost Plus Method’.
However, national tax authorities adopt different methods or combinations
thereof, thus giving rise to discrepancies and causing double taxation.
Conflicting classification rules for PEs and for
MSs entertain different views on whether a place of
business, through which the business of an enterprise is being
conducted, should be classified as a PE for tax purposes,
and on whether a company should be considered resident in a given MS or not.
Consequently, conflicting definitions of PEs and company residency
constitute another cause of double taxation.
Under Article 3(2), terms such as ‘permanent
establishment’ and ‘residence’, which are not defined in the Convention,
“shall … have the meaning which it has under the DTC between the States
concerned.” In the absence of a
DTC, the applicable interpretation is assumed to be that adopted by Article 5
of the OECD Model.
The OECD Model has not settled the dispute between the
taxpayer and the State claiming the right to tax branch profits.
It is up to the domestic courts to decide whether a PE has been
established and the different appeal stages tend to be quite time-consuming.
Similarly time-consuming is determining the conformity of transactions
between a permanent establishment of an enterprise and the other enterprise with
arm’s length principles. All this means that it may take years to adjust
profits and eliminate double taxation, which raises yet other problems due to
some domestic laws which restrict the period in which assessments may be made.
The Convention has addressed these timing problems in Article 6(2).
the Convention remains silent as to the problem of “dual-resident companies”
and therefore fails to settle the contrasts between the “siège réel
(real seat) doctrine” and the “incorporation doctrine”
embraced by the UK and Denmark.
The Convention applies a strictly situs-based
(territorial) criterion. It
excludes from its scope of application PEs situated in non-MSs, even if they
pertain to a EC-based enterprises. It
also excludes transfer pricing disputes relating to transactions between two PEs
situated in non-Contracting States, even if they belong to enterprises of two
The territorial scope of the Convention is that laid down
in Article 299 (ex-Art. 277) of the Treaty, which defines the territories of the
MSs and therefore of the EC. The Convention does not apply to the territories and regions
specified in paragraphs 2, 3 and 4 of this Article and to those specified
in Article 16(2) of the Convention (certain French territories,
the Faroe Islands and Greenland).
This means that the Convention has a narrower scope than
the ‘Merger’ and ‘Parent-subsidiary’ Directives which apply to all
companies resident in the territories and regions covered by the entire Article
299 (ex-Art. 227) of the Treaty. Other
conventions concluded on the basis of Article 293 did not exclude such
territories and regions.
III. Overview of procedural steps and their time limits
The unilateral relief procedure
Under Article 5 of the Convention, a Contracting State
intending to adjust the profits of an enterprises is bound to inform the
enterprise of the intended action in due time.
Such enterprise is then given the opportunity to notify the other party
to the transfer pricing relationship in question, which in turn is given the
opportunity to notify the intended adjustment to the Contracting State in which
the latter associated enterprise is based.
While the proposed primary adjustment proceeds unhindered
by this passage of information, the two related enterprises and the second
Contracting State may, on the basis of this information, agree to the
adjustment. Agreement means that
the primary adjustment undertaken by the first Contracting State has been
unilaterally approved by the second Contracting State which, in the absence of a
dispute regarding the application of the principles contained in Article 4,
undertakes to make a corresponding adjustment, relieving the enterprises of any
double taxation which would otherwise accrue.
Thus, the only requirements imposed by the Convention for
the re-allocation to be valid are the first Contracting State’s duty of
notification and that the adjustment must be satisfactory, i.e. acceptable to
the tax authority of the second Contracting State and to the associated
enterprises in question.
Before and during this phase of the operation of the
Convention, according to the interpretation of Article 5, the tax
authorities are free not only to go ahead with the adjustment but also to assess
and collect the additional taxes due.
Of course, the domestic laws of the Contracting State may allow a stay of
the assessment proceedings or at least a stay of the enforcement of an
assessment, but this is entirely a domestic tax affair.
The ESC believed that the 1976 Draft Directive should require a deferment
of the collection of the additional tax so as to give the competent authority
making the adjustment an incentive to conduct negotiations as expeditiously as
The mutual agreement procedure
Article 6 of the Convention provides in the first
place for a mutual agreement procedure which is very similar to the procedures
set up by bilateral tax treaties.
Under this procedure, the competent authorities of the two Contracting
States are to seek a solution of the transfer pricing dispute by mutual
agreement. However, they are under no obligation to actually reach such
Initiation of the international procedure
In the event that no agreement is reached, the
international procedure is set in motion on an administrative level.
In terms of Article 6, “where an enterprise considers that … the
principles set out in Article 4 have not been observed, it may,
irrespective of the remedies provided by the domestic law of the Contracting
State concerned, present its case to the competent authority of the Contracting
State of which it is an enterprises or in which its permanent establishment is
situated. The case must be
presented within three years of the first notification of the action which
results or is likely to result in double taxation within the meaning of Article 1.”
As the above wording suggests, the initiative of
submission lies not in the hands of the competent authorities but in those of
the victims of double taxation, the associated enterprises.
This Article excludes the initiation of the mutual agreement procedure in
cases not contemplated by the Convention and in cases which are undisputed in
terms of Article 5; these cases can only be dealt with in foro domestico.
The form of submission for the international procedure is
not regulated by the Convention but it has been suggested that the form of
filing a domestic tax complain should be applied, e.g. by registered post,
setting out the reasons for the complaint (involving a breach of the principles
of Article 4) and the resulting or potential double taxation.
Also, unless the competent authorities would have agreed on a common rule
of conduct for the issue, domestic rules apply as to whether the submission
should be accompanied by the arguments and backing documentation or whether
these should follow at a later stage.
Other Contracting States concerned
The affected enterprise should “at the same time”
inform its authority of any third and other Contracting States which could be
concerned and the said authority should “then without delay” notify the
competent authorities of those Contracting States.
It is advisable to effect a protective notification about other states
which could possibly become involved.
Three-year term for complaint
The complaint must be presented within three years of the
first notification. The three year
period sets aside the shorter two year time limit applicable to the mutual
agreement procedure under some bilateral treaties between Contracting States.
The said term starts running from the first notice (in terms of Article 5)
of the threatened adjustment which would result in a second assessment of the
Procedural rules of conduct
The negotiations between the Contracting States are
essentially diplomatic and these will themselves determine the procedural rules
for the conduct of such negotiations. The
Convention does not impose any such rules.
Moreover, the legal position of the affected taxpayer is not recognised
as the latter and its competent national tax court are not bound by the
Time-limit for mutual agreement
Contracting States are allowed to secure a solution by
mutual agreement during a period of two years commencing on “the date on which
the case was first submitted to one of the competent authorities in accordance
with Article 6(1).”
This is considerably shorter than the term applicable under the mutual
agreement procedure contemplated in bilateral DTCs, which may last as long as
However, there are two ways of looking at this.
Hinnekens argues that the competent authorities, “aware of the final
say of the arbitration commission looking over their shoulders, may let this
period expire without active and complete examination of the case.”
On the other hand, other tax lawyers argue that the competent authorities
will tend to utilise this term well so as to decide themselves the applicable
solution, rather than leave it in the hands of the arbitral commission.
Domestic v. international proceedings
The case is referred to and settled by international
procedures, “irrespective of the remedies provided by the domestic laws of the
Contracting States concerned.”
The domestic and international procedures are not mutually exclusive and
do not compete as the mutual agreement is only binding on the Contracting States
and not on the national courts. The
affected taxpayer is therefore placed in a tactical position, facing optional
parallel proceedings. An enterprise
can choose one or the other, or a combination of both.
In the latter case, the enterprise can choose to call for the
discontinuation of the international procedure if it is satisfied with the
elimination of double taxation afforded by a decision of the national courts.
The arbitration procedure
Procedure and composition of the advisory commission
If the mutual agreement procedure yields no results within two
years of the case being presented, the competent authorities must set up an
advisory commission to deliver an opinion on the elimination of double taxation.
If more than one competent authority is involved, the
procedure may be conducted as a single multilateral arbitration in which the
competent authorities of the enterprises concerned participate simultaneously.
In the more complex cases, this option is preferable to separate
procedures, as the latter, done on a piece-meal basis, tend to produce results
which are inconsistent or incoherent in the multilateral context.
advisory commission comprises two representatives of each competent authority
and an even number of independent persons to be appointed by mutual agreement or
by ballot by the competent authorities,
and these representatives and independent persons shall jointly elect a chairman
from a list of independent persons.
There is an uneven number of members of the commission and the majority
decision is likely to depend on the votes of the independent members if the
representatives of the two competent authorities should not be able to reach an
The commission must deliver its opinion within six months
and such opinion shall be adopted by a simple majority of its members.
Where an enterprise pursues a remedy before a domestic
court or tribunal, the two years within which the competent authorities have to
reach agreement run from the date of judgment of the final court of appeal.
Also, the submission of the case to the advisory commission shall not
prevent a Contracting State from initiating or continuing judicial proceedings
or proceedings for administrative penalties in relation to the same matters.
If, within a further six months, agree cannot be reached
on alternative steps to eliminate double taxation, the competent authorities
must accept the commission’s decision.
Legal nature and consequences
There are two schools of thought regarding the legal
nature of the commission procedure and decision under the Convention, as
An administrative act
Some maintain that the mechanism contemplated in Articles
7 to 11 is meant to address the absence of an obligation to reach an
agreement which characterises existing bilateral DTCs
“without going as far as creating an arbitration procedure which would deprive
Member States of their sovereignty over tax matters.”
Thus, according to the proponents of this theory, the
final decision to settle under the Convention remains in the hands of the
Contracting States involved and the commission mechanism merely forces the
Contracting States to come to a mutual agreement and eliminate an instance of
double taxation in a way chosen by common consent. Under Article 12, they are free to take a decision which
deviates from the advisory commission’s opinion and only if they fail to agree
are they obliged to adopt the commission’s opinion.
Moreover, the terminology adopted by the Convention seems
to reinforce this view (“advisory” and “opinion”) as does the two-step
system leaving the decision formally with the competent authorities.
One can be lead to conclude that the legal nature of the procedure is not
a jurisdictional act and that the decision is not taken by a supranational
arbitral body but by the Contracting States.
Furthermore, the changes in legal form from a directive to
the present Convention, forfeiting in the process such characteristics as the
express terminology indicating arbitration (“arbitration”, “decision”)
and the outright arbitral procedure,
constituted a refusal of the Contracting States to accept the entailed incursion
on their sovereign jurisdictional rights.
Thus, the commission procedure is an intergovernmental,
and therefore diplomatic, agreement of a purely administrative nature and
therefore its opinion is not binding on the Courts of the Contracting States.
A jurisdictional act
By way of counter-argument, it is submitted
that a number of factors demonstrate that the Convention is indeed about
arbitration and that it does involve the partial surrender of the Contracting
States’ fiscal sovereignty, even if, and this is not disputed, the Contracting
States’ intention was to minimise the forfeiture of a degree of their fiscal
The commission’s opinion is, even if indirectly, not
only potentially binding, but very likely to be so.
It is argued that if two Contracting States were not able to agree on a
solution at mutual agreement stage, it is highly unlikely that they would agree
differently once an arbitral decision has been pronounced – the vantage point
of the party winning the arbitration is such that it would not be in this
party’s interest to seek such alternative.
So likely is it that the opinion will be binding, that the Convention
uses the terms “opinion” and “decision” interchangeably and, even if
exceptionally, also the term “arbitration procedure” (heading of Chapter II,
We find other factors backing this reasoning: the rules on
the composition of the commissions, the qualifications and secrecy obligations
of their members, the institutional framework and own procedural rules as well
as the recognition of taxpayers’ rights.
All these transform the commission into a truly arbitral body, separate
from the Contracting States, and its opinions into truly jurisdictional and
It follows therefore that the affected enterprises can
only bring before its national court cases of Contracting States’ failure to
fulfil their obligation. A dilemma
concerning cases for the enforcement of an arbitral award is that the national
courts will have the choice of enforcing either the award or a different
bilateral transfer pricing ruling on the same transaction, to which the
competent authorities would have agreed earlier on and which would be binding on
them. The only determinative and
available solution for the Court is to uphold the award.
Appeals in domestic courts
One’s approach to the question of whether an appeal is
possible from the solution resulting after the four-stage process introduced by
the Convention is determined by whether one forms part of one or the other of
the above schools of thought.
The advocates for the administrative nature of these
procedures conclude that, being simply a decision agreed to between the
administrative authorities of the Contracting States, such decision should be
appealable before the domestic courts.
This also applies to agreements between Contracting States under the
mutual agreement procedure applicable under their bilateral DTCs.
The proponents of the alternative doctrine maintain that
the commission’s decision is jurisdictional in nature and the commission’s
authority supranational and therefore, such decisions do not admit of appeals in
foro domestico. However, is it
possible to have an international review of the arbitral decisions?
Appeals in the international domain
Such a possibility can be interpreted of Article 13
which lays down that “the fact that the decisions taken by the Contracting
States … have become final shall not prevent recourse to” the mutual
agreement procedure and the arbitration procedure afresh.
Some authors have interpreted this to imply the possibility of
enterprises to start over the international procedure if they believe that the
“decision taken by the Contracting States” does not comply with the
Convention’s principles or that the decision did not consider facts which
happened subsequently. Even if this
view should be upheld, the time limits laid down by the Convention will not
start running afresh and therefore, taxpayers will be able to re-open
international procedures, subject to the three year time limit imposed by
In a democratic society, it is expected that the
post-award controls incorporated in the 1950 European Convention for the
Protection of Human Rights, in the 1981 International Covenant on Civil and
Political Rights and in the 1958 Convention on the Recognition and Enforcement
of Foreign Arbitral Awards should form part of the taxpayers’ rights under the
Convention and that therefore, the final decision imposed by the Convention
should be subject to appeal also on the basis of these treaties.
The nature of the final decision
The commission’s decision is only potentially binding as
its binding force is conditional to the competent authorities, within a further
six months, failing to reach an alternative solution to the dispute.
The authority of the advisory commission, therefore, is limited to
submitting an opinion containing a decision capable of being implemented by the
tax authorities if these should fail to reach an agreement..
It is up to competent authorities to act by common consent on the basis
of Article 4 and take the final decision which has to eliminate the double
taxation. They may deviate from the
opinion rendered by the advisory commission or allow the commission’s opinion
to become binding.
Whichever of the two options is chosen, the double
taxation of profits shall be regarded as eliminated if “the profits are
included in the computation of taxable profits in one State only; or the tax
imposed on those profits in one State is reduced by an amount equal to the tax
payable on them in the other State.”
The solution will depend on whether the Contracting States adhere to the
exemption method or the tax credit method.
Cases where the right to tax an item of income is divided between the two
States would be incompatible with the exemption method.
The Convention fails to address cases where one or both of the
enterprises incur a loss even though they are expressly targeted by Article 1(3).
According to Killius, “where a loss carry-forward is
available, a reduction of a loss as a result of the re-allocation of income will
just defer the incidence of double taxation. Conversely, if a loss carry-back is
available, a lower amount of tax paid in previous years would be refundable.
These problems may be aggravated under imputation systems or where other tax
credits may be available. None of these aspects are specifically addressed in