Tải bản đầy đủ - 0 (trang)
3 The Arm´s Length Standard as an Independent Benchmark (Hypothetical Reference System/Comparability Approach)

3 The Arm´s Length Standard as an Independent Benchmark (Hypothetical Reference System/Comparability Approach)

Tải bản đầy đủ - 0trang

Double Taxation Relief, Transfer Pricing Adjustments and State Aid Law


actually created system in order to find that all companies so established would be

in a comparable situation in light of that objective, even though it has nowhere been

stated (nor was it obvious from the way the various taxes were designed) that this

would indeed be the objective. In fact, a more appropriate description of the

system’s objective would have been to “create a tax system for all companies

with physical presence in Gibraltar”. In light of that objective, offshore companies

would clearly not be comparable to companies with physical presence there.51

In its earlier decisions, the Commission appeared to be hesitant about its

approach. It typically started with the claim that the determination of an advantage

required comparing the method of profit calculation applied to a MNE “to the

ordinary tax system, based on the difference between profits and losses52 of an

undertaking carrying out its activities under normal market conditions”.53 This

could be understood as an attempt to ascertain the actual “ordinary tax system” of

the Member State under scrutiny as a point of reference, with an ensuing search for

a derogation in the tax base calculation used under the specific circumstances

concerning a certain taxpayer.54 But in fact, that is not what the Commission was

looking for: instead, it examined whether the tax system55 of a Member State took

measures to respond to the different situation groups of companies find themselves


The ECJ thus seemed to employ a similar strategy to the one uses in its case law on economic

double taxation, where it ignores certain elements of a Member State’s goals with a certain tax

measure: in Manninen, the ECJ held the relevant objective of the tax credit system was merely to

“avoid economic double taxation” and thus easily found comparability between income derived

from domestic companies as compared to income derived from foreign companies, as both could

be subject to economic double taxation. The ECJ thus ignored the real objective of the system,

which was to avoid economic double taxation occurring as a consequence of Finland’s own

taxation. This approach may be sound in substance if it can be shown that the “real” objectives

of a Member State are themselves contrary to the internal market, but the ECJ did not undertake

that analysis and simply re-interpreted the national tax systems underlying principles.


Note: “the difference between profits and losses” does not make much sense, and should be

explained as a mere mistranslation from the original French “la diffe´rence entre les produits et les

charges” which rather means “the difference between receipts and costs”, i.e. a simple calculation

of commercial profits.


E.g. Preliminary Decision SA 38944 (Amazon), paragraph 53; see also Preliminary Decision SA

38375 (FFT), paragraph 60.


One should note that finding such derogation, e.g. with respect to the profit allocation of

companies that form part of an international group, would not necessarily result in a finding of

prohibited aid, as such derogation could be consistent with the nature or general scheme of the tax

system. A tax system may legitimately distinguish between domestic and cross-border situations

because of the need for coordination with the tax jurisdiction of other countries; see the discussion

above concerning the potential different treatment under different relief mechanisms to the extent

that they lead to double taxation relief. I shall revisit this issue again below when discussing the

search for the right comparator.


As explained above, the fact that the concrete cases before the Commission relate to individual

tax rulings appears to be of little import: there is no reason to believe that the Commission would

take a different view if the treatment of the multinational enterprises under scrutiny were fully

compatible with the tax rules laid down in domestic law.


W. Haslehner

in as compared to independently operating entities, because the former, but not the

latter, are in a position to manipulate prices and thus their accounting profits. As the

Commission stated in its Amazon decision:

When accepting a calculation method of the taxable basis proposed by the taxpayer, the tax

authorities should compare that method to the prudent behaviour of a hypothetical

market operator, which would require a market conform remuneration of a subsidiary

or a branch, which reflect normal conditions of competition.56

The Commission’s reliance on an “EU arm’s length principle” independent from

the OECD rules and domestic regulations on transfer pricing has now become most

clear from is formulation in its decision on the Belgian Excess Profit Regime, where

it noted:

The arm’s length principle therefore necessarily forms part of the Commission’s assessment under Article 107(1) of the Treaty of tax measures granted to group companies,

independently of whether a Member State has incorporated this principle into its national

legal system and in what form.

… the arm’s length principle that the Commission applies in its State aid assessment is

not that derived from Article 9 of the OECD Model Tax Convention and the OECD TP

Guidelines, which are non- binding instruments, but a general principle of equal treatment

in taxation falling within the application of Article 107(1) of the Treaty, which binds the

Member States and from whose scope the national tax rules are not excluded.57

Underlying this approach is thus really a comparison (and assumption of comparability) of independently acting undertakings and such entities that also

(or exclusively) transact with associated enterprises. As the former (necessarily)

transact at market prices, a tax system that calculates taxable profits to consist of the

result of receipts minus costs as achieved in a free market (i.e. based on such market

prices), must in turn apply measures to ensure that a comparable measure of profit is

applied to companies that do not (necessarily) transact at market prices. The

Commission supports this (implicit) comparison with the Court of Justice’s

endorsement of this approach in its decision concerning Belgian Coordination

Centres.58 This certainly appears to be sound, if the pair of comparison is and has

to be that described, taking, as the Court suggests, the objective of the tax system as

a whole as the relevant starting point. It also seems to match perfectly with the idea

of the arm’s length standard, which is why the Commission can complete the circle

to this with stating:


E.g. Preliminary Decision SA 38944 (Amazon), paragraph 55; see also Preliminary Decision SA

38375 (FFT), paragraph 62 (emphasis added).


Decision SA 37667, published 4 May 2016, paragraph 150.


ECJ C-182/03 and C-217/03, Belgium and Forum 187 v. Commission, ECLI:EU:C:2003:385,

paragraph 95: “Pour examiner si la de´termination des revenus imposables, telle que pre´vue dans le

re´gime des centres de coordination, procure un avantage a ces derniers, il y a lieu, comme le

sugge`re la Commission au point 95 de la de´cision attaque´e, de comparer ledit re´gime a celui de

droit commun fonde´ sur la diffe´rence entre produits et charges pour une entreprise exerc¸ant ses

activite´s dans des conditions de libre concurrence”.

Double Taxation Relief, Transfer Pricing Adjustments and State Aid Law


In this context, market conditions can be arrived at through transfer pricing established at

arm’s length.59

Yet it is debatable whether this comparison is indeed appropriate. Prior to this,

however, two problems with the Commission’s reasoning merit separate scrutiny.

First, the Commission is not entirely clear (at least in its preliminary decisions)

as to what comparison it suggests, and states that the advantage from a tax base

calculation that is not at arm’s length “provides the taxpayer with a more

favourable treatment as compared to other companies which are in a similar

factual and legal situation. Those companies are either domestic, i.e. non-multinational companies whose taxable profit is calculated on the basis of the difference

between a company’s income and charges or, if they are multinational companies,

their taxable profit is derived at on the basis of a correct application of the arm’s

length principle.”60 Thus, the Commission leaves it open whether the taxpayer in

question should be compared to a domestic group company or to another multinational. Yet both comparisons are flawed in the context of the “comparability

approach”. The first comparison because a company that is a member of a domestic

group of companies obviously faces the same opportunities to manipulate its prices

in transactions with other group companies; it might not face the same incentives to

do so, although effective tax rates can also differ for companies of a domestic

group, but it is clear that the profits as calculated by a simple subtraction of costs

from receipts where both are from transactions with associated enterprises will not

reflect “normal market conditions”. What is more, Member State may generally

reserve the application of strict transfer pricing rules to cross-border situations and

are not obliged to impose similar adjustments on purely domestic groups. If

domestic groups and cross-border groups were comparable ‘in light of the objective

of the tax system’, that different treatment would result in prima facie advantage

provided to companies that deal solely with domestic associated enterprises. It is

unlikely that this would result in a finding of actual state aid, however: as the Court

accepts the application of (strictly proportionate!) transfer pricing rules to be

confined to cross-border situations within the scope of the fundamental freedoms,61

it would probably also find such differential treatment to be justified by the nature

and general scheme of the tax system. In the absence of a need to allocate profits

properly to taxpayers in line with an agreed allocation of taxing rights vis-a-vis

other countries, there is no reason to require transfer pricing adjustments in the case

of purely domestic transactions, which is entirely logical and consistent.62 This only


E.g. Preliminary Decision SA 38944 (Amazon), paragraph 54; see also Preliminary Decision SA

38375 (FFT), paragraph 61.


E.g. Preliminary Decision SA 38944 (Amazon), paragraph 60.


See e.g. SGI (C-311/08, EU:C:2010:26).


One might also be tempted to argue that there can be no “aid” in that case because the State does

not renounce any resources: any tax forgone from one member of the group would be collected

from another member of the same group. For this to be true, however, it is necessary to look not at

the individual company, but rather the group as a whole. If not the group, but the individual

company is the taxpayer under the corporate tax system, such an approach does not seem


W. Haslehner

serves to show, however, that a comparison between an international group and a

purely domestic group makes little sense in the first place.63 The second comparison, between two multinational enterprises, seems more straightforward. However,

it only makes sense if the Member State in question actually applies the arm’s

length standard to certain companies; it fails where the arm’s length standard is

simply not part of domestic law. But if it is, then there is no need for the use of the

“comparability approach” at all, as it would be easy to identify the application of

the arm’s length standard as the domestic reference system and the lack of its

application in certain circumstances as derogation therefrom.

Second, the Commission treads on treacherous terrain with its reference to a

“prudent hypothetical market operator”, who “would not accept that its revenues

are based on a method which achieves the lowest possible outcome if the facts and

circumstances of the case could justify the use of other, more appropriate

methods”.64 This could be stating the obvious, namely that an arm’s length profit

needs to be determined by the use of appropriate methods and comparables, in

which case there was no particular reason to refer to the concept of a prudent

hypothetical market operator; it might also signify, however, something different:

potentially a deviation from the OECD Transfer Pricing Guidelines, which rely on

“actual comparables” and shy away from mere “hypotheticals”.65 It has been

suggested that the concept bears resemblance to the “market economy investor

principle”, which is used to assess the Member State’s behaviour allegedly as

private investors.66 However, there is no clear link between the two other than a

certain similarity in the criteria applied to the assessment of “market-like”

(i.e. exclusively self-interested) behaviour. The more interesting question is

whether the concept indicates a deviation from the way the OECD interprets the

arm’s length standard. It would not be convincing to argue that the arm’s length

warranted: Theoretically, any resources the State uses to provide “aid” to a certain entity have to be

collected elsewhere—in the absence of any productive market activity of the State, necessarily

from another taxpayer. If we were to look beyond the concrete taxpayer to define the “renunciation

of revenue”, we would thus never find any aid. Practically, the distinction between taxpayers

matters: the tax not collected from one entity may not be collectible from the other entity,

e.g. because it is insolvent; also, transfer pricing manipulation may be used domestically to offset

losses and thus defer taxation, which would again result in a reduction of tax revenue.


In fact, the “justification” with the “nature and general scheme of the tax system” makes

similarly little sense as it merely repeats the exercise already undertaken in the comparability

analysis. This is notably different from the situation under the fundamental freedoms: there, the

ECJ (also) accepts justifications which lie “outside” the internal logic of the system in question,

whereas it only looks for “internal” justifications applying state aid rules (cf. Commission State

Aid Draft Notice at paragraph 138; see also Paint Graphos (C-78/08, EU:C:2011:550, paragraph



E.g. Preliminary Decision SA 38944 (Amazon), paragraph 55; see also Preliminary Decision SA

38375 (FFT), paragraph 62.


See Bullen (2011), p. 335.


See Gunn and Luts (2015), p. 119 at 123.

Double Taxation Relief, Transfer Pricing Adjustments and State Aid Law


standard is a relevant benchmark for state aid analysis simply because it is a widely

accepted international standard. Rather, the arm’s length standard as understood in

international tax law is only incidentally the “correct” standard from a state aid

perspective, despite the fact that the objectives of the OECD on the one hand and

the EU rules on state aid on the other hand are quite different: the former aims to

achieve a generally acceptable measure to allocate profits between sovereign

countries torn between interests of competition and coordination; the latter aims

to protect (or rather: create) an internal market in which Member States do not

subsidize certain enterprises and thus distort free market outcomes. It is thus

perfectly possible (although it appears unlikely in practice) that the EU law concept

of the arm’s length standard for purposes of state aid review deviates from the arm’s

length standard as proposed by the OECD Transfer Pricing Guidelines.67 Indeed,

the reference to a hypothetical private market operator seems to echo not so much

the OECD Transfer Pricing Guidelines as the concept of the “hypothetical arm’s

length comparison” (hypothetischer Fremdvergleich) carried out under Germany’s

§ 1 III AStG, which entails the determination of transfer prices on the basis of what

two hypothetical prudent and diligent directors (“ordentliche und gewissenshafte


aftsleiter”) would have accepted in a fully-transparent negotiations. This

approach, which appears sound in principle, is easily criticized on the basis that it

does exactly not reflect true market behaviour, not least because true markets are

not characterized by complete information symmetry.68

Returning to a more general reservation about the Commission’s approach, is

the comparison between companies that engage in cross-border transactions with

other members of one multinational enterprise and companies that engage in crossborder transactions with independent contracting partners the correct one? When

measured against the alternatives, this ought to be answered in the affirmative.

Since the arm’s length standard is designed specifically for a problem that occurs in

cross-border situations, it makes little sense to involve a purely domestic taxpayer

in the comparison. Similarly, while trying to determine the independent normative

value of the arm’s length system under state aid law, it is not useful to compare two

multinationals where one is exceptionally not subjected to its more rigorous rules.

It follows naturally therefrom that the correct comparison should hold everything

else equal apart from the source of the potential distortion of competition, which is

the possibility to reduce the tax burden through the application of non-market price

transactions that follows from membership in an international group.

However, the choice of this pair of comparison does not mean that the arm’s

length standard provides the only right measure of profits for a company that is part

of an international group. It is not certain that independent companies are truly in


This notwithstanding the fact that recent developments surrounding the BEPS debate raise

questions as to the sustainability of the arm’s length standard, as various suggestions for changes

to the Transfer Pricing Guidelines seem to signify effective derogations from its content.


See e.g. Eigelshoven (2015).


W. Haslehner

similar legal and factual circumstances with companies that are members of a group

of companies. This needs to be assessed separately. Two views appear prima facie

relevant to such assessment: an “economic” or a “doctrinal” perspective. The

economic approach, in short, posits that the arm’s length standard is not the correct

standard to assess the profits of a member of a group, as it necessarily ignores any

benefits that result from the integration in the group even though such benefits

explain its very existence.69 The doctrinal approach, by contrast, seeks to assess

comparability in light of state aid law’s concern for free competition between

independent undertakings. That latter approach would thus create a benchmark

that deliberately ignores economic differences between integrated and independent

businesses, in order to prevent the former from gaining an advantage as a consequence of their integration. This appears to resonate with the argument advanced by

Lang that the proper comparison should be made on the basis of the competitive

relationship between entities.70 In principle, there is no reason to assume that group

companies and independent companies are not in direct competition with each

other—this holds true even if a group company only provides services within its

group and thus does not appear to participate in the “free” market—competition

still exists as the group has to decide whether to outsource activities to an independent enterprise or keep it within the group. If state aid law is concerned with

potential distortions of that decision, it might thus be justified that tax considerations should not influence that decision. Under that approach, the imposition of the

arm’s length standard as an independent standard flowing directly from the EU

competition rules could be justified. There should be no illusions about the meaning

of such approach, however. If comparability of independent companies with companies that are members of a group were confirmed, it would not result in an

alignment of tax rules with economic reality. It would rather appear to transform

tax rules into an anti-trust tool that negates efficiency advantages from the integration of separate enterprises. Paradoxically, however, if that approach were applied

consistently in all countries—i.e. to mandate downward adjustments as well as

upward adjustments in order to arrive at a taxable profit that corresponds to the

market return achieved by non-integrated enterprises, it would lead to the dislocation of any excess profit generated by group synergies and thus inadvertently even

promote integration. If, as appears more likely, the approach will only be employed

to instigate upward adjustments, it might, by contrast, result in double or multiple

taxation of such excess profits.


Brauner (2013), p. 387; Sch€

on (2012b), p. 47 at 53 et seq.; Wilkie (2012), p. 137; Vann

(2010), p. 291.


See Lang (2012), pp. 85–115 at 99. See also Lang (2011), p. 593 at 598.

Double Taxation Relief, Transfer Pricing Adjustments and State Aid Law



Adjustment Mismatches and the (Ir)relevance of “White

Income”/Double Non-taxation

So far, the analysis has concerned the application of the arm’s length principle in

one Member State under the implicit assumption of corresponding rules applied in

other countries of business for a multinational enterprise. The case of adjustment

mismatches, which can result in tax-specific advantages that are not related to

efficiency gains from integration and are thus not defendable on the basis that

they reflect economic reality, are quite another matter. It appears that at least certain

of the recent investigations of the Commission into transfer pricing arrangements

were driven by concerns about such practices.71

As briefly outlined above, a result of double non-taxation is a priori neither

necessary nor sufficient for a finding of state aid. However, where a Member State

seeks to justify a derogation from the normal tax system on the basis of that

system’s nature and general scheme and, in particular, the principle of ability to

pay, double non-taxation as a consequence of such derogation is highly suspicious.

If a tax reduction in one country is offset by a tax burden in another jurisdiction, the

ability-to-pay principle suggests that no relevant advantage exists. The same will

not be true if the actual tax burden is lower compared to the purely domestic

situation or even results in “white income”. Although that outcome could be

dismissed as a consequence of the other country’s unilateral decision concerning

its domestic scope and level of taxation in combination with the effect of a bilateral

agreement with that other state, it remains the case that the first state has reduced its

tax claim without attendant justification as provided by the ability-to-pay principle.

The arguments for this position are even stronger in the context of mismatching

transfer pricing adjustments for related companies in comparison to the case

discussed previously,72 where white income resulted from the application of the

exemption method in a situation of merely virtual double taxation. This is so

because the case of mismatching transfer pricing adjustments concerns separate

taxpayers and the ability-to-pay principle as a taxpayer-oriented (rather than an

enterprise- or state-oriented) principle makes the tax burden of each separate entity

the relevant benchmark of a normative corporate income tax system. A unilateral

downward adjustment that cannot be expected to be matched by a corresponding

upward adjustment by another jurisdiction is thus likely to fall foul of state aid


A difficulty remains, however, to determine the ability to pay of a company that

is entirely embedded in a multinational group on a stand-alone basis: in the absence


See e.g. European Commission (2015). Similar arguments have been raised with respect to the

Starbucks case, where payments deducted by the Dutch company to its British sister entity was

allegedly not taxable in the UK as a consequence of that entities hybrid nature.


Supra Section 2.


This is clearly the Commission’s view in the case of the Belgian unilateral adjustments under its

Excess Profits tax scheme. See European Commission (2016).


W. Haslehner

of its relationship to the other member of the group, its profit may well be zero.

Even though a certain downward transfer pricing adjustment may thus appear to

constitute an advantage, the correct remedy is not obvious: such adjustment might

indeed have been required by the arm’s length principle. One is thus thrown back to

the question as to whether the arm’s length standard can act as the proper benchmark to determine the “correct” profit. Two examples of transfer pricing adjustments are briefly discussed here. Both examples are based on the view rejected

above, that the arm’s length standard is a mandatory feature of Member States’ tax

system: The first leads to “white income”, but should still be accepted from a state

aid perspective (even) if the arm’s length standard is used as a normative standard.

This example reinforces the point made earlier, namely that mismatches are not per

se relevant for state aid review. The second aims to show that transfer pricing

adjustments that are contrary to the arm’s length standard should still not be

considered state aid (even) if the arm’s length principle were accepted as an

independent benchmark, when they are based on the need to avoid economic double

taxation, which renders concerns regarding distortionary effects of the adjustment

measure unfounded.

As concerns the first situation, if a Member State makes a downward adjustment

to a resident company’s profits in line with the arm’s length standard, it does not

grant a selective advantage to the taxpayer even if there is no equivalent upward

adjustment in another jurisdiction. This is so despite the fact that “white income”

arises as a consequence to the extent that a part of the overall economic profit of the

associated entities remains untaxed as a consequence. If the arm’s length standard is

mandatory as a matter of state aid law, it must also be sufficient to comply with its

rules. The fault in this case lies thus squarely with the other country. If the other

jurisdiction were a Member State, the Commission would certainly try to require an

upward adjustment from that state in order to avoid the mismatch, but it obviously

has no such power where that jurisdiction is a third country. Such limitation cannot,

however, affect the legitimacy of the Member State’s conduct.

With regard to the second situation, if a Member State makes a corresponding

(downward) adjustment to avoid economic double taxation in a situation where the

corresponding first (upward) transfer pricing adjustment was not in line with the

arm’s length principle, the question arises whether such relief would constitute

illegal state aid. The company in question will certainly receive a benefit from state

resources (if the comparison is a domestic independent company making a higher

“normal” market profit) and the tax treaty cannot provide a justification, as it does

not require an adjustment in cases where the other country does not follow the arm’s

length standard. Similarly, the ability-to-pay principle will not easily justify that

treatment as it is focused on the situation of the single taxpayer. However, state aid

law is not confined to so narrow a perspective. Instead, it applies to the undertaking

as a whole, which is the multinational enterprise.74 In the case at hand, however, the

granting of relief, if fully matched by an upward adjustment in the other country,


See Rossi-Maccanico(2015), p. 63 at 67.

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

3 The Arm´s Length Standard as an Independent Benchmark (Hypothetical Reference System/Comparability Approach)

Tải bản đầy đủ ngay(0 tr)