1. Preface. The tax regime of corporate financing between national anti-abuse measures and international double taxation

Corporate financing can be conducted primarily via the allocation of own capital or via debt contracts with third subjects (independent financial institutions or financing companies within the same multinational group). The national tax provisions applicable to dividends and interest are generally different. Therefore, one could wonder whether European law tends to favour one of these two forms of corporate financing or if on the contrary, it is neutral towards them.

The existence of different tax regimes for interests and dividends at the level of a single Member State can have a twofold effect. On one hand, there could be a disproportion of the tax treatment of interest compared to dividends. This situation could have the effect of stimulating companies to increase the financing via debts stocks only because of tax convenience (thin capitalization). Therefore, this phenomenon is generally related to the creation of anti-abuse provisions at the national state level. On the other hand, a potentially different tax treatment can have an impact at an international level due to the lack of coordination between national tax systems. Thus, double taxation could take place (or, more rarely, a double non-taxation).

Furthermore, it is necessary to verify whether and within which limits the introduction of such anti-abuse measures and the phenomena of double taxation are admissible in European law, especially considering the judgements of the European Court of Justice (ECJ).

At the same time, this analysis will verify whether the judgments of the ECJ also allow the identification of a preference for one of the two methods of company corporate financing (equity or own capital) at a European level, that indirectly influence the choices of tax legislators at the member state level.

Such an analysis will also consider the impact of the limited, but still existing, EU secondary legislation in the area of direct taxation ^[2], as it is interpreted by the ECJ.

2. National regimes of (in)deductibility of dividends and interest. Comparison

As highlighted, there are obvious disparities of treatment in the regime applicable to these two categories of income. Prima facie interest is definable at an internal level as well as an international level as remuneration of debt capital. It is usually considered as a deductible cost in the calculation of the base of income tax paid to the state in which the company receiving the loan is established. On one hand the payer is exempted from the taxation related to the amount paid as interest and can further deduct it from the tax base. On the other hand, this is counterbalanced by the fact that the beneficiary of the interests will be taxed. Nevertheless, at the national level restrictions and limitations on the deductibility of interest may exist especially vis-à-vis the tax avoidance schemes related to the payment of interest to debtors established in countries with a lower taxation ^[3]. In Europe, it is necessary to verify the compatibility of such provisions with the Treaty FUE and with the EU directives.

Furthermore, broadly speaking, dividends are the positive economic result of corporate financing via equity or own capital. They are taxable according to different principles. Like interest, the profits of a company derived from the management of own capital are generally taxable in the state where the company that realized the investment is established ^[4]. Differently from interest, dividends cannot be deducted from taxable net profits and with the exception of corrective measures are generally subject to economic double taxation, on both the internal and international levels. Sometimes, a tax credit or a tax exemption may be granted – relative to the withholding tax and/or tax already paid by the company, based on internal law provisions (most of the time limited to internal situations) or in an EU context, based on the parent-subsidiary directive.

Such differences in the tax treatment are less important if these items of income are received by non-residents. This is due to the fact that, usually, the national tax systems provide a withholding tax on dividends or interest received and a variable tax rate according to the type of income and to the different legal system involved. The withholding tax might create situations of double taxation in the country of source, both for dividends and for interest (in the case that they are considered non deductible). However, this negative impact can be reduced because of the application of the EU directives ^[5], international agreements or internal law provisions.

Besides EU law provisions aimed at reducing economic double taxation of intra-group dividends, domestic me a sures have been adopted with the same aim, like for example the Belgian so-called notional interest ^[6] regime. It allows the deduction of an amount equal to a fixed percentage of the company’s own capital (as if own capital had been borrowed with interest at market price). In other words, Belgian or foreign companies with an permanent establishment in Belgium ^[7] are allowed to deduct the “notional” interest on the tax base of the company’s income tax ^[8]. On this basis, a tax benefit has the positive effect of reducing the inconvenient of the non deductibility of dividends. Consequently, in the Belgian system, one could say that tax law tends to favour own capital over debt.

At first glance, the notional interest deduction cannot be qualified as state aid prohibited by European law as it is applicable to all companies in Belgium and not only a certain category of companies (as it happened with the previous system of the coordinating centres, substituted by the notional interest system). Nevertheless, if on one hand, the notional interest cannot be legally qualified as state aid, on the other hand such a system actually gives a great advantage to the companies that are strongly capitalised and in particular to the financing companies managing the infra-group loans within multinational groups ^[9] .

3. Characterization of dividends and interest in EU law: the influence of international law

In the EU context, the definition of dividends and interests is of major importance ^[10]. In the absence of equal treatment of the different ways of financing, it is indeed essential to distinguish among dividends, interest and other forms of income. The first problem to deal with is thus exploring whether in EU law, tax provisions could force Member States to uniformly qualify the existing categories of income deriving from corporate financing. The existence of differences of qualification is one of the reasons of international double (non-)taxation: in the presence of a neutral taxation system, it would be irrelevant, for tax purposes, that a company finances itself through equity or debt ^[11].

Regarding interest, EU law (Article 2 of the Directive 2003/49/EC ^[12]) lays down legislative definitions of interest (and royalties) ^[13]. Article 4 of this Directive enumerates four cases of exclusion of payments in the form of interest from the scope of the Directive. The definition of the Directive reflects the provision of Article 11 of the 2003 OECD Model. Therefore, in this field, EU law uses concepts and definitions included in the OECD Model ^[14].

The definition of interest appears to be particularly extensive. It also includes income originating from hybrid financial instruments that enable debtors to participate in profits. However, on the basis of the articles 2 and 4 of the Directive, in certain cases member states can deny its application. In fact, Article 4 refers to different situations that potentially allow the restriction of the definition of interest. This provision allows member states to counter possible abuses of the Directive, e.g. when interest paid on income is actually finalised to avoid the profit distribution among shareholders (besides the general anti-abuse clause contained in Article 5).

Regarding dividends, the Parent-Subsidiary directive lacks a precise definition of distribution of profits that would permit a clear distinction between dividends and interests. This could be explained by the fact that some definitions already existed in international tax law. Therefore, it seems that EU law has considered the definition of dividends and interest of the OECD Model (articles 10 and 11) as a reference point.

Article 11 of the OECD Model defines interest in an autonomous way, differently from what is said in paragraph three of Article 10 OECD MC concerning dividends. The origin of the definition of interests can be historically traced back to the working group 11 of the OECD in 1959. Originally, even interest from bonds related to profits was defined as dividends, which was later used in the definition of interest. Throughout the different versions of the OECD Model the definitions of dividends and interest contained a list of requisites that could be utilised as a basis for the definitions established on the national legislation level. Although the national legislation clarifies if a payment can be considered as interest or dividend, it is necessary to verify whether the national definition of interest can include one or more categories considered as “dividends” by the OECD model.

Therefore there is a potential overlap between the two above-mentioned definitions, which is univocal. The definition of dividends can include aspects related to the definition of interest, but not vice versa.

The OECD Model does not offer solutions for this overlap. Therefore a threefold observation can be made. Firstly, the reference to the participation of profits of the debtor on the basis of Article 11 OECD MC was originally only stated with regard to dividends. Secondly, the second point of Article 10 excludes debt-claims from the notion of dividends. Thirdly, point three of Article 10 excludes the “normal” debt. When considering all these aspects at the same time it is clear that Article 11, regarding interest, could have originally been intended to have priority over Article 10 (dividends). In other words, the fact that payments are considered as interest implies that they can not automatically be considered as dividends under the provision of Article 10.

Thus, interest paid as the price of a structured-bearing loan and included profit can be considered as “interest” as well. Nevertheless, if it is related to a profit quota and to the risks of the company’s activity it cannot be considered as a loan. In this case, the economic result should be considered as a participation in partnership profits, rather than a thin capitalization case. Nevertheless, the Thin Capitalization Report ^[15], considers this situation as a case of thin capitalisation. Therefore, one can argue that Article 10 OECD MC is not only related to dividends, but also to interest from loans if the creditor actually takes part in the risk of the company.

Such overlaps can only be excluded by a national or European ad hoc legislation explicitly stating that the received payments in the form of interest automatically exclude all amounts ascribable to the definition of dividends (thin capitalization). If de jure condendo such a definition should be acknowledged in the draft of an OECD Model Tax Convention and by EU law, in the current state of play it creates some difficulties and confusion. Moreover, member states are not entitled to fulfil the non-binding interpretations provided by the OECD Model or by the EU. Therefore it is possible that the same capital flow could be qualified as interest in one member state and as dividends in another. This circumstance could create a situation of a double (non) taxation.

4. European limits on the anti-abuse provisions related to thin capitalization ^[16] and on interest deduction. The ECJ case-law

The problem of the characterization of income is the result of a situation in which companies are almost completely free to choose their financing structure, and therefore the type and location of the associated companies to which they prefer to transfer net profit to their shareholders as interest rather than as dividends. In this case, there is the risk that, in the presence of a group of companies, the parent company is mainly financed by debt capital instead of being financed by its own capital. Usually the national legislations try to combat these situations by enacting provisions against thin capitalization, in order to avoid that the own capital of a company is consistently less than the borrowed capital. Apart from the tax advantage,, such situations of thin capitalization, are also attractive from a civil and commercial law viewpoint, as the amounts financed by the parent-company cannot be seized by the creditors.

According to thin capitalisation provisions, the above described situation is regarded as the concealment of the contribution of own capital and can be considered an abuse by the national tax administration. In such a circumstance the payment of interest could, from a legal point of view, be treated as a hidden distribution of dividends. In addition, from the tax point of view, interest paid in the presence of such financing operations could be considered excessive if the companies/persons involved are not completely independent. Their deductibility could then be challenged by the national tax administration, at least on the limits in which they exceed the amount usually paid for interest in similar operations by companies which are not involved in parent-subsidiary relations.

The Court of Justice, in its case-law, has clarified the limits and the conditions concerning national anti-abuse provisions especially aimed to limit the deductibility of interest paid to non-resident companies. In general, an anti-abuse tax provision must respect the limits laid out by the Treaty FEU and must not limit the freedom of establishment and the free circulation of capital ^[17]. The Court of Justice dealt explicitly with these issues in the Lankhorst-Hohorst GmbH ^[18] case . The interest paid by a resident subsidiary as a remuneration of a debt from a non-resident holding company was automatically subject to the taxation of the national tax administration (German) as hidden dividends. Whereas, if it were a subsidiary resident and the holding company (resident as well) could benefit from the tax credit, the interest paid would have been considered as current expenses and not as hidden dividends ^[19].

The Court of Justice considered that a differentiated treatment was made on the basis of the location of the head office of the holding company. In the case of a subsidiary resident receiving a loan from a non-resident holding company, the characterization of interest paid as a distribution of dissimulated net profit is the direct consequence of the application of the national anti-abuse provision. This disparity of treatment between the subsidiary in relation to the head office of their holding company was recognised as a restriction of the freedom of establishment, prohibited by the EU law. In this case the tax legislation had an impact on the exercise of freedom of establishment by companies located in other member states, which would have consequently renounced the acquiring, creating or maintaining of a subsidiary in another member state. On the basis of these provisions, laying down an equal treatment for loans to resident companies in the EU, different member states revised the national legislation regarding thin capitalization ^[20].

Furthermore, the Court of Justice settled other limits regarding the national prohibitions of deducting interest in cross- border situations to counter thin capitalization and profit shifting. For example, in the judgement Test Claimants in the Thin Cap Group Litigation ^[21], the Court of Justice confronted the case in which a national legal system (such as the United Kingdom) imposed restrictions on the possibility for a borrowing resident company in a member state to deduct, for tax purposes, interests on loans granted to a subsidiary company. Such restrictions could directly or indirectly be verified and imposed upon the resident companies in another member state. If, instead, the parent company had an office in the state of the borrowing company, it would not be immediately subject to restrictions.

The Court of Justice confirmed that the freedom of establishment can be jeopardised by the disparity of treatment between the borrowing resident companies depending on the location of the corporate seat of the money-lending subsidiary. The national legislation can be considered as restricting the freedom of establishment, if it restricts the freedom in a member state with regard to companies located in another member state. On the contrary, it is not necessary to demonstrate that the provision has the genuine effect of determining some companies to renounce the acquisition, creation or maintenance of a subsidiary in the first member state. On the other hand, the Court reaffirmed that in order to justify the limitation of an anti-abuse provision (that limits one of the Treaty freedoms) the fact that a resident company is granted a loan by a related company which is established in another Member State is not sufficient ^[22]. Conversely, a national measure restricting freedom of establishment may be justified on the ground of prevention of abusive practices where it specifically targets wholly artificial arrangements which do not reflect economic reality and are designed to escape the tax normally due on the profits generated by activities carried out on national territory. ^[23]

Another example of the effects of the Court’s jurisprudence is the judgement Lammers & Van Cleeff BV ^[24]. The Court of Justice handled the case in which in the Belgian resident subsidiary had paid the interest to a parent company resident in another Member State vis a vis a credit previously granted. The tax administration (Belgian), in accordance with the internal revenue code had qualified a part of the interest as dividend and it had therefore subjected it to taxation. However, on the basis of the national legislation, interest payments made by a resident company to a director which was a resident company could not be reclassified as dividends, and thus they were not taxable. Notwithstanding, if they were paid to a director which was a foreign company, such interests, if considered excessive, could be recharacterized as dividends and were taxable.

These provisions created a disparity of treatment between resident companies according to the place of establishment of the company which, as director, had granted them a loan. Therefore, it appeared to be a restriction of the freedom of establishment because they made it less attractive for companies establishing in other Member States to exercise that freedom and they may, as a consequence, refrain from managing a company in the Member State which enacted that measure, or even refrain from acquiring, creating or maintaining a subsidiary in that Member State ^[25]. The Court of Justice maintained that the fact that a resident company is granted a loan by a related company which is established in another Member State cannot be the basis of a general presumption of abusive practices and justify a measure which compromises the exercise of a fundamental freedom guaranteed by the Treaty ^[26]. Consequently, even if the application of such limits aims to fight abusive practices, the provisions exceeds the limits set out by EU law.

This case-law of the ECJ is fundamental because it clarified the room left to the member states to introduce anti-abuse measures. On one hand, it settled some important limits on the automatic recognition or on an absolute presumption of the abuse, as, for example, mechanisms of under-capitalisation or dissimulated capitalization ^[27]. In this framework, the mere circumstance that a resident company creates a secondary establishment, for example a subsidiary, in another member state does not automatically determine a general presumption of tax fraud, nor justifies ex se a measure that jeopardises the exercising of a fundamental liberty guaranteed by the treaty ^[28].

On the other hand, the jurisprudence of the Court of Justice clarified that the anti-abuse practices must always be based on legal reasons. In particular, the reduction of tax revenue does not constitute a necessary motive of general interest that can justify a measure in line with principles in opposition to a fundamental freedom ^[29]. In the same way, a possible tax relief resulting from the low taxation to which a subsidiary established in a member state is subject, different from the tax relief of a parent company in another Member State, cannot inherently allow this latter State to reserve (in exchange) a less favourable tax treatment for the parent company ^[30].

5. Recent developments of European law: from abusive practices to the balanced allocationof taxation power between the member states

However, all anti-abuse provisions are not automatically in conflict with the freedom of establishment ^[31]. On one hand, the Court considered that the national provisions restricting the freedom of establishment are compatible with the Treaty only if they specifically target the artificial arrangements designed to circumvent the legislation of the Member State concerned ^[32]. This freedom involves an effective settlement of the company in the member state where it is established and the exercising of a real economic activity ^[33]. Wholly artificial arrangements can therefore justify anti-abusive tax provision if they do not reflect economic reality and are designed to escape the tax normally due on the profits generated by activities carried out on national territory ^[34].

On the other hand, the Court of Justice has admitted the possibility for the national legislations to provide rules aim at preventing tax evasion, even outside cases of abusive practices. The Court admitted that the freedom of movement cannot grant companies the free choice of the State in which they can carry out the taxation of net profits. For example, the Court considered that the presence of different tax rates in the member states applicable to different companies of the same group are, however, able to justify an anti-abuse clause ^[35]. These situations can, according to the Court, justify the presumption that a financing subsidiary was only established in order to organise a transfer of net profits so as to benefit from a lower tax rate. These kinds of provisions were therefore considered acceptable according to Court’s case-law. At the same time, it does not exclude that the rules adopted by a Member State, to specifically regulate the situation of cross-border group, can in some cases represent a restriction of the freedom of establishment of the interested companies.

The judgement Oy AA, concerned the transfer of losses within a group of companies. If an intra-group cross-border transfer was deductible from the taxable income of the transferor, it would result in allowing groups of companies to freely choose the Member State in which the profits of the subsidiary are to be taxed. Therefore, in such a case, the allocation of taxing powers between member states would be jeopardised. Depending on the choice made by the group of companies, the member state of residence of the subsidiary would be obliged to abandon its right to tax the income of the latter to the potential advantage of the member state of the parent company ^[36]. The Court dealt with the issue of the eligibility of a different tax treatment in the two hypotheses. In the first hypothesis, the transfer, conducted by a subsidiary in favour of a parent-company that had the corporate seat in one of the ember states (Finland) was considered as an intra-group financial transfer under such laws, deductible from the taxable income of the subsidiary. In the second hypothesis, in case of a transfer realized by a subsidiary in favour of a parent company where the corporate office was in a different member state than Finland, it could not be considered as such and therefore could not be deducted from the taxableincome of the subsidiary. In other words, the subsidiaries of foreign parent companies thus receive less favourable tax treatment than the treatment enjoyed by the subsidiaries of Finnish parent companies. The Court of Justice affirmed the compatibility of the national legislation with the principles of the Treaty based on two aspects: the tax power of the states and the prevention of tax evasion. Regarding the latter aspect, the possibility to transfer the taxable income of a subsidiary to a parent company that has the corporate office in another member state involves the risk that, via wholly artificial arrangements, transfers of income will be organised within the group of companies. These transfers would occur in the direction of the companies where the corporate office is found in the member states that apply lower tax rates of taxation or in member states in which such income could not be taxed at all ^[37].

In light of the joint consideration of safeguarding the balanced division of tax powers amongst the member states and preventing tax evasion, the Court recognised that the national system pursued objectives that were legitimate and compatible with the Treaty, as referable to overriding reasons of public interest. Such reasoning was confirmed in various judgements including, the 2010 judgements X Holding ^[38]and Société de gestion industrielle ^[39].

It may be inferred from this case-law that EU law does not grant EU protection for operations aimed at, for example, the transfer of net profits – such as in the form of interest – towards other companies established in member states that apply low tax rates. The balanced allocation of ta xing powers between the Member States ^[40]is therefore protected, also beyond situations of evident abuse of rights.

These new jurisprudential orientations – highly objectionable because they constitute a step backwards in the protection of the taxpayer in respect to previous jurisprudence - tend to render the via debt less interesting. Member States are indeed allowed to deny the taxpayer the application of the tax regime that is normally provided for interest, namely the deductibility of the interests.

6. International double taxation, internal market and neutrality

In addition to the measures designed to fight against evasion, the tax regimes of financing activities of European companies are strongly influenced by the existence (or by the elimination) of double taxation.

With regard to interest and dividends, the taxing powers of the states are generally concurring. International double taxation generally occurs if the internal rules of the two different states are jointly applicable to the same taxable income. Situations of double taxation can arise in the presence of a concurring power of the state of residence and the state of the source, of two states of the source or two states of the residence ^[41].

The unilateral identification of taxing power might be linked to the lack of t ax coordination between Member States. If there are conventional tax powers, there is less risk of double taxation. Nevertheless, even the international conventions, through a system of provisions regulating state taxation powers, cannot eliminate completely the double taxation risk. Moreover, depending on International Conventions, conflicts of categorization are generally resolved through the interpretation of the law on a case-by-case basis.

The taxation of dividends is linked to the fact that the company profits are already taxed. The problem regarding whether the second taxation i.e. shareholder should be avoided is tightly linked to the form of company taxation. However, the two taxations – company’s income and shareholder – are object of a regular taxation on a European and international level. The Community legislation has considered different mechanism such as tax credits and dividend exemptions ^[42].

Tax systems do not usually consider if the revenues related to distributed dividends have already been taxed, at the time of the tax declaration, prior to the distribution. Therefore, corporate financing via equity capital is less advantageous due to this economic double taxation. In order to eliminate such problems, the member states might provide a mechanism of tax credit or dividend exemption.

There is a clash between the existence of an international double taxation of dividends and interests and the two principles of tax neutrality and existence of the European internal market. On the basis of the neutrality principle, the tax system should not prefer the company financing via risk or debt capital. As a consequence, a national tax policy opting to enhance one of these two forms of corporate financing should automatically be in contrast to the criterion of neutrality ^[43].

Most of the time the neutrality in the area of European taxation is often connected, at least partially, to the double taxation. There are examples of tax neutrality criteria in various EU directives. Regarding indirect taxations, the neutrality of the VAT is implemented through the tax deduction and is present in the definition of value added tax (VAT). The latter is able to guarantee the perfect neutrality of taxation of economic activity subject to this regime ^[44]. This neutrality is also verified in trade between member states, via a harmonised determination of both categories of taxable operation and criteria of localisation. This harmonization implies the near-complete elimination of any double taxation situation ^[45]. On the topic of direct tax, the criterion of tax neutrality can be found in the Directive 90/434/EEC ^[46]. The latter states that the operations of mergers, divisions, transfers of assets and exchanges of shares concerning companies of different Member States must not be hindered by restrictions, disadvantages and distortions derived from tax rules of member states. The taxations cannot prevent decisions on the restructuring or to the reorganization of the company ^[47].

The idea that the pursuit of neutrality can concretely enhance the free circulation of capital within Europe, through the elimination or at least the reduction of double taxation at European level, also characterises the EU directives related to the direct taxation that have been progressively implemented by the member states.

The Council Directives of 23 July 1990 90/435/EEC was amended by the Directive 2003/123/ EC ^[48]on the common system of taxation applicable in the case of parent companies and subsidiaries of different Member States. This directive alo ng with the Directive 2003/49/EC ^[49]on a common system of taxation applicable to interest and royalty payments made between associated companies of different Member States were laid down by the European Council.

The Directive 2003/48/EC ^[50] of 3 June 2003 introduced a uniform taxation of savings income in the form of interest payments and it has had a great impact. Although the authority of the member states to tax the interest at the source was not involved, the exchange of information between the state at the source and the state of residence should be logically accompanied by the abolition of taxation at the source of the interest, thus eliminating corporate double taxation. Nevertheless, until now there have been no common European positions regarding this issue ^[51].

7. The double taxation, the repeal of Article 293 CE and the case-law of the Court of Justice

Currently the neutrality (and as a consequence, the prohibition of the double taxation), has not been not recognised, beyond the specific application already mentioned, as a real and true general principle of European law.

a) The double taxation and the repeal of Article 293 CE

The need to regulate the relations between member states stipulating conventions that would prevent a double taxation was indicated in Article 220 of the CEE Treaty ^[52]. It is necessary in order to eliminate or reduce the negative effects on internal market operations, which are consequence of the co-existence of national tax systems. However, article 220 EEC (later on 293 EC) had the sole purpose of encouraging the member states to establish actions that aimed to reduce or abolish the double taxation. Some scholars argued that such a provision did not foresee the abolition of the double taxation. This interpretation was emphasised by the fact that Article 293 of the Treaty was not self-executing and did not lay down any direct obligations or exclusive responsibilities of the EU institutions. After the Treaty of Lisbon came into force, the general obligation of the member states to abolish the international double taxation in the non-harmonised area of the direct taxation remained largely unchanged. The obligation previously laid down by the Article 293, however, can still be considered existent on the basis of the loyalty principle (Article 10 of the Treaty) and by the objectives of the internal market. Therefore, after the repeal of Article 293 the obligation of the member states regarding the elimination of the double taxation is not always exempt from potential ambiguities ^[53]and can thus present problems in its application ^[54].

In order to understand the importance of such obligations it is necessary to verify the conclusions reached by the jurisprudential interpretations of the Court of Justice before the recent modifications of the EU legislation.

Regarding direct taxations, the Court of Justice had previously considered the freedoms of the Treaty primarily as provisions prohibiting discrimination towards citizens of other Member States ^[55]. Subsequently, the case-law decisions of the Court of Justice changed. In some judgements, it stated that the Treaty’s provisions not only prohibit the discrimination of non-resident citizens but it also forbids all other restrictions placed on both residents and non-residents that are involved in the freedom of circulation of goods within the EU ^[56]. Even though, on the basis of the latter orientation, discrimination towards non-residents regarding tax credit or exemptions is not acceptable, it must be considered that the national tax systems are not harmonised in the area of direct taxation ^[57]. The Court of Justice could therefore affirm that the freedoms guaranteed by the Treaty were not able to prevent an international double taxation ^[58]. This is due to the fact that EU rules do not establish general criteria for the division of responsibilities between member states that relate to the elimination of the double taxation within the EU ^[59]/^[60].

Despite the Treaty of Rome’s objective to abolish double taxation between the Member States ^[61], the Court of Justice recognised that in the absence of EU rules of unification or of harmonization (in particular under the repeal of Article 293, second part of the EC Treaty) the member states would be capable of defining the criteria to share their tax power, especially in order to eliminate double taxation, on the basis of international agreements or internal law ^[62].

b) Double taxation of dividends and bilateral conventions

Via the judgement Kerckhaert-Morres ^[63], the Court of Justice clarified whether the general principle of free circulation of capital could clash with the legislation of a member state regarding income tax and taxation of dividends. In this case-law, the national legislation stated that the taxation of dividends distributed by companies established in the same member state or by companies established in other member states depended on a single tax rate. The national legislation (Belgian), therefore, did not acknowledge the deduction of the tax at the source level applied in the other member state regarding dividends distributed by the external company ^[64]. The remuneration of the direct investment carried out by natural persons in a company established in another member state were subject to a higher taxation (for example, 15 percent in France and 25 percent in Belgium, applied to the net dividend) compared to the income related to the participation in a Belgian company.

Prima facie, the Belgian tax legislation did not generate any discrimination between shareholders resident in Belgium receiving dividends from a company established in Belgium or from a company established in another Member State. In both cases the Belgian law applied a uniform rate of income tax equal to 25% as a tax claim on income. However, taking into consideration the absence of a specific prevision that would have allowed a different treatment for diverging situations – considering that foreign dividends were also taxed at the source at a rate of 15 percent (in the state of residence of the EU company), whereas dividends distributed by Belgian-residing companies were not in the same situation – the creation of a double taxation effect was inevitable.

However, the Court of Justice stated that the negative consequences of the Belgian tax legislation were caused by the existence of common areas of tax competence between the two Member States. Moreover, these consequences were not caused by the infringement of the economic freedom protected by the Treaty. Therefore, the ECJ clarified that Member States were responsible for taking the measures necessary to prevent such situations by applying, in particular, the apportionment criteria followed in international tax practice.

The repeal of Article 293 of the Treaty did not concern the pre-existing acquis commnautaire, involving many case-law, including the judgement Kerckhaert-Morres ^[65]. On the basis of this judgement the member states are required to create measures limiting the double taxation through ad hoc tax conventions. At the same time, member states are requested implementing the provision laid down in the Article 293 (now repealed). Because of the repeal of this article it is necessary to clarify, on a case-by-case basis, whether a treaty is necessary. The Court of Justice’s case-law was endorsed by theDamseaux ^[66] case . The member state of residence is not responsible for preventing the double taxation between the member state of the source of the dividends and the member state of residence of the shareholder. In such a situation the member state of the source would have the priority in the taxation of the income. On one hand, this share responsibility would comply with the international legal procedures and, above all, with the OECD model. However, at the same time, the Court of Justice stated that the EU law does not set out general criteria concerning the qualification of areas of competence between the Member States, in relation to the elimination of double taxation within the EU. Therefore, the Court reached the following conclusion: firstly a member state cannot rely on a bilateral convention to avoid looming obligations in force by the Treaty ^[67]. Subsequently, both the member state in which the dividends are paid and the member states in which the shareholders reside are responsible to tax those dividends. However, this does not imply that the member state of residence is obliged, by EU law, to prevent the disadvantages which could arise from the exercise of power endorsed by the two Member States.

c) Double taxation and withholding tax: comparison between interest and dividends

Regarding interest, Truck Center ^[68] is arguably the most important case ^[69]. The Court of Justice resolved a situation that, for several reasons, was similar to the case Kerckhaert-Morres despite the fact that the payment was related to interests and not dividends ^[70]. This case dealt with the verification of the compatibility of the EC Treaty with a member state legislation (Belgium) providing the retention of tax at source on interest paid by a company resident in that Member State to a recipient company resident in another Member State, while exempting interest paid to a recipient company resident in the first Member State from that retention.

The Court of Justice considered that the freedom of establishment was not an obstacle to this discrepancy because the differences of treatment (a consequence of the Belgian tax legislation) were not comparable. There was no similar situation foreseen by the national legislation concerning beneficiary companies of capital income with different tax methods according to their residence either in Belgium or in another member state.

According to the Court, the difference in treatment between companies receiving income from capital, established by the tax legislation consisting in the application of different taxation arrangements to companies established in a certain member state, and companies established in another Member State, relates to situations which are not objectively comparable. Therefore, it is necessary to differentiate the cases between the company that distributes the interest and the beneficiary company of this interest both reside in the same member state. or, rather, the cases in which the resident company in a specific member state pays interest to a non-resident company.

In the same way, there are different tax conditions depending on whether the payment of the interest is carried out by a resident company to another resident company, or whether this payment is carried out by a resident company to a non-resident company.

The possibility of a different treatment gives rise to a series of observations. In the first place, even if the withholding tax does not involve interest paid from a resident company to another resident company, normally (as happened in the Truck Center case) such interest is subject to the corporate tax of the same state.

In the second place, the resident beneficiary companies are controlled by the national tax administration. The latter can recover the tax through an enforcement procedure. Vice versa, this does not happen in the case of non-resident beneficiary companies as the recovery of the tax requires the assistance of the tax administration of their state of residence.

In the third place, the difference in treatment resulting from tax legislations does not necessarily affect the resident beneficiary companies, because they are usually obliged to pay an advance on the total amount of the corporate tax. Moreover, the rate of the withholding tax calculated on the interest paid to a non-resident company is consistently less than the rate of the corporate tax (calculated on the total income of the resident companies which receive the interest).

In similar situations related to the application of deductions at the source on dividends, the Court of Justice reached opposite conclusions. In the recent case Commission vs Italy ^[71], the Court of Justice confirms the previous cases law (such as Denkavit Internationaal ^[72], Amurta ^[73], and Aberdeen Property Investment ^[74]). The Court ruled on the Italian legislation providing a less favourable tax treatment for dividends distributed to companies established in other member states rather than in Italy. According to the ECJ it constitutes a prohibited restriction on the free circulation of capital.

The Italian legislation previously exempted from taxation the dividends that were distributed to resident companies, in the amount of 95%. It taxed the remaining 5% at the normal rate of corporation tax, which was 33%. Dividends distributed to companies established in other Member States were subject to a withholding at the source at the rate of 27%, four-ninths of the sum that was capable of being repaid on application. A withholding at source at a reduced rate may also be applied, by virtue of the provisions of the various conventions for preventing double taxation, whereby certain conditions regarding the size and duration of the holding were fulfilled. However, this rate remained higher than the rate imposed on dividends distributed to resident companies. Such a difference in treatment was likely to deter companies established in other Member States from making investments in Italy. It therefore constituted a restriction on the free movement of capital, prohibited in principle by Article 56(1) EC.

The Court examined whether the beneficiary companies of dividends that resided it Italy and those established in another member state were in similar situations. Independent from potential taxation in another member state, the tax responsibility of a member state creates the risk of multiple taxation or economic double taxation. The state of residence of the distributing company must monitor whether the non-resident shareholding companies are subject to a treatment equal to that given to the resident beneficiary shareholding company. In this way, the freedom of establishment is not restricted ^[75]. This monitoring action must be aimed at preventing or reducing multiple taxation or double taxation.

The Court observed that the Italian legislator chose to exercise its taxing power regarding dividends distributed to companies established in other member states. As a consequence, the non-resident beneficiaries of these dividends found themselves in a similar situation to residents. There was a risk of an economic double taxation of dividends distributed by resident companies, for which the non-resident beneficiaries cannot be treated differently from resident beneficiaries. For this reason Italian law was not considered as complying with the principle of free circulation of capital.

Comparing this decision with the Truck Center case ^[76] (concerning interest) it is possible to observe that the Court allowed a different treatment between dividends and interest in nearly-identical situations. Therefore it appears questionable, in certain cases, to accept justifications based on coherence of the tax system and at the same time reducing the impact of the double taxation of dividends, whereas in other cases the double taxation of interest (and even of dividends) is considered acceptable (as happened in the aforementioned judgements Kerckhaert-Morres ^[77] and Damseaux ^[78]).

8. Conclusion

The Court of Justice commenced dealing with the notions of interest and dividends focusing on the obstacles that limit the freedoms stated by the treaty, such as the national anti-abuse previsions and situations of international double taxation. It seems clear that the European case-law does not have a preference, not even implicitly, for one of these two forms of financing. The limits of the Union’s responsibility regarding direct taxes and the consequent absence of an effective European legislative policy in the EU directives place important restrictions on the current state, even considering the interpretative activity of the Court of Justice. It is possible to trace examples of such limits by considering that, regarding dividends and interests, the Court systematically referred to the freedom of the member states to freely choose criteria related to division of their tax powers. Thus, the prohibition of the cases of double taxation – not specifically foreseen by the EU directives – cannot be considered a general principle of EU law.

Moreover, on the basis of the analysis of EU law it is difficult to reconstruct a EU definition of interest and dividends. This creates a further obstacle to the enforcement of the internal market.

On one hand, the lack of a legislative definition of interest and dividends at the OECD level and consequently in Europe, creates the risk of a double taxation or of a lack of taxation. On the other hand, it is possible to observe that the thin capitalization is the field where the definitions of dividends and interests might overlap. In this domain important case-law were decided by the European Court of Justice. However, the ECJ in principle only verifies the reasons behind the measures, limiting the freedoms of the treaty at the national level. Moreover, the problem of the thin capitalization is not sufficiently tackled by EU law, mainly because of the Community concept of interest. Therefore, considering the above-mentioned observations, on the basis of the European law, a national legislation that does not consider any kind of interest as dividends, such as the Belgian one, cannot be consideredas contrasting with EU law. Rather, it has the final effect of not allowing the deduction of this interest.

Furthermore, EU law does not take any position in relation to the national legislations – such as the Belgian one – establishing notional interest systems. The Belgian system made it possible to mitigate the provisions contrasting the thin capitalization. Notional interest allows a deduction from the taxable income of an amount related to the own capital. Notional interest creates a tax appeal for the own capital, whereas the utilisation of debts and of interest deduction systems is reduced.

The absence of a unique typology of dividends and interest in EU law involves a real risk of competition among tax systems. The current trend in the national legal systems involves reducing the tax rates in order to improve tax competitivity. These reductions are normally financed via the enlargement of the tax base, which can also be obtained via the extension of anti-abuse measures or through a limit of the deductibility of certain costs, such as interest . Systems foreseeing notional interest pursue a similar objective, even though advantages are much more evident. They allow the exploitation of capital flow and the drawing of tax on the income resulting from the positive difference between notional interest and the income from interest paid to the company. The coexistence within the European Union of these different choices of tax policies enhances the tax competitivity among member states, in a framework characterised by the economic crisis. In particular, the existence of hybrid forms of corporate financing allows taxpayers to take advantage of the absence of harmonization, structuring their activities in order to maximising the different characterizations of interest and dividend used at the national legislation level.

De jure condendo the necessity to overcome the current differences in the definitions of dividends and interest among the national le gislations appears to be clear. The goal consists of the pursuit of a greater level of harmonization at the European level, among the tax legislations concerning the correct definition of these concepts. This harmonization should be achieved either via directives or through the interpretation of the European Court of Justice. The final objective of more integration of the national tax systems, according to the principle of the tax neutrality is unanimously recognised. At the same time, it is the only way to avoid the tax competition between member states, which could create major problems for the competitivity of the European economic system at the global level.

© Copyright Seast – All rights reserved