Cross-Border Loss Relief Jurisprudence

1. Introduction

Cross-border loss relief is an important part of economic group’s strategy when choosing the place to establish business which means that if the EU wants to keep up with the Lisbon Agenda it must make strong efforts to ensure the possibility of cross-border loss relief between Member-States becomes a reality. However, the present situation is far from ideal with Member-States not willing to accept the lost of tax sovereignty which comes with the implementation of a working intra-states loss relief system.

The first part of this paper will analyze the recent European Court of Justice (ECJ) jurisprudence in the area of cross-border loss relief between Member-States.

We will discuss the most important judgments in this area beginning with AMID concerning a Belgian head office with a permanent establishment (PE) in Luxembourg. The next judgment is Marks & Spencer which was the landmark in this area concerning a UK parent company with subsidiaries in France, Germany and Belgium.

We will then analyze the importance of the Oy AA judgment in deepening the Court’s understanding on allocation of taxing rights and risk of tax avoidance.

The second analysis will focus on Deutsche Shell judgment which made clear the difference between losses and operating expenses. Finally, We will analyze the recent Lidl Belgium judgment which for now brings a final standing point from the Court in this area.

In the final chapter we will give our opinion on the guidelines the ECJ is following in the area of direct taxation and we will talk about the balance achieved by the Court in the area of cross-border loss relief with its possible origins in the freedom of workers. Also, we will try to draw a conclusion on where Portugal currently stands and what steps it must take to avoid future problems with the ECJ and the incompatibility of its rules with the ECT.

2. Cross-border loss relief jurisprudence analysis

2.1 – AMID[1]

In AMID there was a Belgium head-office with a PE in Luxembourg. In 1981, the company incurred losses in Belgium but earned profits in Luxembourg. These profits were taxed in Luxembourg and were not subject to tax in Belgium according to the Double Taxation Treaty celebrated between the two Member-States.

In 1982, the head office made profits and applied for a deduction of the losses incurred during the previous year. However, a Belgium provision (Article 69 of the Royal Decree) denied the deduction of the losses if they were hitherto capable of being offset, or were previously covered by profits exempted by Treaty.

In this case the provision denied AMID the possibility of deducting losses incurred in Belgium from profits subsequently earned in Belgium on the ground that those losses should have been deducted from profits accrued in Luxembourg in 1981 which covered all of the losses incurred by the head office.

Because under the Luxembourg corporation tax system, it was not possible to offset the Belgian loss against the Luxembourg profit, AMID was denied the possibility of deducting their losses in either Member-State which led to legal action.

The Court’s analysis on comparability is interesting because it establishes that there is no objective difference with regards to the deduction of losses between a company with a Belgian PE and company with a foreign PE[2].

We can see the Court approaches comparability on a case by case analysis and does not use a general formula to determine if two taxpayers are in a comparable situation. The Court always considers the specific facts of the case and the type of area in which the provision under analysis applies. In this case it was loss relief.

Thus, even if Belgian companies with no PE abroad have whole of their income calculated globally and taxed at the rate applicable in Belgium and Belgian companies with foreign establishments are taxed in respect of the income of the PE in accordance with the tax provisions of the Member-State where the PE is situated[3], when looking at the specific area of loss relief these two companies (migrant and non-migrant)[4] are in a comparable position with no objective difference because both want to deduct their losses in following years profits.

The Court has already done the same with regard to dividends in ACT IV Group Litigation[5] and Manninen[6].

Having established that there was no objective difference between the migrant and non-migrant in these circumstances, the Court concluded:

“(...) the legislation at issue in the main proceedings limits the possibility of carrying forward losses incurred in that Member State during a previous tax period where, during that same tax period, those companies made profits in another Member State through the intermediary of a permanent establishment, whereas it would be possible to set off those losses if the establishments of those companies were situated exclusively in the Member State of origin”[7].

This case is concerned with an origin state case where a Belgium domestic rule is impacting on the way Belgian companies conduct their business. Thus, we need to apply the migrant/non-migrant test to check whether a company choosing to establish itself outside Belgium is disadvantaged when compared to a company which chooses to remain in Belgium. If the domestic rule is hindering the freedom of establishment of the company in another Member-State it will be incompatible with the EC Treaty unless it is justified and proportional.

The possibility of not being able to deduct losses from the head office if any profits were to arise in the PE in the same year is of major importance for a company when deciding if they should establish a PE in another Member-State. The impact of this domestic provision would discourage companies from moving to another Member-State. When trying to create a situation of cross border loss relief, in practice, this system went against the balance achieved by the two countries in the DTT where there was a complete separation of taxing rights. This was the reason why the Luxembourg authorities when asked to accept the losses incurred by the head office denied the request supporting their decision with the current allocation of taxing right negotiated in the DTT between the two States.

Luxembourg was not entitled to tax profits from a head office situated in Belgium and, thus, was not prepared to accept the losses incurred by the same head office according to the “two sides of the same coin” argument[8].

Hence, a domestic rule of one country (Belgium) was creating the problem. It was not a situation of disparity with regards to the tax rules of both Member-States, because on that level an agreement had been reached on the DTT. However the domestic rules introduced by Belgium disrupted that balance and created a problem for Belgium companies with foreign PEs[9].

And by disrupting the balance achieved by both Member-States, and in particular by introducing the obligation to deduct losses incurred by the Belgium head office in foreign PE’s profits, Belgium created a situation of final losses from an origin state perspective. Final because the losses incurred by the head office could not be deducted in 1981, in Luxembourg PE’s profits, since the Luxembourg tax authorities would not allow it and, in addition, it could not be deducted from the profits earned by the head office in 1982 because of the Belgian domestic provision.

Thus, the key point in this judgment was that the Belgian domestic rules led to the situation where the losses would be relieved in neither State when in fact the two Belgian companies were in a comparable position.

2.2 – MARKS & SPENCER[10]

Marks & Spencer[11] concerned a UK parent company with subsidiaries in France, Germany and Belgium which were opened to extend the retail business across Europe. Unfortunately, the business did not go as expected and the subsidiaries incurred large losses for Marks & Spencer. This led to the decision of closing down the subsidiaries by winding up the company or selling the facilities. However, the subsidiaries had accumulated large amounts of losses which, because the business was never profitable, had never been deducted in France, Germany or Belgium.

The UK parent tried to deduct the losses incurred by its subsidiaries cross border to the profits earned in the UK. The request to deduct the losses was denied on the ground that group relief could only be granted for losses recorded in the United Kingdom.

Marks & Spencer appealed against this decision and the High Court eventually decided to refer some questions to the ECJ for a preliminary ruling. In this judgment there was an origin state case where the a UK domestic rule denied Marks & Spencer the possibility of relieving losses incurred cross border by its subsidiaries. Applying the migrant/non-migrant test[12] we must check if the UK parent company which chose to exercise its freedom of establishment cross border is being treated less favourably compared to a company that chose to remain in the UK.

The conclusion is that if is we had a UK parent with a UK subsidiary, losses incurred by the subsidiary could be deducted from profits earned by the parent company using Section 402 and 403 of the ICTA concerning group relief for losses. However, because foreign subsidiaries cannot be part of the group the same advantage does exist for the migrant company situation. The possibility of deducting losses incurred by their subsidiaries constitutes an advantage given to non-migrant companies[13].

Therefore, when deciding whether to extend their business cross border, UK companies will be hindered to do so because of the disadvantage they will suffer with regard to the deduction of losses.

In practice, this led the Court to confirm the existence of a restriction on the freedom of establishment arising from different treatment for tax purposes to losses incurred by a UK subsidiary and losses incurred by a subsidiary resident in another Member-State which is prohibited by the ECT unless a proportional justification by imperative reasons of public interest exists. Within the justifications brought forward by the UK is the now known argument[14] of the preservation of the balanced allocation of taxing rights.

First, the argument states[15] profits and losses are two sides of the same coin and must be treated symmetrical in the same tax system in order to protect the balanced achieved between Member-States by way of the DTT[16]. This means Member-States have negotiated between them a balance on the allocation of the taxing rights concerning business profits through the existent DTTs. In this case, the UK had negotiated with France, Germany and Belgium that profits made by UK parent subsidiaries situated outside the UK were not subject to taxes in the UK. Thus, because the UK did not tax the profits arising in these subsidiaries it did not want to relieve the losses either. If the UK had to take the foreign losses into consideration it would disrupt the balance achieved between the Member-States.

Second[17], if the UK was obliged to give cross border loss relief there was a risk the losses would be taken into account twice, in the Member-State where the parent company was located and in the Member-State where the subsidiary was located. Thirdly[18], if companies were able to deduct losses in the country of their parent company there was a big risk of tax avoidance. This risk came from the fact companies would use this advantage to make “trafficking of losses”. This means companies might transfer their losses to countries where the tax rate was higher and, thus, where the tax value of those losses would be bigger.

In this particular case the Court chose to accept these three justifications taken together and decided the restrictive provision pursued legitimate objectives which were compatible with the ECT constituting overriding reasons in the public interest apt to ensure the attainment of said objectives. The ECJ felt that to give companies the option to have their losses taken into account in both Member-States would jeopardize the allocation of profits because the taxable basis would be increased in one Member-State and reduced in the other to the extent of the losses transferred. Also, the ECJ said it was legitimate for Member-States to be able to prevent the losses from being used twice and that a provision like the UK domestic provision was adequate to fulfill that purpose. Finally the Court also agreed with the third justification on the transferring of losses to countries with higher tax rates.

It is submitted that the second justification should not have been accepted. Notwithstanding the fact losses could be taken into account twice, Member-States have at their disposal instruments which would allow them to control if the losses had already been taken into account in the host state. Using the Directive 77/799/EEC on Mutual Assistance in the field of direct taxation[19], tax authorities would be able to confirm this situation either using the exchange on request in article 2 or the spontaneous exchange of information in article 4 of the Directive[20].

Hence, Member-States can use this Directive to check whether a loss is being used twice which means a domestic provision denying the possibility of group relief in general terms is not necessary for this purpose. Also, the ECJ has consistently used this argument to deny administrative difficulties arguments brought by Member-States.

Nevertheless, having concluded the UK domestic provisions were justified by reasons of imperative public interest the ECJ went on to analyze[21] whether the restrictive measure went beyond what was necessary to attain the objectives pursued according to the Kraus and Gebhard judgments[22].

This is the most important part of the judgment. After analyzing the possibility of less restrictive measures which could be adopted in substitution of the general exclusion rule for group relief, the Court reached the conclusion that the UK provisions on group relief went beyond what was necessary to attain the essential part of the objectives in the following situations:

“- the non-resident subsidiary has exhausted the possibilities available in its State of residence of having the losses taken into account for the accounting period concerned by the claim for relief and also for previous accounting periods, if necessary by transferring those losses to a third party or by offsetting the losses against the profits made by the subsidiary in previous periods, and

- there is no possibility for the foreign subsidiary’s losses to be taken into account in its State of residence for future periods either by the subsidiary itself or by a third party, in particular where the subsidiary has been sold to that third party”[23].

The ECJ considered that a domestic provision denying the deduction of cross border losses is disproportionate in situations where, for example, the time limit for the deduction of losses has expired and the subsidiary is still accumulating losses. In the first example the Court states that if the subsidiary is in a situation where the limit is surpassed and there is no possibility of transferring the losses to a third party or deducting the losses against previous profitable years the provisions are disproportionate.

The second example was the actual situation of the case. The subsidiaries had ceased trading or had been wound up[24] and there was no other subsidiaries of the group in those countries there was no possibility of deducting those losses which otherwise would be lost and could not be deducted in either Member-State[25].

This judgment was significant because it drew a difference between temporary losses and final losses. The first type of loss can still be deducted in following years of the subsidiary whether the losses are still within the time limit established by the domestic law or because the company is still trading and can still earn profits in the following years.

If any of these situations change the losses will become final due to the fact that in the Member-State where the subsidiary is located the possibility of deducting losses has been exhausted and the domestic rules of the other Member-State are denying the possibility of offsetting those losses cross-border. This means Marks & Spencer would not receive loss relief in either Member-State.

2.3 – Oy AA[26][27]

In this case there was a Finnish company, oy AA, owned indirectly through two other companies by a United Kingdom company (AA, Ltd). In 2003, the parent company incurred losses and the subsidiary in Finland earned profits. Alleging the business of the parent company was extremely important to the Finnish subsidiary oy AA wanted to make an intra-group transfer of profits to help its parent company deal with its financial problems.

The Finnish domestic rules allowed these intra-group transfers to be deducted from the taxable business income of the transferor company which would be regarded as a taxable business income of the transferee company instead.

Thus, an intra-group transfer would be taxed only once, it would be deducted from the taxable income of the company which provides that it is added to the taxable income of the recipient company. However, Finnish domestic rules only allowed the deduction if the intra-group transfers were between Finnish share companies. The purpose of the provision was to put groups consisting of parent and subsidiary companies in the same position as a firm which has a number of PEs.

In this situation we have a host state case. The rules in the host state (Finland) are granting different treatment to subsidiaries established in Finland depending on where their parent company is situated. If the subsidiary has a Finnish parent company the deduction would be allowed. However, the same does not happen if the parent company is situated in any other Member-State.

According to the national treatment principle[28], Member-States are obliged to grant subsidiaries owned by parent companies, established in other Member-States, the same treatment granted to national companies. In this case the residence is the criteria used to determine the nationality of the companies. However, this could only be considered discriminatory towards the foreign parent company if the purely domestic and the cross border situations are comparable, because if there is no comparability Finland could treat different cases differently according to the principle of non-discrimination established in Schumacker[29].

Therefore, the Court needed to establish if an intra-group financial transfer between a Finnish subsidiary and a Finnish parent was in- a comparable position to an intra-group financial transfer between a Finnish subsidiary and a UK parent. German, Netherlands, Swedish and United Kingdom Governments argued that these two situations were different because the foreign parent company is not subject to tax in the Member-State where the subsidiary is established. There is a distinction between subsidiaries with a parent company which is also taxed in Finland and subsidiaries with a parent company which is not subject to tax in Finland. In their opinion[30] this difference was relevant because the Finnish tax authorities could not be sure if the intra-group transfer would be subject to tax at the level of the transferee because the transferee would not be a subject to Finnish tax laws. Thus, a situation of double non-taxation could arise.

The Court once more made an analysis considering the purpose of the rule according to what was done concerning dividends in ACT IV and Manninen. This type of analysis considers what is the purpose of the provision under analysis and if the difference argued by the Member-State has any impact on that purpose.

The purpose of the provision as discussed above is to remove tax disadvantages inherent in the structure of a group of companies by allowing a balancing out within a group that comprises both profit-making and loss-making companies, promoting the interests of a group of companies. The conclusion reached by the Court was that a group comprising cross border companies is clearly interested in using this benefit and the final purpose of this rule could not be violated if extended cross border. Cross border groups suffer the same financial problems and the need to balance out the group may also arise in a cross border group.

Of course, the need for the income to be taxed at least once at the level of the transferee is important. However, simply denying the benefit to cross border situations in a type of “over kill” provision was not a proportional way of avoiding tax abuse. The ECJ suggests Finland could make the deduction subject to conditions concerning the treatment to be applied to the transfer by that other Member-State where the burden of proof would rest with the company applying for the benefit.

With concern to comparability the ECJ states that:

“in relation to the aim pursued by the Finnish system of intra-group financial transfers, the mere fact that parent companies which have their corporate establishment in another Member State are not subject to tax in Finland does not differentiate the subsidiaries of those parent companies from the subsidiaries of parent companies which have their establishment in Finland, and does not render the positions of those two categories of subsidiary incomparable[31]”.

We agree with the ECJ’s since only by considering the purpose of the rule we are better able to establish if there is a comparable situation. For example, in Lasertec[32] and Thin Cap GLO[33] the Court considered the purpose of the rules as rules concerning controlled companies to decide which freedom applied, freedom of capital or establishment. Also, in ACT IV GLO concerning UK rules granting credit for tax paid on dividends the Court considered the purpose of the rule, to avoid double taxation, to decide whether two companies were comparable.

In this case the need to treat a group of companies as a single entity for taxing purposes is the same whether in a cross border group or in purely domestic group. The same happens in dividends and provisions on double taxation relief, both domestic and a cross border distributions of dividends should be granted double taxation relief even if the dividends are ultimately taxed in different Member-States.

Therefore the Court found the Finnish provision constituted a restriction on freedom of establishment unless there was an overriding reason of public interest which was adequate and proportional. The justifications submitted by some Member-States included the cohesion of the tax system, the principle of territoriality and balance allocation of taxing rights. However this analysis will focus on the latter, since it is the most relevant issue for discussion in this paper.

The argument, taken from Marks & Spencer, stated that the Finnish rule was necessary to ensure the balance attained by Finland in the allocation of taxing rights. Allowing deductions on cross border intra-group financial transfers would ultimately result in the waiving of the Finland taxing right of profits arising in the territory. Since companies would get a deduction on the intra-group financial transfers to group companies cross border, Finland would never be able to tax those amounts which would be out of their tax jurisdiction permanently. In reality, by allowing deduction of cross border intra group financial transfer Finland would disturb the balance achieved by the Member-States in the DTTs.

This argument was accepted by the Court as it already arisen in Marks & Spencer because looking at the consequences the ECJ stated:

“to accept that an intra-group cross-border transfer, such as that at issue in the main proceedings, may be deducted from the taxable income of the transferor would result in allowing groups of companies to choose freely the Member State in which the profits of the subsidiary are to be taxed (…) That would undermine the system of the allocation of the power to tax between Member States because, according to the choice made by the group of companies, the Member State of the subsidiary would be forced to renounce its right, in its capacity as the State of residence of that subsidiary, to tax the profits of that subsidiary in favour, possibly, of the Member State in which the parent company has its establishment[34]”.

The second part of the justification concerning the possibility of double deduction of losses was denied by the Court simply because the case under analysis did not concern deductibility of losses but intra-group financial transfers.

The third part of the justification concerning the prevention of tax avoidance argued that if the Finnish Government allowed the deduction of cross-border intra-group financial transfers income transfers could be organized within a group of companies towards companies established in Member-States applying the lowest rates of taxation or in Member-States in which such income is not taxed.

The Court accepted that Member-States are allowed to prevent this risk by introducing provisions precluding the transfer of profits to a lower or zero rate jurisdiction especially because in this situation the Finnish system of intra-group financial transfers did not require the company transferring the funds to have incurred losses.

We agree with the ECJ on this point since notwithstanding the fact “wholly artificial arrangements”[35] could be denied the benefit, this would constitute an opportunity for Finnish companies to transfer their profits to lower tax jurisdictions and shift large amounts of profits which would normally be taxed in Finland to other Member-States. The intra-group transfer would be used only to avoid paying the Finnish tax rate on corporate profits.

This was also the reason why the ECJ considered the Finnish provision to be proportionate. Any other situation would severely undermine the balance of the allocation of taxing rights negotiated by Member-States and open way to tax avoidance opportunities. In this judgment the ECJ accepted two of the three justifications taken together in Marks & Spencer and extended his understanding to an area outside losses confirming the flexibility of these justifications which can be used by Member-States in several areas of direct taxation.

We could also conclude that the ECJ is not prepared to allow the transfer of profits between Member-States allowing the shift of profits to lower tax jurisdictions and gravely interfere with the sovereign power of a Member-State to impose tax to their residents, unless a final situation of double non deduction arises from the specific facts of the case.

2.4 – DEUTSCHE SHELL[36]

Deutsche Shell[37] was a case concerning a German parent company with an Italian PE. In 1974, Deutsche Shell set up a branch in Italy. For this purpose the head office provided the PE with start-up capital accounted for in the two separate accounts, one in Germany (DEM) and one in Italy (LIT).

During the life of the branch this start-up capital was partially repaid through the repatriation of profits which were being converted from LIT to DEM.  On 28 February 1992 Deutsche Shell transferred the assets of the branch to a wholly-owned subsidiary, Sierra Gas SrL., and closed down the branch. Deutsche Shell then sold the shares acquired through the transfer to an independent Italian company, Edison Gas SpA.

Because the transaction was carried out in LIT, at the time of the repatriation of the proceeds of the sale, the sums were converted back into DEM. Part of that amount was destined to repay the outstanding amount of the start-up capital provided by Deutsche Shell to the Italian branch.

After all the start-up capital was repaid and this sum was set against the historic acquisition costs of the start-up capital, Deutsche Shell alleged they had suffered a currency loss of DEM 122 698 502 arising from the devaluation of the LIT against the DEM.

Thus, Deutsche Shell granted the Italian branch a start-up capital sum in LIT which generated a cost in DEM for the head office. The start-up capital was later repaid in LIT to the head office and had to be converted back into DEM at the time of the repayment. Due to the consistent devaluation of the LIT against the DEM created a currency loss since the start-up amount granted from 1974 was worth much less DEM in 1992 at the time of repayment. Deutsche Shell tried to deduct this currency loss in Germany but the tax authorities refused the deduction alleging the depreciation in the value of the start-up capital was merely part of the PE’s profits which were not subject to tax in Germany and even considering the depreciation Deutsche Shell had achieved a positive result in the financial year concerned.

Thus, because the DTT between Germany and Italy exempted profits made by a PE established in Italy from German tax, the German tax authorities refused to take into consideration an expense which was under the taxing powers of Italy and was exempt in Germany.

This case is another origin state case where a German domestic rule is preventing a company from deducting an expense because of its cross-border nature. The migrant/non-migrant test is applied in a different way in this situation. This happens because in our opinion it is impossible to find a comparable purely domestic example. The problem arises from the use of two different currencies in two different countries since if we were to find a similar situation, German head office with a German PE, that problem would not arise since both entities use the same currency.

However, the test can be applied in a slightly different way. Because in a purely domestic situation these problems never arise, a migrant company is being discouraged to establish itself cross-border. The migrant company knows it will have to face further difficulties in the deducting expenses arising from these situations.

Thus, the German domestic rule renders less attractive the exercise of the freedom of establishment. [38] However, the truly important aspect of this case is the analysis of the Court on another type of definitive situation. With effect, the main problem generated by the German domestic provisions was the final situation it created for Deutsche Shell. By denying the possibility of deducting the operating expense the company could not deduct in the host state, German domestic rules were creating a final loss which could not be relieved in either Member-State.

In Italy there was no currency loss. The currency loss can only crystallize at the time of the conversion (into the currency of the Member-State where the company is established) of the money transferred to the head office. Thus, the problem of currency losses only arose when the LIT were converted into DEM.

In the Italian branch accounts which only worked with LIT, the amounts (start-up capital and payments of that capital) were exactly the same which meant no loss existed. However, in Germany because German law regarded currency losses arising from the operation of a foreign branch of a German company as concerning to the income from that branch and that income was exempt in Germany, the currency loss was refused creating a situation where this expense could not be taken into account neither by the branch nor by the parent company.

We can immediately see the parallel with the situation of final losses in Marks & Spencer. Although in this case we have an expense suffered by Deutsche Shell and not a loss situation as in the Marks & Spencer case, the definite situation of a final situation is the same. This helps establish the guidelines that drive the ECJ in the area of direct taxation.

Furthermore, it is interesting that in this situation the ECJ refused the justification submitted by the German government on the allocation of tax sovereignty. Germany alleged that by DTT the two Member-States decided to exempt PE’s profits located on the territory of the co-Contracting State from income tax, which would exclude the currency loss from being taken into account.

The ECJ refused the justification of the allocation of taxing rights stating the following:

“(…) the tax disadvantage concerned relates to a specific operational factor which is capable of being taken into consideration only by the German tax authorities. Although it is true that any Member State which has concluded a double taxation convention must implement it by applying its own tax law and thereby calculate the income attributable to a permanent establishment, it is unacceptable for a Member State to exclude from the basis of assessment of the principal establishment currency losses which, by their nature, can never be suffered by the permanent establishment”[39].

This is an excellent opportunity to analyze the reason why this argument was accepted in Marks & Spencer and Oy AA but not in Deutsche Shell. In our opinion the main reason is the type of transaction under analysis. In Marks & Spencer and Oy AA we had the transfer of profits or losses between parent companies and subsidiaries. However, in Deutsche Shell we had a business expense which should have affected the taxable income of the German company.

The difference between these two is simple. In this case we are not before the transfers of losses or profits from one Member-State to the other which could mean the waiving of a Member-State taxing rights on income generated in their territory. In Deutsche Shell there was a business expense suffered by a German company arising from a cross-border investment which was not being accepted because of its cross border nature.

In both Marks & Spencer and Oy AA the justification on the balance allocation of the taxing rights was accepted on the basis of at least the combination of two arguments: disruption of the balance achieved in the DTT and tax avoidance possibility. In this situation neither existed because we were not before the transfer of profits or losses between two entities but a operating expense suffered by the German company similarly to the situation in Rewe Zentralfinanz concerning a write down on the value of shares.