Vassilis Dafnomilis (Assistant Professor at the University of Amsterdam / Senior Manager Tax at PwC Netherlands Tax Knowledge Centre)
Francesco Semonella (Associate at Maisto e Associati, Italy)[1]
A palindrome is a word, phrase, or sentence that reads the same backward as forward. The term “palindrome” originates from the Greek words “palin” (meaning “again”) and “drome” (meaning “run”). An illustrative example is the word “Lexel”, which not only embodies the characteristics of a palindrome but also mirrors its effects: it can be read in both directions. Similarly, someone could argue that the concept of abuse can be perceived from different angles. This concept – following the CJEU judgment in Lexel[2] – has sparked considerable debate among academics and practitioners: are the terms and conditions of an intra-group loan or just its purpose that should be the determining factor in assessing whether an intra-group loan is abusive? This question might be addressed thoroughly in the pending case X BV[3], in which the CJEU is asked to clarify the position previously taken in Lexel, on whether and to what extent intra-group loans may be regarded as abusive when they are carried out on an arm’s length basis. So far, the Advocate General (AG) Emiliou has expressed his opinion on this controversial issue, arguing that the intention behind intra-group loans should be the key element in identifying abuse[4] and that, consequently Lexel should be overturned.
Long story short…
Long story short, when addressing the compatibility with EU law of the exception to interest deduction set forth in Article 10d of Chapter 24 of the Swedish law on income tax in the Lexel case, the CJEU has ruled that:
“56 It must be held that the exception may include within its scope transactions which are carried out at arm’s length and which, consequently, are not purely artificial or fictitious arrangements created with a view to escaping the tax normally due on the profits generated by activities carried out on national territory.” (emphasis added)
Overall, the CJEU found the Swedish interest limitation rule to be incompatible with EU law because it imposes an unjustified restriction to the freedom of establishment.
Slightly more than one year after Lexel in the pending X BV case, the CJEU was asked whether – in light of paragraph 56 of Lexel – the interest limitation rule set forth in Article 10a of the Dutch CITA is in line with EU law when it applies in cases where the terms and conditions of the loan or the interest paid under that loan are at arm’s length. Article 10a of the Dutch CITA is very similar to Article 10d of Chapter 24 of the Swedish law on income tax in the Lexel case. In his opinion, the AG acknowledges that what paragraph 56 of Lexel states is “neither vague nor ambiguous and, thus, open to different readings”[5] i.e., interest limitation rules that apply to arm’s length transactions are incompatible with EU law. However, he advises the CJEU to depart from this judgment.[6]
Further, and perhaps even more interestingly, in footnote no. 43 of his opinion the AG delves into the dilemma that the CJEU will face in this pending judgment X BV. Considering that the case is under review by a chamber of 5 judges, just like Lexel, the question arises on whether the Court, in its current composition, will be in the position to reconsider the verdict previously expressed in Lexel.
On this intriguing dilemma, the AG states that Lexel has already been implicitly overruled and that, therefore, “[t]he Court does not need the Grand Chamber’s authority to confirm that fact in the present case”.[7] This is essentially based on the argument that Lexel is not consistent with the notion of wholly artificial arrangement as outlined by the Grand Chamber of the CJEU both before and after such a judgment.
Nevertheless, the possibility for the CJEU to overturn a previous judgment without creating uncertainty, along with the arguments presented by the AG supporting the current chamber’s ability to do so, warrant a closer examination in our opinion.
Has Lexel been overturned?
In the AG’s view, Lexel has been implicitly superseded by judgments issued by the Grand Chamber, both preceding and subsequent to the Lexel judgment. Consequently, the present chamber of the CJEU is not obliged to seek the intervention of the Grand Chamber to overturn Lexel. However, the validity of this argument is questionable. To begin with, Lexel does represent, in our view, the most recent examination of abuse in the context of intra-group loans. There has been no subsequent CJEU judgment post-2021 that addresses the notion of abuse contrary to what the AG contends (without any reference to such post-2021 case law).
Moreover, from our perspective, judgments of the Grand Chamber predating Lexel that refer to the “wholly artificial arrangement” concept are not decisive in this analysis, as the issue at hand is whether Lexel has been implicitly invalidated by later cases, not earlier ones. In addition, after Lexel the CJEU has never ruled on the notion of abuse in the specific context of interest limitation rules (and one may wonder whether the general concept of wholly artificial arrangement must be nuanced or better clarified to take into account the peculiarities of abusive arrangements involving intra-group loans).[8] As a result, we remain unconvinced as to why the current five-judge chamber in Lexel should encounter no impediment in overturning the Lexel judgment.
How often has the CJEU overruled its previous judgments (explicitly or … implicitly)?
The question is whether such an overruling is possible in any scenario, given that the CJEU is not known for overturning its judgments readily, or to put it more precisely, so explicitly. Consider, for example, the Lidl Belgium case[9], which dealt with the issue of final losses and marked the first instance where the CJEU applied the concept of final losses to foreign permanent establishments. It could be argued that this case was implicitly overruled by the subsequent W AG case[10], in which the CJEU determined that a foreign permanent establishment (PE) cannot be equated with a domestic PE if the tax treaty between the EU Member States employs the object/base exemption method. This was also the case in Lidl Belgium, but at that time, the Court overlooked this aspect and instead assessed whether the losses were final, an analysis that inherently assumes that a domestic PE is objectively comparable to a foreign one. Notably, the CJEU did not explicitly state whether its judgment in the W AG case overruled Lidl Belgium, but it seems to be the case —from our perspective—an implicit overruling. Conversely, the CJEU did take into account that the facts of the case differed from those in Bevola[11], where the base exemption was derived from national law rather than a tax treaty. Here, one might contend that the CJEU nuanced the position previously taken in Bevola without overturning it.
It is true, however, that in matters of direct taxation, the discourse often revolves around the clarification of newer case law in relation to older precedents (for instance, the X case elucidating Schumacker in tri-state situations)[12] or, as previously mentioned, W AG nuancing Bevola. Similarly, in Hirvonen[13], the CJEU was prompted to address the connection between this case and Gielen, as inquired by the national court. While Hirvonen was differentiated from Gielen[14], some scholars have suggested that Hirvonen appears to have overturned the “puzzling” Gielen case.
Given the history of the CJEU’s reluctance to explicitly overrule prior cases, it is imperative that the Court now makes an explicit reference to Lexel. In our opinion, the Court in its current composition should either affirm Lexel or refer the matter to the Grand Chamber, which should possess the authority to reverse Lexel. There should be no middle ground in this matter, as it would compromise legal certainty. Furthermore, a mere “clarification” of X BV in relation to Lexel is not advisable, especially considering that:
- the Dutch 10a rule, which serves as the foundation for the Swedish 10d rule (found incompatible in Lexel), is at stake; and
- we would end up with two judgments having equal standing and providing different views on substantially the same issue, unless the reasons for departing from the previous judgment in Lexel are explicitly and convincingly dealt with in this new decision (which, as mentioned, seems unlikely in light of the case law of the CJEU).
From this angle, if the Court decides not to refer the case to the Grand Chamber and for legal certainty purposes, it would be better that in X BV the Court rules consistently with its decision in Lexel and provides further explanation on why Lexel must be deemed consistent with the notion of wholly artificial arrangement as depicted by the case law of the CJEU. This may also shed a light on why the Court in Lexel decided without an opinion of an AG. In our view there might be room to argue that the arm’s length test should be the decisive proof of abuse when dealing with abusive techniques aimed at achieving increased interest deduction, considering that the reference in the anti-avoidance rule to the purpose of the loan or of the overall transactions revolving around the loan may sometimes be too vague.
Is a referral of the X BV case to the Grand Chamber possible?
The intervention of the Grand Chamber may be sought pursuant to Article 60(3) of the Rules of Procedure (RoP) of the CJEU which provide that “[t]he formation to which a case has been assigned may, at any stage of the proceedings, request the Court to assign the case to a formation composed of a greater number of Judge”.[15] As also mentioned in literature, an attribution to a particular chamber may be changed at any stage “even after the Advocate General has handed down his opinion”[16], and a reference back to the Grand Chamber in the present case would also be consistent with the case law of the CJEU.[17]
A Grand Chamber judgment
Leaving the case to the Grand Chamber would give the CJEU the opportunity to clarify how to shape national interest limitation rules to ensure their compatibility with EU law. Indeed, for national governments it is extremely difficult to tackle abusive debt transactions only aimed at achieving profit shifting and connected tax advantages without restricting too much the taxpayers’ freedom to decide the financial structure of their investments.
The Court’s judgment should address whether the arm’s length principle can be the decisive criterion through which determining whether financial arrangements are abusive, or otherwise stated, clearly address the requirements that must be met for an interest limitation rule to be deemed as a rule targeting abusive financing arrangements. To this end, the CJEU should outline possible indicators of abuse that an interest limitation rule shall contain (e.g., departure from the arm’s length principle, lack of substance of the financing entity, diversion of funds from the original borrower, etc.) and should clearly establish what is not to be regarded as an indicator of abuse (e.g., the circumstance that the contracting entity are related party, the existence of a tax advantage, etc.). Lastly, the Court should elaborate on how to ensure proportionality of the anti-avoidance measure, by describing situations in which the taxpayer can rebut the presumption of abuse and the role of the arm’s length principle in that rebuttal. On this specific aspect we share the view expressed in literature that while the arm’s length can be at certain instances a proportionality criterion, it is not a one-size-fits-all standard for all types of interest limitation provisions and appears to be particularly weak where the indicia of abuse are not determined by reference to features of the intra-group loan itself (such as interest rate, loan amount, characterization of the loan for transfer pricing purposes).[18]
Furthermore, to ensure consistency between primary and secondary EU law, in our view the judgment of the Court should not ignore the functioning of the ATAD GAAR (Article 6 of ATAD). For a debt transaction to be qualified as abusive under the ATAD GAAR, it is necessary that two tests are met: (i) the transaction must be regarded as non-genuine (i.e., artificial), and (ii) the transaction must give rise to a tax benefit that defeats the object and purpose of the applicable law. Because interest deduction should generally be considered legitimate in the light of the rules under which the corporate tax base is determined (as mentioned, taxpayers should be left free to decide the appropriate debt/equity ratio depending on the business needs), more clarity on how to perform the artificiality test is necessary and more emphasis could also be placed on the non-tax reasons that may justify the debt transactions. As mentioned in the AG opinion in the X BV case (paragraph 95), “courts should consider all valid economic reasons, including financial ones” (emphasis added), and financial reasons may frequently be crucial in practice, e.g., when external acquisitions are carried out at a central level through the main holding company of the multinational group and then the debt arising from the acquisition is subsequently re-allocated to the various entities of the group by means of intra-group re-leverage transactions (e.g., intra-group share deals financed by way of intercompany loans).
[1] Alumnus of the Adv. LL.M. in International Tax Law of the University of Amsterdam, Francesco was awarded the LL.M. in 2023 with a thesis entitled “The Lexel Case – Interest Limitation Rules, Arm’s Length Principle and Abuse” (supervised by Dr. Vassilis Dafnomilis and co-supervised by Prof. Dr. Daniël Smit).
[2] Lexel AB v Skatteverket, Case C-484/19, EU:C:2021:34.
[3] X BV, Case C-585/22, Request for a preliminary ruling from the Hoge Raad der Nederlanden (Netherlands) lodged on 7 September 2022 — X BV v Staatssecretaris van Financiën.
[4] Opinion of the AG Emiliou in X BV, Case C-585/22, paragraph no. 71.
[5] Opinion of the AG Emiliou in X BV, Case C-585/22, paragraph no. 70.
[6] Opinion of the AG Emiliou in X BV, Case C-585/22, paragraph no. 71.
[7] See the opinion of the AG Emiliou in X BV, Case C-585/22, footnote no. 43.
[8] See however PRA Group Europe AS vs Staten v/Skatteetaten, Case C—3/21 delivered by the EFTA Court on 1 June 2022.
[9] Lidl Belgium, Case C-414/06, EU:C:2008:278.
[10] W AG, Case C-538/20, EU:C:2022:717.
[11] Bevola and Jens W. Trock, Case C-650/16, EU:C:2018:424.
[12] Finanzamt Köln-Altstadt v Roland Schumacker, Case C-279/93, EU:C:1995:31; X v Staatssecretaris van Financiën, Case C-283/15, EU:C:2017:102.
[13] Hirvonen, Case C-632/13, EU:C:2015:765.
[14] F. Gielen v Staatssecretaris van Financiën, Case C-440/08, EU:C:2010:148.
[15] Rules of Procedure of the Court of Justice, available here.
[16] See Wägenbaur, B., Court of Justice of the European Union: Commentary on Statute and Rules of Procedure, C.H. Beck, 2013, p. 284.
[17] See Lindorfer v. Council, Case C-227/04 P, EU:C:2007:490, para. 40.
[18] See S. Buriak, Should the Court of Justice reconsider the Lexel decision?, in Kluwer Blog of 30 April 2024.
________________________
To make sure you do not miss out on regular updates from the Kluwer International Tax Blog, please subscribe here.
