At the request of the Commission, Fiat provided detailed financial data on the group treasury companies. In particular, Fiat provided the annual report for the years 2011 to 2013 of FFT, FF, FFNA, FFC and FFB. Fiat also provided a description of the functions of the group treasury companies. According to that information, FF has 52 employees, out of which [20-30] are working for the treasury department and [10-20] in accounting. FFT has 14 employees, composed of one director, [0-10] front office employees (the front office is based in the United Kingdom), [0-10] back office employees and [0-10] accounting and control employees. FFNA has five employees and FFC has limited activities.

According to the submission of Fiat, the book value of the subsidiaries of FFT (FFNA and FFC) […]. The annual report of FFT states that both FFNA and FFC have been acquired from Fiat SpA and FF in 2011. Starting from 2011, FFT prepared consolidated financial statements in Luxembourg.

FF's participation in affiliated undertakings at the end of 2013 (and at the end of 2012) consisted of a participation in FFT for EUR 157 269 000 (representing 60 % of the share capital of FFT), in FFB for EUR 2 013 000 and non-controlling interests in other group companies for EUR 725 000. At the end of 2011, 100 % of the shares of FFT was recorded in the participations of FF for EUR 262 102 000. At the end of 2010, before the sale of FFNA and FFC to FFT in 2011, the participations of FF consisted of EUR 7 213 000 in FFC, EUR 262 077 000 in FFT, EUR 87 055 in FFNA and EUR 2 013 000 in FFB, totalling EUR 358 million of participations in fully-controlled Fiat subsidiaries in 2010, as reported in the Table 8.

The information submitted by Fiat shows the maturity of the intra-group financing transactions. At the end of 2013, EUR 12 613 000 out of the total of EUR 12 858 000 of receivables (presenting the cash and loans to group companies in Table 9) had a contractual maturity shorter than one year.

The information provided on individual intra-group transactions shows that many transactions are overnight. However, FFT also concludes transactions with different maturities […]. FFT provides loans to the group in different forms […]. Almost all deposits had a maturity of […] at the end of 2013. The notes issues had different maturities, […].

Fiat also provided the Group Liquidity Policy document referred to in the transfer pricing report underlying the ruling request. The document sets out internal rules for investments of cash by the group. […]

In its letter of 20 February 2015, the Commission requested Fiat to explain which mechanism was used to achieve a net result equal to a stable proportion of equity over 2009, 2010 and 2011 (as is apparent from Table 4 in recital 51 of the Opening Decision), despite important variation in assets, liabilities and financial charges and revenues.

Fiat provided a document titled ‘Transfer Pricing Policy’ in response to that request to explain the pricing of intra-group loans and deposits. The Transfer Pricing Policy document clarifies that Fiat sets the prices of intra-group loans provided by the group treasury companies in such a way so as to achieve a pre-determined return for the treasury companies.

The functions described as performed by FFT and the risk assumed are identical to the description in the transfer pricing report, as reproduced in recitals 38 to 51, with the exception of some information. […] the Transfer Pricing Policy document relates to the credit risk and counterparty risk, which is described in the ruling request as non-existent concerning group assets, whereas the document qualifies this risk as ‘limited’.

The Transfer Pricing Policy document explains that for estimating the pricing of intra-group transactions the following steps are taken. […] the amount of the intra-group loans to be financed is estimated. […], on that basis a margin is estimated to be applied to intra-group loans by dividing the sum of the target return to which the operating costs are added, by the total loans to be financed. Finally, this margin is added to the costs of funding of the treasury companies to obtain an estimate of the intra-group loan pricing.

The document further distinguishes between pre-defined term loans, established in the past, for which it is not possible to change pricing conditions and floating revisable intra-group loans. […]

As a general rule, for the purposes of the selectivity analysis a reference system is composed of a consistent set of rules that apply on the basis of objective criteria to all undertakings falling within its scope as defined by its objective.

While the determination of taxable profits in the case of non-integrated/domestic standalone companies that transact on the market is rather straightforward, as it is based on the difference between income and costs in a competitive market, the determination of taxable profits in the case of integrated group companies like FFT, requires the use of proxies. Standalone, non-integrated companies can take their accounting profits as a starting point for determining the tax base to which corporate income tax applies, since those profits are dependent on prices dictated by the market for the inputs acquired and the products and services sold by the company. By contrast, an integrated company that transacts with companies of the same corporate group will have to estimate the prices applied to those intra-group transactions for determining their taxable profits for tax purposes, that estimate being determined by the same company controlling the group instead of being dictated by the market.

Finally, in response to the various arguments FFT advances to support its claim that group companies and stand-alone companies do not belong to the same reference system, the Commission notes the following.

In any event, contrary to what FFT claims, neither decision confirms that where a tax measure is granted in favour of an integrated company, the reference system must necessarily be limited to those types of company. Moreover, the objective of the tax measure at the basis of both decisions is not comparable to the present case and therefore, the conclusions that can be drawn from those decisions are not applicable to the present case.

By contrast, the objective of the contested tax ruling is to determine FFT's tax base to calculate the tax due for the purposes of levying the Luxembourg corporate income tax on that amount. First, while it could be argued that the objective underlying the Groepsrentebox decision is only valid in a group context (such as the fact that stand-alone companies are not faced with the issue of arbitrage between different forms of financing), the determination of the taxable base for the computation of the annual corporate income tax liability is equally relevant and applicable to entities that are part of a group as well as stand-alone companies.

The Commission therefore concludes that the reference system against which the contested ruling should be examined is the general Luxembourg corporate tax system in the form of the Luxembourg corporate income tax rules (IRC) as set out in recitals 193 to 208. In particular, that reference system is composed of a consistent set of rules that apply on the basis of objective criteria for the taxation of profits of stand-alone companies, where the determination of the taxable profit usually coincides with the accounting profit (subject to certain adjustments based on tax law) and group companies, which resort to transfer prices to allocate profits, alike. In light of the intrinsic objective of that system, both types of companies — non-integrated and integrated companies — should be considered to be in a similar factual and legal situation.

The Commission does not accept this line of reasoning.

As explained in recital 198, the objective of the Luxembourg corporate income tax system is to tax the profits of all companies that fall under its tax jurisdiction, irrespective of whether those companies are integrated or non-integrated companies. As explained in recital 194, Luxembourg corporate income tax is levied on the worldwide profits of domestic companies and foreign companies resident in Luxembourg (unless a tax treaty applies), including Luxembourg branches of foreign companies, while non-resident companies are only taxed on specific Luxembourg-sourced income.

By considering, as Luxembourg and FFT do, that the reference system only includes group companies, since only those companies need to revert to the arm's length pricing to determine their tax base, an artificial distinction is introduced between companies based on their company structure for the purposes of determining their taxable profits that the general Luxembourg corporate income tax system does not recognise when taxing the profits of companies falling within its tax jurisdiction.

Accordingly, the Commission concludes that, in the present case, the reference system against which the contested tax ruling should be examined is the general Luxembourg corporate income tax system as set out in recitals 193 to 208, irrespective of whether corporate income tax is imposed on the profit of group or stand-alone companies and irrespective of the activities they carry out.

In principle, the function of a tax ruling is to establish in advance the application of the ordinary tax system to a particular case, given a set of facts and circumstances specific to that case, for a certain period of time and provided that there is no material change over the application of the ruling in that specific set of facts and circumstances. Where a tax ruling is based on a method of assessment that deviates from what would result from a normal application of the ordinary tax system without justification, that ruling will be considered to confer a selective advantage upon its beneficiary, in so far as that selective treatment results in the lowering of that beneficiary's tax liability in the Member State concerned as compared to companies in a similar legal and factual situation.

As explained in recitals 52 et seq., by the contested tax ruling, Luxembourg accepted a methodology for determining FFT's taxable profit in Luxembourg, as proposed by the latter's tax advisor in the transfer pricing report, which allows FFT to determine its corporate income tax liability in Luxembourg on a yearly basis for the period during which that ruling is valid. More specifically, the transfer pricing report endorsed by the contested tax ruling determines, in the absence of transactions dictated by the market as would exist for a non-integrated independent company, the profit to be allocated to that company of the Fiat group and which translates into the pricing of the transactions it concludes with the other group companies of the Fiat group.

The Court has thus accepted that a tax measure which results in a group company charging transfer prices that do not reflect those which would be charged in conditions of free competition, that is prices negotiated by independent undertakings negotiating under comparable circumstances at arm's length, confers an advantage on that group company, in so far as it results in a reduction of its taxable base and thus its tax liability under the ordinary corporate income tax system.

The principle that transactions between intra-group companies should be remunerated as if they were agreed to by independent companies negotiating under comparable circumstances at arm's length is generally referred to as the ‘arm's length principle’. In the Belgian coordination centres judgment, the Court of Justice endorsed the arm's length principle as the benchmark for establishing whether a group company receives an advantage for the purposes of Article 107(1) of the TFEU as a result of a tax measure that determines its transfer pricing and thus its taxable base.

The purpose of the arm's length principle is to ensure that transactions between group companies are treated for tax purposes by reference to the amount of profit that would have arisen if the same transactions had been executed by independent companies. Otherwise, group companies would benefit from a favourable treatment under the ordinary corporate income tax system when it comes to the determination of their taxable profits that is not available to independent companies, leading to unequal treatment between companies that are factually and legally in a similar situation in light of the objective of such a system, which is to tax the profits of all companies falling under its tax jurisdiction.

The Commission's assessment of whether Luxembourg granted a selective advantage to FFT must therefore consist in verifying whether the methodology accepted by the Luxembourg tax administration by way of the contested ruling for the determination of FFT's taxable profits in Luxembourg departs from a methodology that leads to a reliable approximation of a market-based outcome and thus from the arm's length principle. In so far as the methodology Luxembourg accepted by the contested tax ruling results in a lowering of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies whose taxable profit under that system is determined by the market, that ruling will be deemed to confer a selective advantage to FFT for the purposes of Article 107(1) of the TFEU.

Finally, in response to the argument of Luxembourg and FFT that, because transfer pricing is not an exact science, the assessment by the Commission of the transfer pricing arrangement agreed in the contested tax ruling should necessarily be limited, the Commission recalls that the approximation component of transfer pricing has to viewed in the light of its objective. While the OECD TP Guidelines do indeed acknowledge that transfer pricing is not an exact science in paragraph 1.13 thereof, that same point first explains that ‘[i]t is important not to lose sight of the objective to find a reasonable estimate of an arm's length outcome based on reliable information’. The objective of the OECD TP Guidelines is to develop, for the benefit of tax administrations and multinational enterprises, the most appropriate methods for estimating arm's length prices of cross-border transactions between associated enterprises for taxation purposes. The pursuit of that objective would be impossible if the approximative nature of the transfer pricing exercise could be used to disregard the consensus on appropriate transfer pricing methodologies which those guidelines represent. The approximative nature of the arm's length principle can therefore not be invoked to justify a transfer pricing analysis that is either methodologically inconsistent or based on an inadequate comparables selection.

In conclusion, if it can be shown that the methodology accepted by the Luxembourg tax administration by way of the contested ruling for the determination of FFT's taxable profits in Luxembourg departs from a methodology that leads to a reliable approximation of a market-based outcome and thus from the arm's length principle, that ruling will be found to confer a selective advantage on FFT for the purposes of Article 107(1) of the TFEU, in so far as it leads to a lowering of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies whose tax base is determined by the profits they generate under market conditions.

The first doubt expressed by the Commission in the Opening Decision was that FFT's tax base as agreed to by the contested tax ruling appeared to constitute a fixed range. In addition, the documents submitted during the formal investigation show that FFT's tax base had been fixed at EUR 2 million annually before the contested tax ruling was issued. Although the Circular required FFT to renew its previous ruling request on the basis of a transfer pricing report, the transfer pricing report prepared for the contested tax ruling reached an outcome in terms of tax base, which is practically identical to the previously agreed lump sum amount.

The contested tax ruling accepts a methodology for determining a profit allocation to FFT, which is based on a transfer pricing analysis prepared by the latter's tax advisor that calculates a remuneration for the intra-group financing and treasury functions performed and the risk borne by FFT.

Several methodological choices underlie that transfer pricing analysis: (i) the choice to use the TNMM to estimate FFT's taxable profit in Luxembourg; (ii) the choice of capital as profit level indicator in the application of the TNMM; (iii) the choice of the Basel II framework to calculate that capital and (iv) the choice of the CAPM based on equity prices to determine a required return on that capital.

As explained in the following sections, in recitals 241 to 301, choice (i) is a choice between the five methods described and discussed in detail in the OECD Guidelines. The OECD Guidelines also discuss choice (ii) of a profit level indicator when using the TNMM. However, subsequent methodological choices (iii) and (iv) are not covered by the OECD Guidelines.

Based on these methodological choices, choices of parameters for their implementation and […] adjustments, the tax advisor arrives at a level of remuneration for FFT's intra-group financing and treasury activity that is accepted by the contested tax ruling as respecting the Circular and the arm's length principle.

In the following sections, in recitals 241 to 301, the Commission will explain why it considers several of those methodological choices, choices of parameters and […] adjustments to result in a lowering of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies whose taxable profits are determined by transactions concluded on market terms and, thus, that the contested tax ruling, by accepting those choices and adjustments, confers a selective advantage on FFT for the purposes of Article 107(1) of the TFEU by deviating from the arm's length principle.

The use of a method further requires the use of one or more parameters, so that the method can result in an actual outcome. The choice of method and the choice of parameters cannot be an arbitrary choice. Leaving those choices to the sole discretion of the taxpayer would advantage integrated cross-border groups over companies that transact on the market, the former being given the choice of method and parameters for pricing intra-group transactions to determine their tax base, while the latter transact on market terms without any possibility to adjust their tax base. Those choices must therefore be guided by the objective of achieving an arm's length price for intra-group transactions.

However, when estimating an arm's length price for transfer pricing purposes, the use of a second-best method does not automatically give rise to an advantage for integrated cross-border groups. For example, where such a method is chosen, but that method is used in combination with an overly conservative set of parameters, the remuneration arrived at might nevertheless result in an outcome which is in that specific case equal to a market-based outcome or it might result in an overestimated tax burden, in which case a tax ruling accepting that second-best method would not give rise to an advantage for the purposes of Article 107(1) of the TFEU.

By contrast, even if the most appropriate method is chosen, when it is used in combination with overly favourable parameters the remuneration arrived at using that method might nevertheless underestimate the tax liability of the taxpayer and thus give rise to an advantage for the purposes of Article 107(1) of the TFEU.

In the Opening Decision, the Commission expressed a doubt that the TNMM might not be the most appropriate method for determining an arm's length remuneration and thus the taxable profit of FFT. Compared to the other four methods described in the OECD TP Guidelines, the CUP method is more direct and would, if applicable, provide for a more reliable approximation of a market-based outcome.

However, in light of the information submitted in the course of the formal investigation procedure, the Commission accepts FFT's argument that the CUP method might not be the most appropriate method in the case of FFT. That information shows that FFT concludes transactions with different counterparties of the Fiat group and that the applicable rates, the form and the maturity of the loans provided and the bonds issued by FFT vary, even if the seniority of the transactions does not seem to differ. Consequently, since using the CUP method for transfer pricing purposes would require finding comparable transactions for each individual loan FFT provides, the tax advisor's use of the TNMM to estimate an arm's length remuneration for the functions performed and the risk borne by FFT appears appropriate.

While the Commission considers the tax advisor's use of the TNMM in the case of FFT to be appropriate, the Commission considers several of the methodological choices, choices of parameters and […] adjustments employed by the tax advisor in the application of that method inappropriate for calculating FFT's tax base in Luxembourg.

First and foremost, the Commission does not consider the tax advisor's choice of regulatory capital as an appropriate profit level indicator in the application of the TNMM for estimating an arm's length remuneration for the functions performed by FFT. Instead, the Commission considers that the tax advisor's choice of the TNMM, in combination with its use of a return on equity estimated through the CAPM, necessitates that FFT's accounting equity is used as the profit level indicator against which a return on equity is applied to calculate that remuneration if the outcome is to result in a reliable approximation of a market-based outcome.

More specifically, to ensure that FFT's tax base reflects a reliable approximation of a market-based outcome in line with the arm's length principle, the method used to arrive at an arm's length remuneration for the functions it performs should be methodologically consistent from an accounting perspective, which the choices made by the tax advisor are not.

A return on equity is a profitability ratio. Equity is remunerated through the net profits of a company, that is, income minus all charges incurred in the course of the business, but also all financial charges paid to debt holders. That net profit is thus the profit left to the company to remunerate equity holders: it constitutes a return on equity either through distribution or through increased value of the company. It is therefore a matter of consistency that return on equity is equal to the net profit in accounting terms left to shareholders after all other charges have been paid, divided by the value of the shares in accounting terms (i.e. equity) that that profit is remunerating.

By contrast, it is inconsistent to consider the accounting net profit of the company to remunerate regulatory capital. Regulatory capital is the estimate of a regulator of a minimum capitalisation level to be maintained by a bank or other financial institution; it does not constitute, as such, a claim in that proportion to the profits of the regulated entity. Moreover, financial institutions must hold this level of capital at all times, so that, in practice, they generally hold more than the capital required as a buffer to avoid breaching minimum regulatory requirements in case of losses, which in turn reduces available capital. Any capital additional to the minimum required needs to be equally remunerated from the perspective of investors.

Since FFT's equity level is observable and the CAPM provides for an estimate of return on equity, the tax advisor should have used FFT's accounting equity in the application of the TNMM, instead of a hypothetical level of its regulatory capital, to determine FFT's Luxembourg tax base in line with the arm's length principle.

In any event, while accounting equity is commonly used in the financial sector as a base for calculating a company's profitability, the Commission accepts that a different capital base could, in principle, be used for the application of the TNMM for transfer pricing purposes, so long as the methodology used to arrive at a remuneration for intra-group transactions is consistent. However, in the specific case of FFT, the Commission considers the tax advisor's choice of FFT's hypothetical regulatory capital for this purpose to be inappropriate for arriving at a reliable approximation of a market-based outcome for the following reasons.

Firstly, FFT is not a regulated financial entity to which the Basel II framework applies, so that the application of that framework to estimate its hypothetical regulatory capital is difficult to verify when used only for tax purposes. Indeed, the Basel II framework defines required regulatory capital as a proportion of assets held by the institution weighted by the underlying risk of each such asset. Concretely, the risk-weighting of each asset for that regulatory purpose depends, in particular, on the credit rating of the counterparty, but also on other criteria assessed asset by asset. The burden of the administration for verifying individual asset risk weighting is removed from its prudential supervision context when used to calculate the tax base for transfer pricing purposes instead of minimum capital requirements. It is therefore unlikely that any outcome arrived at as a result of that exercise can be easily verified by the tax administration as constituting a reliable approximation of a market-based outcome in line with the arm's length principle.

Thirdly, the 66 comparables selected by the tax advisor for estimating a return on equity using the CAPM are clearly inadequate for estimating average regulatory capital in the industry or the required return on that capital, since several of those companies are not regulated entities falling under the Basel frameworks (such as stock exchanges), so that those companies might not in fact calculate an estimate of their regulatory capital requirement and it would be impossible to estimate minimum regulatory requirements for individual unregulated institutions on the basis of publically disclosed information alone.

Moreover, using accounting equity as profit level indicator in the present case would have obviated the need for the tax advisor to calculate a separate ‘functions remuneration’, the second component of FFT's estimated tax base in Luxembourg, which itself does not appear to be based on any sound methodology, as explained in recital 80 of the Opening Decision. Indeed, what FFT's tax advisor designates as the ‘Capital used to perform functions’ in the transfer pricing report does not seem to correspond to any customary capital component used in the calculation of return requirements in market valuation. This concept is not defined in the transfer pricing report and there is no indication that such a risk based on the denomination used by the tax advisor would not be covered by either of the categories of regulatory capital, for example, capital to cover operational risk, and in particular risk related to process, if FFT's hypothetical regulatory capital would have been correctly estimated by the tax advisor, which is not the case. Further concerns regarding the separation of the capital into distinct components to which different levels of return, as low as zero, are applied is detailed below, in recitals 277 to 289.

In addition, and without prejudice to the Commission's objections to the use of FFT's hypothetical regulatory capital in the application of the TNMM, the Commission further considers that the inconsistent manner in which FFT's tax advisor applied the Basel II framework by analogy to arrive at that hypothetical level of regulatory capital confers a selective advantage on FFT for the purposes of Article 107(1) of the TFEU, since it also results in a lowering of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies which transact on market terms.

A proper application of the Basel II framework requires, first, an estimate of FFT's RWAs and, then, applying an appropriate regulatory capital ratio to that estimate. FFT's tax advisor underestimates both elements in the transfer pricing report.

Moreover, while under the Basel II framework half of the 8 % requirement could be provided in the form of Tier 2 capital, FFT does not seem to have any Tier 2 capital which could have been used for the regulatory requirement, had that requirement been applied. In the absence of other eligible forms of capital available, the 8 % would have had to be covered by equity.

The Commission therefore maintains its view that the capital requirement ratio under the Basel II framework is 8 % and that the tax advisor's use of 6 % and its acceptance by the Luxembourg tax administration calls into question the conclusion reached in the contested ruling that the resulting profit allocation to FFT reflects a reliable approximation of a market-based outcome in line with the arm's length principle.

In sum, the Commission concludes that even if FFT's hypothetical minimum regulatory capital would have been acceptable as a profit level indicator for the application of the TNMM, the tax advisor underestimated that capital by using an arbitrary and low risk weighting on the assets (excluding most assets from the risk weighting), by applying a lower ratio than the minimum prescribed by the Basel II framework and by not including income from group assets and liabilities in FFT's gross income. The Commission therefore concludes that the contested tax ruling, by accepting those choices, departs from a market-based outcome in line with the arm's length principle. Since those choices lead to a reduction of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies which transact on market terms, the contested tax ruling should be considered to grant a selective advantage to FFT for the purposes of Article 107(1) of the TFEU.

In any event, for the reasons given in recitals 281 to 290, the tax advisor's choice to accord that portion of FFT's equity designated as the ‘Equity supporting the financial investments in FFNA and FFC’ a zero remuneration is inappropriate. Since that choice effectively leads to an unjustified deduction from the capital to be remunerated, and thus from FFT's tax base for taxation purposes, any transfer pricing methodology based on that choice cannot be considered to result in a reliable approximation of a market-based outcome in line with the arm's length principle.

In any event, FFT's annual reports do not seem to confirm Luxembourg's claim that those acquisitions were financed through own funds, if this assertion means that the group provided additional own funds to FFT to acquire FFNA and FFC. Indeed, FFT's equity level (i.e. capital and reserves) in 2010, that is, before the acquisition, was EUR 286 million and that level remained at EUR 287 million in 2011 after the acquisition took place.

Thirdly, in the application of the TNMM based on capital, any estimate of such capital should ensure that FFT is properly capitalised in line with industry standards. This also follows from the approach advocated in the Circular, which considers the functions performed by intra-group financing companies, in substance, comparable to the functions performed by independent financial institutions.

Moreover, that inconsistent result would equally arise in the case FFT acquires new participations during the lifetime of the contested tax ruling. By validating a methodologically flawed deduction for participations, the contested tax ruling could result in FFT effectively being taxed nothing in Luxembourg, provided the value of the subsequent participations acquired results in reducing FFT's capital to be remunerated to nil or even a negative amount. In that case, the intra-group financing functions performed by FFT could remain identical, but the taxation would be much lower or even nil when applying the method validated by the contested tax ruling.

In light of these observations, the Commission concludes that the contested tax ruling, by accepting the tax advisor's application of a return estimated through the CAPM to a hypothetical regulatory capital, where that hypothetical regulatory capital was underestimated as a result of a misapplication of the Basel II framework and inappropriate deductions, departs from the arm's length principle. Since that departure results in a lowering of FFT's tax base under the general Luxembourg corporate income tax system as compared to non-integrated companies which transact on market terms, the contested tax ruling should be considered to grants a selective advantage to FFT for the purposes of Article 107(1) of the TFEU.

In addition to the Commission's conclusion on the inconsistent manner in which FFT's tax advisor arrives at the estimated amount of capital to be remunerated for the application of the TNMM, the Commission considers that the manner in which FFT's tax advisor arrives at the estimated level of required return to be applied to that capital base does not result in a reliable approximation of a market-based outcome and therefore is not in line with the arm's length principle for the reasons presented in recitals 293 to 300.

The Commission notes, in this regard, that the beta represents the non-diversifiable risk of a capital return. Against this background, loan portfolios of banks would, in principle, be more diversified than FFT's portfolio, the exposure of which is concentrated on car companies of the Fiat Group. For this very reason it could be argued that a higher point in the range of comparables should have been selected by FFT's tax advisor for the determination of the beta, arguably higher than the median, rather than the 25th percentile.

Indeed, the Circular itself specifically points to the counterparty risk and more specifically the sector risk, which is amplified here by FFT's concentration on only one sector, when setting an appropriate remuneration. Although the indications of the Circular are not specific in terms of how those risks are to be factored in an actual remuneration calculation, the method chosen by FFT's tax advisor, where risks of intra-group loans (as opposed to third party receivable) do not have any risk-weight and therefore also do not result in any capital requirement or any remuneration, is not in line with the Circular. Therefore, considerations regarding the reduced risk of FFT's activities cannot be accepted.

In light of these observations, the Commission considers that neither the implicit nor the explicit guarantees materially decrease the risk borne by FFT when performing its functions. This risk is higher than has been presented in the transfer pricing report because of the existence of guarantees provided to group companies leading to a significant off-balance sheet exposure, so that a beta higher than the 25th percentile should have been chosen.

In conclusion, the Commission considers that the contested tax ruling, by endorsing the tax advisor's choice of a beta of 0,29 in the application of the CAPM for determining the return on capital to be applied to FFT's hypothetical regulatory capital, results in a profit allocation to FFT that departs from market conditions in line with the arm's length principle. Since that departure results in a lowering of FFT's tax liability under the general Luxembourg corporate income tax system as compared to non-integrated companies which transact on market terms, the contested tax ruling should be considered to grant a selective advantage to FFT for the purposes of Article 107(1) of the TFEU.

Given the tax advisor's choice of the TNMM in the transfer pricing analysis and in light of the observations above, the Commission is of the opinion that, to ensure an appropriate market-based level of remuneration to FFT for the financing and treasury functions it performs within the Fiat group in line with the arm's length principle, that remuneration should be established on the basis of FFT's accounting equity on the basis of its specific facts and circumstances.

The Commission accepts 2012 as reference year for the assessment of FFT's tax base in Luxembourg. The Commission also does not object to the use of comparables database search to estimate arm's length returns, but maintains its objections regarding the appropriate choice of comparables expressed in recital 294.

Indeed, the analysis in Section 7.2.2.9 indicates that the risks borne by FFT to be considered for the calculation of an arm's length remuneration are higher than the risks presented in the transfer pricing report and that, therefore, the pre-tax return on equity of 6,05 % (and the corresponding 4,3 % post-tax) accepted by the contested tax ruling, calculated by FFT's tax advisor using the CAPM, falls well below the required returns on capital in the financial sector, which has consistently remained at and above 10 %, confirming the Commission's conclusion that the tax advisor's choice of comparables was not appropriate.

As follows from the tables above, presented in recital 306, for the period 2010-2013, the average return on equity was 7,2 % for FF and 0,5 % for FFT, which again confirms that the tax advisor's choice of comparables in the transfer pricing report and the resulting return on capital were not appropriate.

To determine an appropriate return on equity to apply to FFT's accounting equity, the aggregate statistics for the banking sector seem more adequate than the set of 66 financial companies, which seem in large part to engage in specific financial activities, as the former would in any case better approximate a central tendency.

The Commission further observes that an arm's length remuneration must not be lower than the difference between the income and the charges of the company. Therefore, if the remuneration of loans provided or deposits received by FFT were to be adjusted and the resulting remuneration of FFT would be higher than a remuneration calculated through transfer pricing, the entire recorded profit must be taxed, because a third party would not accept to reduce its remuneration if none of its counterparties would make justified claims to receive a higher remuneration on their deposits or justified claims to pay a lower remuneration on their loans received.

The Commission is therefore of the opinion that, if the TNMM is used for transfer pricing purposes to calculate an appropriate remuneration due to FFT for the functions it performs with the Fiat group, the correct estimate of FFT's taxable base in Luxembourg corresponds to at least 10 % post-tax applied to the full amount of its accounting equity, which is considered to be broadly in line with the leverage corresponding to the industry average. A taxable base calculated on that basis would result in a profit allocation reflecting market conditions in line with the arm's length principle, since it is a level of profit standalone financial institutions could expect on the market, so that any tax ruling accepting that base for determining FFT's tax liability under the general Luxembourg corporate tax system would not give rise to a selective advantage for the purposes of Article 107(1) of the TFEU.

As explained in recital 193, a reference system is composed of a consistent set of rules that generally apply on the basis of objective criteria to all undertakings falling within its scope as defined by its objective. The Commission considers that the Circular, as regards group financing and treasury companies that make a ruling request, cannot constitute such a system, due to the lack of objective criteria that allow a consistent application of the arm's length principle to intra-group financing transactions.

The Circular does not, however, contain any information as to how to estimate the expected return of capital, whether the use of the CAPM is acceptable and, if so, how the relevant components of the CAPM should be determined. The Circular also does not contain much information as to the equity to be considered except that it must be ‘sufficient in order to assume the risks’.

Therefore, by concluding that the transfer pricing analysis ‘has been realised in accordance with the Circular 164/2 of the 28 January 2011 and respects the arm's length principle’, the Luxembourg tax administration did not rely on any objective criteria laid down in the Circular, since that document does not include any precise and objective criteria that could be used to allow a consistent application of the arm's length principle to intra-group financing transactions.

In addition to the broad wording of the Circular, the Commission has examined the tax rulings submitted to it by Luxembourg and found that its tax ruling practice in relation to group financing and treasury companies further evidences that there is no consistent set of rules that generally apply to all similarly situated taxpayers on the basis of objective criteria.

• A ruling of […] 2013 (No 2)184 concerns the application of the Circular to intermediary financing activities […].

• A ruling of […] 2013 (No 4) also deals with the analysis of the arm's length principle under the Circular to intra-group financing activities. […]

• A ruling of […] 2012 (No 1) concerns a financing activity […].

• A ruling of […] 2012 (No 5) concerns, among others, a tracking loan […].

• A ruling of […] 2013 (No 22) concerns, among others, a financing activity […].

• A ruling of […] 2012 (No 6) concerns a company restructuring. It contains, among others, a tracking loan […].

• A ruling of […] 2012 (No 21) concerns the acquisition of […].

• A ruling on treasury functions of Company E consists of meeting notes produced by Company E's tax advisor and signed on […] by representatives of the Luxembourg tax administration. […]

• A ruling on treasury functions of Company G dates from […], approving a request of the same day by company G. […]

While the Commission does not agree with those arguments, it submits that the lack of a consistent set of rules that generally apply to financing and treasury companies making a ruling request on the basis of objective criteria is best demonstrated by the existence of an alternative tax ruling in favour of [company F]. The [company F] ruling, issued on the same day and covering the same functions as the contested tax ruling (with [company F] covering the industrial segment […] in an alternative structure of group treasury companies), arrives at a substantially different conclusion as regards the companies' respective tax bases, without any apparent justification for that difference in treatment. Indeed, the contested tax ruling agrees on a tax base of around EUR 2,5 million, whereas the tax ruling in favour of [company F] agrees to a remuneration of similar financing and treasury activities resulting in a tax base of EUR […]. This difference in treatment of practically identical companies without any apparent justification further demonstrates that Luxembourg's tax ruling practice, based on the broad wording of the Circular, does not constitute an appropriate reference system for the purposes of the selectivity analysis.

In sum, an examination of the tax ruling practice of the Luxembourg tax administration does not allow for the identification of a consistent set of rules that generally apply on the basis of objective criteria to all undertakings falling within its scope as defined by its objective. That practice can therefore not constitute the framework against which the contested tax ruling is examined to determine whether FFT obtained a selective advantage for the purposes of Article 107(1) of the TFEU.