6.11.2019

Official Journal of the European Union

C 375/1

COMMUNICATION FROM THE COMMISSION

Commission Notice on the calculation of the cost of capital for legacy infrastructure in the context of the Commission’s review of national notifications in the EU electronic communications sector

(Text with EEA relevance)

(2019/C 375/01)

1. Introduction

This Notice sets out the Commission’s methodology for estimating the weighted average cost of capital (WACC) (1) as a reference in the context of reviewing draft measures notified by national regulatory authorities (NRAs or regulators) under Article 7 of the Framework Directive, which is part of the EU’s regulatory framework for electronic communications (the Framework) (2). As of 21 December 2020, Article 32 of the new European Electronic Communications Code (3) will replace Article 7 of the Framework Directive.

The WACC measures companies’ cost of capital. In line with economic theory (4), the cost of capital is the opportunity cost of making a specific investment instead of a different investment with equal risk. Thus, the cost of capital is the rate of return required by a company to make a given investment. The cost of capital can be split into cost of equity and cost of debt, depending on the source of financing.

NRAs use the WACC in the context of regulating operators designated as having significant market power (SMP). NRAs may impose price control obligations for the provision of specific types of interconnection and/or access in situations where there is a lack of effective competition. In cases of price controls, NRAs shall take into account the investment made by the operator and allow a reasonable rate of return. (5)

In the context of Article 7 notifications, the Commission has observed significant discrepancies in estimating the WACC for services provided over electronic communications networks. There is not only a lack of consistency across Member States but also over time (i.e. the same NRA uses different methods at different points in time). The Commission considers that such methodological inconsistencies are likely to distort investment incentives in the Digital Single Market (6) and undermine the development of the internal market by hindering the creation of convergent conditions for investment in electronic communications networks. (7)

The Framework aims at ensuring that NRAs contribute to the development of the internal market by cooperating with each other and with the Commission to ensure the development of consistent regulatory practice and the consistent application of the Framework. (8) In line with the objectives of the Framework, this Notice aims at increasing the consistency of WACC calculations across the Union.

The scope of the Notice is limited to the WACC calculation for legacy infrastructure. For the purposes of this Notice, legacy infrastructure means infrastructure of an SMP operator not subject to a Next Generation Access (NGA) premium. (9) The Notice does not address the applicability or the calculation of NGA risk premiums and excludes any consideration of the appropriateness of price control obligations for new very high capacity networks as defined in Article 2(2) of the Code (10).

In accordance with the objectives of the Framework, NRAs, BEREC and the Commission should all contribute to the development of the internal market by promoting consistent regulatory approaches and the consistent application of the Framework. (11) NRAs should also apply objective, transparent, non-discriminatory and proportionate regulatory principles. (12)

In line with the objectives of the Framework, this Notice is based on the following four regulatory principles: (i) consistency in the methodology used to determine the parameters in the WACC formula; (ii) regulatory predictability to limit unexpected variations in the regulatory approach and in the value of the parameters over time; (iii) the promotion of efficient investment and innovation in new and enhanced infrastructures, taking account of the risk incurred by the investing undertakings; and (iv) transparency of the method to determine the reasonable rate of return on their investments, avoiding unnecessary complexity.

This Notice seeks to achieve these objectives and contribute to a stable regulatory environment that supports efficient investments in electronic communications networks in the Union to the benefit of end users. It will discourage the distortion of investments by inconsistencies in NRAs’ approaches over time and across the Union, which could harm the functioning of the Digital Single Market.

10.

With this Notice, the Commission seeks to provide greater transparency and predictability of its policy and decision-making in the area of electronic communications regulation. Based on the methodology of this Notice, the values of each WACC parameter will be calculated and published annually. The Commission will use these values as reference for its review of draft measures notified under Article 7 of the Framework Directive.

11.

This Notice, together with the annual publication of WACC parameter values consistent with it, will provide greater transparency regarding the Commission’s assessment of the cost of capital, where this is an element of the measures notified to it under Article 7 of the Framework Directive.

12.

This Notice and the accompanying Staff Working Document are without prejudice to the interpretation that may be given by the Commission to the cost of capital in other economic sectors.

2. Estimating the weighted average cost of capital (WACC)

13.

The WACC is calculated as a weighted average of the cost of a company’s two sources of finance: debt and equity. The weights reflect the relative share of each financing source in the total value of the firm:

where:

—	RE is the cost of equity;

—	RD is the cost of debt;

—	E is the value of equity, with being the share of equity in the company value (D+E); and

—	D is the value of debt, with being the share of debt in the company value (D+E).

2.1. The cost of equity

14.

Cost of equity is the return that a company needs to deliver to its shareholders to compensate for the risk of investing/owning a portion of the company. The most commonly used method for estimating cost of equity is the Capital Asset Pricing Model (CAPM) (13) . All NRAs in the electronic communications sector estimate the cost of equity by using the CAPM. The main reasons for this are the relative simplicity of its calculation and its long track record (when compared to alternatives). (14) The Commission considers this as the appropriate approach to promote a predictable and homogeneous methodology across NRAs.

15.

According to the CAPM, in a competitive market investors are willing to hold risky equities if their return is higher than the return they could earn on a risk-free asset (the risk-free rate, RFR) (15). This additional return is called market risk premium or equity risk premium (ERP) (16).

16.

According to the basic assumption of the CAPM, risks can be divided into two categories:

—	Specific or diversifiable risks (17): risks that can be eliminated by diversification. Specific risks typically affect an individual firm or sector, such as a change in the price of an input used by a company or sector.

—	Systematic or non-diversifiable risks (18): risks affecting all firms in the economy, e.g. changes in interest rates and general price levels or macro-economic shocks in the demand side.

17.

In the CAPM framework, specific or diversifiable risks should not be taken into account to estimate the cost of capital. This is because in efficient capital markets investors should be able to reduce such risks by holding a diversified investment portfolio. (19)

18.

In contrast, systematic risks cannot be diversified away. For this reason, in the CAPM framework the estimated cost of capital only reflects the compensation for systematic risks.

19.

According to the CAPM, the cost of equity is the sum of the RFR and ERP, multiplied by the beta coefficient (20). The beta coefficient is included to capture and adjust for the specific sensitivity of the equity to market fluctuations.

20.

In this regard, in the CAPM framework, the cost of equity is calculated as follows:

RE = RFR + β × ERP

where:

—	RE is the cost of equity;

—	RFR is the risk-free rate;

—	β is the beta; and

—	ERP is the equity risk premium.

2.2. The cost of debt

21.

The cost of debt can be measured directly as the interest paid by the company on its debt, but it is often expressed as the sum of the RFR and a debt premium:

RD = RFR + Debt Premium

where:

—	RD is the cost of debt;

—	RFR is the risk-free rate; and

—	The debt premium is the additional return that lenders require from a company with a given credit risk, over and above the RFR.

2.3. Estimating the parameters of the WACC

22.

The formula for estimating the WACC is the following:

23.

There are two sets of methodological assumptions affecting the WACC values estimated by NRAs across the Union. Some assumptions are common to several parameters (e.g. length of the averaging period) while others are specific to a single parameter in the WACC formula (e.g. adjustment to the estimation of the beta). Further, two types of parameters need to be distinguished: those that reflect general economic conditions (RFR and ERP) and those that reflect conditions affecting specific companies (or industry sector) (debt premium, beta and gearing (21)).

24.

The next sections address both sets of methodological assumptions: common vs single parameters and general economic vs company-specific conditions.

3. Assumptions common to several WACC parameters

25.

The next subsections describe assumptions that are common to several WACC parameters.

3.1. The length of the averaging period

26.

The averaging period is the length of the reference period that regulators use to derive an average value for the RFR, beta and cost of debt.

27.

To ensure consistency in the estimation of WACC parameters, the Commission considers it appropriate to use the same averaging period for all parameters. Longer averaging periods are likely to promote greater predictability and stability in the value of the parameters, albeit at the cost of lower static efficiency. A 5-year averaging period, which is the one most commonly used by NRAs, is likely to strike the right balance between predictability and efficiency. (22)

3.2. The averaging method

28.

When estimating WACC parameters, regulators need to decide which averaging method to use. Typically, they use (i) an arithmetic average, (ii) a geometric average or (iii) a median.

29.

The Commission considers the arithmetic average method the most appropriate for estimating WACC parameters. Firstly, using a single averaging method is likely to be more transparent for stakeholders than combinations of several methods. Secondly, the arithmetic average is the most commonly used approach and the easiest to calculate. (23)

3.3. The frequency of the sampling period

30.

The frequency of the observations used by regulators to estimate the RFR and the beta and debt premium is typically daily, weekly or monthly. The choice of the frequency of the observations, together with the length of the averaging period, determines the size of the sample and is therefore liable to influence the parameter values and, ultimately, the final WACC value.

31.

Using the same frequency of observations to estimate all parameters ensures consistency. A weekly frequency is likely to be an efficient choice, as, combined with a 5-year averaging period, it is likely to provide sufficient observations to derive a robust estimate and mitigate problems of low liquidity of shares (if any).

4. Parameters reflecting general economic conditions

32.

There are two WACC parameters reflecting general economic conditions: the risk-free rate and the equity risk premium.

4.1. The risk-free rate (RFR)

33.

The RFR is the expected rate of return on a risk-free investment. For an investment to be free of risk, the risk of default on payments needs to be zero and there must be no reinvestment risk (i.e. the investor can reinvest future interest payments at the same rate of return as at the time the asset was first purchased). NRAs typically approximate RFRs using the yields of domestic government bonds.

34.

Government bond yields are likely to reflect appropriately the respective domestic RFR. The Commission considers that the use of domestic government bonds, together with a consistent methodology, will ensure that differences in RFRs reflect actual differences in financing conditions between Member States.

35.

NRAs frequently use government bonds with a residual maturity (24) of 10 years. Yields of 10-year bonds tend to be less volatile than shorter-term bonds and more consistent with the long lifetime of investments in electronic communications networks. The Commission therefore considers them the most appropriate benchmark for the purposes of this Notice. (25)

36.

To ensure consistency in the estimation of the RFR, it is preferable to use a single, reliable, transparent and easily accessible source of information for government bond yields (e.g. Eurostat (26)). An adjustment for central bank quantitative easing programmes is not necessary. (27)

4.2. The equity risk premium (ERP)

37.

The ERP is the expected return on equities over and above the RFR, i.e. the expected additional interest for holding higher-risk equities compared with the interest for holding risk-free assets. The ERP compensates for the added risk of investing in equity rather than in a risk-free asset.

38.

The Commission identified disparities in the approaches used by regulators to estimate the ERP, with some using an EU benchmark, others a national ERP or a combination of both. A Union-wide ERP is consistent with empirical evidence suggesting that financial markets in the Union are increasingly integrated (as shown by their increased correlation) and therefore have convergent ERPs. (28) It is also consistent with the evidence suggesting that investors in the EU telecoms sector do not display ‘home bias’, as a significant share of electronic communications companies’ shareholders are non-nationals. (29) The Commission therefore considers a single Union-wide ERP to be appropriate for the purposes of this Notice. Further, in line with the approach most commonly used by regulators, the Commission considers it appropriate to estimate this Union-wide ERP using historical series of market premiums in Member States.

5. Company-specific parameters

39.

Some WACC parameters capture general economic conditions (RFR and ERP) while others (beta, gearing and debt premium) reflect the economic conditions of the specific company for which the regulator estimates the WACC.

40.

To derive the values of company-specific WACC parameters, regulators typically rely on a group of electronic communications companies (‘peer group’), usually including their national SMP operator(s). NRAs use the parameter values of the companies in the peer group as reference to derive appropriate values for company-specific parameters in their regulatory measures.

41.

The first important step in estimating company-specific parameters is to decide on the criteria for selecting the companies that will be part of the peer group.

5.1. Criteria for selecting the peer group

42.

For the peer group, companies listed on a stock exchange and having liquidly traded shares (30) are the most suitable. Further, companies that own electronic communications infrastructure, as opposed to renting it, and have their main operations in the Union are best suited for calculating a WACC in accordance with the aims of this Notice.

43.

Efficiency considerations favour confining the peer group to companies with investment grade credit rating (31) and companies not involved in any substantial mergers and acquisitions, as the latter would affect a company’s value in ways unrelated to its underlying systematic risk.

44.

In summary, the following criteria would be in line with the regulatory principles presented in this Notice. The companies selected for the peer group:

—	are listed on a stock exchange and have liquidly traded shares;

—	own and invest in electronic communications infrastructure;

—	have their main operations located in the Union;

—	have an investment grade credit rating; and

—	are not, or have not been recently, involved in any substantial mergers and acquisitions.

5.2. The equity beta

45.

Within the CAPM framework, equity beta reflects the systematic risk faced by a company relative to the average company in the market. In practice, the equity beta is estimated through a regression analysis, i.e. by estimating the correlation between the returns of a company’s share and the returns of a market index (which is meant to approximate the whole economy).

46.

A European market index is more consistent with the use of a Union-wide ERP value than a domestic one. The academic literature expresses a preference for broadly-based indices and for value-weighted indices. (32) In line with these criteria, there are several market indices encompassing European equities such as STOXX Europe TMI (33), S&P Europe 350 (34), Eurostoxx50 (35), and MSCI Europe (36). The Commission considers it appropriate to use a market index representing a large share of the free float market capitalisation in the Union (e.g. the STOXX Europe TMI).

47.

In estimating the equity beta, the Commission does not consider it appropriate to make adjustments (37), as these are unlikely to improve the efficiency of the beta estimator and make the estimation unnecessarily complex and less transparent.

48.

To compare companies’ betas within the peer group, it is necessary to convert (unlever) their equity betas into asset betas. (38) Asset betas reflect companies’ systematic risk, free of financial risk (i.e. risk associated with the level of financial leverage). A simple formula to derive asset betas from equity betas is the following:

where:

—	βA = asset beta

—	βD = debt beta

—	βE = equity beta

—	E = company's equity

—	D = company's net debt

—	V = company's value (E + D)

49.

There are significant practical difficulties in estimating debt betas but values typically range from 0-0,2 (39). Using a single value for the debt beta would reduce complexity and improve transparency of the WACC calculation. An intermediate debt beta value of 0,1 seems a reasonable choice.

50.

For estimating the WACC, the asset beta of the peer group is converted (back) into equity beta (i.e. add back the impact of debt on the company beta) using the formula above, which, when solving for the equity beta, results in:

5.3. The gearing

51.

As mentioned above, companies can finance themselves through debt (D) or equity (E). The sum of debt and equity equals the company value (V):

V = D + E

52.

The share of debt in the company value (i.e.

) is called ‘gearing’. Gearing is a leverage ratio. It shows to what extent a company is funded by lenders as opposed to shareholders.

53.

The share of debt (i.e.

) and the share of equity (i.e.

) are used to compute the cost of debt (i.e.

) and the cost of equity (i.e.

) when estimating the WACC.

54.

The most common approach for estimating the gearing, which is considered appropriate for the purposes of this Notice, is to use the book value of a company’s net debt, including the value of financial leases. (40)

5.4. The debt premium and the cost of debt

55.

The cost of debt can be measured directly as the interest paid by a company on its debt or indirectly as a premium on the RFR.

56.

The debt premium can be estimated as the spread between the domestic RFR and the yield of long-term corporate bonds (as close as possible to the 10-year maturity used for the RFR). (41)

57.

For reasons of consistency, the Commission considers it more appropriate to estimate the cost of debt (indirectly) as the sum of the RFR and the debt premium.

6. Taxes and inflation

The tax treatment of debt and inflation expectations affect the WACC and thus need be incorporated in the WACC calculation.

6.1. Taxes

58.

Interest on debt is a tax-deductible expense for corporations. The post-tax WACC accounts for this favourable tax treatment of debt as follows:

where:

—	RE is the cost of equity and RD is the cost of debt;

—	and are the weights of equity and debt, respectively, on the value of the firm; and

—	Tc is the marginal tax rate.

59.

NRAs typically uplift the (post-tax) cost of equity (42) – to meet the requirements of equity investors, as estimated by the CAPM – to a pre-tax cost of equity. The pre-tax WACC is calculated by dividing the post-tax WACC by (1 – Tc ), in order to account for corporate taxes, and can be expressed as follows:

60.

The Commission considers it appropriate to use the relevant domestic corporate tax rate, which is the common approach adopted by NRAs, to estimate the pre-tax WACC.

6.2. Inflation

61.

Investors maximise their inflation-adjusted or real returns. There are typically two ways in which NRAs take inflation into account: (43)

a)	inflation is compensated for through the annual indexation of the company's assets and only a real WACC return is allowed; or

b)	inflation expectations are included in the return on capital, by using a nominal WACC, without any adjustment to the company’s asset base.

62.

NRAs using the first approach typically convert the nominal WACC to the real WACC. One common conversion method is the Fisher equation:

where π is the inflation rate.

63.

The Commission considers it appropriate to use a Eurozone-wide inflation estimate for Eurozone Member States; for non-Eurozone Member States national inflation estimates may be justified. In both cases, forward-looking estimates are more appropriate and ideally cover a period equal to the 10-year maturity of government bonds used to estimate the RFR. In practice 10-year inflation forecasts are rarely available, thus shorter term forecasts may be used (e.g. inflation forecast 5 year ahead by ECB).

7. Role of BEREC and the Commission in the calculation of WACC parameters

64.

In the preparatory steps leading to the adoption of the Notice, the Commission services worked in close cooperation with BEREC. In the context of this collaboration, the Commission invited BEREC to estimate the WACC parameters consistent with the approach described in this Notice. BEREC agreed to estimate the parameter values and publish them on an annual basis. (44) This will greatly facilitate the work of NRAs in preparing periodic WACC reviews and the Commission’s review of such notifications.

65.

In the annual calculation exercise, BEREC, in close collaboration with the Commission, will estimate (i) the parameters reflecting general economic conditions and (ii) the company-specific parameters for the peer group.

66.

Regarding the parameters reflecting general economic conditions, BEREC will estimate the RFR for each Member State and a single Union-wide ERP.

67.

Regarding the company-specific parameters, BEREC will prepare a list of companies suitable for the peer group and estimate the equity beta, gearing, debt premium and cost of debt for each company included in the list. Further, BEREC will describe factors that may justify NRAs removing one or more companies from the list to take account of national specificities.

8. Frequency of WACC reviews

68.

NRAs set/review the WACC with different periodicity, ranging from more than once a year up to once every 4-5 years. This difference in the frequency is one of the factors behind differences in WACC values across Member States.

69.

Increasing the consistency in the frequency of WACC calculations across Member States would reduce unjustified differences in WACC values.

70.

The Commission considers that updating the national WACC value at least once per year is appropriate to take account of recent economic conditions. (45)

9. Transitional period towards the adoption of the methodology in this Notice

71.

When reviewing notifications under the Article 7 procedure, the Commission will, as a rule, use the methodology described in the present Notice from 1 July 2020. However, in justified cases and at the request of the notifying NRA, the Commission will not base its review of draft measures on this methodology during a transitional period of up to one year (starting from 1 July 2020) (46). For example, this may be justified when the review based on this methodology, if applied by the national regulator, would result in significant changes in the WACC value undermining regulatory stability and predictability. During the transitional period of one year, the Commission will also take into consideration if the full set of WACC parameters to be published by BEREC is available and the possibility for the NRAs to rely on those parameters in their analysis.

(1) The WACC represents the value for which the investor needs to be compensated for an investment. In the context of telecoms regulation, the WACC is calculated by the relevant national regulatory authority and added to the maximum allowed wholesale price that the regulated operator can charge for access to its infrastructure. For a comprehensive explanation of the concept of WACC, see section 2 in the Staff working document accompanying this Notice.

(2) The regulatory framework currently in force consists of (i) Directive 2002/21/EC of the European Parliament and of the Council of 7 March 2002 on a common regulatory framework for electronic communications networks and services (Framework Directive) (OJ L 108, 24.4.2002, p. 33), as amended by Directive 2009/140/EC (OJ L 337, 18.12.2009, p. 37), and Regulation (EC) No 544/2009 (OJ L 167, 29.6.2009, p. 12); (ii) Directive 2002/20/EC of the European Parliament and of the Council of 7 March 2002 on the authorisation of electronic communications networks and services (Authorisation Directive) (OJ L 108, 24.4.2002, p. 21); (iii) Directive 2002/19/EC of the European Parliament and of the Council of 7 March 2002 on access to, and interconnection of, electronic communications networks and associated facilities (Access Directive) (OJ L 108, 24.4.2002, p. 7); (iv) Directive 2002/22/EC of the European Parliament and of the Council of 7 March 2002 on universal service and users’ rights relating to electronic communications networks and services (Universal Service Directive) (OJ L 108, 24.4.2002, p. 51) and (v) Regulation (EU) No 531/2012 of the European Parliament and of the Council of 13 June 2012 on roaming on public mobile communications networks within the Union (OJ L 172, 30.6.2012, p. 10), as amended by Regulation (EU) 2015/2120 of the European Parliament and of the Council of 25 November 2015 (OJ L 310, 26.11.2015, p. 1) and Regulation (EU) 2017/920 of the European Parliament and of the Council of 17 May 2017 (OJ L 147, 9.6.2017, p. 1).

(3) Directive (EU) 2018/1972 of the European Parliament and the Council of 11 December 2018 establishing the European Electronic Communications Code (the Code) (OJ L 321, 17.12.2018, p. 36). Pursuant to Article 124(1) of the Code, Member States are required to adopt national transposition measures by 21 December 2020 and apply them from that date. References in existing laws, regulations and administrative provisions to the Directives repealed by the Code shall be construed as references to the Code.

(4) See for example Brealey, Myers, Allen (2014), Principles of Corporate Finance, McGraw Hill, Chapter 9.

(5) In accordance with Article 13 of the Framework Directive. See also Commission Recommendation 2013/466/EU of 11 September 2013 on consistent non-discrimination obligations and costing methodologies to promote competition and enhance the broadband investment environment (OJ L 251, 21.9.2013, p. 13), on circumstances under which wholesale pricing flexibility may be justified and, as from 21 December 2020, Article 74 of the Code.

(6) The Digital Single Market strategy (https://ec.europa.eu/digital-single-market/en/policies/shaping-digital-single-market) aims at ensuring access to online activities for individuals and businesses under conditions of fair competition, consumer and data protection.

(7) In accordance with the objectives set out in Article 8(2)(b) and 8(2)(c) of the Framework Directive.

(8) In accordance with the objective set out in Article 7(2) and 8(3)(d) of the Framework Directive.

(9) The Notice does not prejudge whether additional premiums for specific investments (e.g. next generation access networks) are justified. On this issue, see Commission Recommendation 2010/572/EU of 20 September 2010 on regulated access to Next Generation Access Networks (NGA) (NGA Recommendation) (OJ L 251, 25.9.2010, p. 35), in particular point 25 and Annex I. The lower risk profile of investment into FTTN/VDSL (compared to fibre to the home) is discussed in Annex I, section 6 of the NGA Recommendation. In such cases, NGA networks fall within the scope of legacy infrastructure.

(10) The Code tasked the Body of European Regulators for Electronic Communications (BEREC) with issuing guidelines on the criteria that a network is to fulfil to be considered a very high capacity network, by 21 December 2020. Further, the Code includes several provisions regarding the process of migration from legacy infrastructure to new very high capacity networks, for example, in Article 81.

(11) See Article 8(3)(d) and 8(5)(a) of the Framework Directive.

(12) See Article 8(5) of the Framework Directive.

(13) W.F. Sharpe, ‘Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk’, The Journal of Finance, Vol. 19 (September 1964), pp. 425-442; and J. Lintner, ‘The Valuation of Risk Assets and the Selection of Risky Investments in Stock Portfolios and Capital Budgets’, The Review of Economics and Statistics, Vol. 47 (February 1965), pp. 13-37.

(14) For more details, see section 2.1 of the Staff working document accompanying this Notice.

(15) The return required by investors on a risk-free investment.

(16) The return in excess of the RFR expected by investors for the additional risk involved in a market investment.

(17) Specific risks may also be referred to as unsystematic risks, residual risks, unique risks or diversifiable risks.

(18) Non-diversifiable risks may also be referred to as market risks, systematic risks or undiversifiable risks.

(19) Diversification works because prices of different stocks do not move exactly together (from a statistical point of view, stock price changes are less than perfectly correlated). In a diversified investments portfolio, movements in individual assets are likely to be compensated by movements of other assets in the portfolio over the long term.

(20) The beta is the covariance between the return of an asset (typically, the market value of the company) and the market return (typically, the market value of a stock index that is taken to represent the whole market or economy) divided by the variance of the market return. Shares with betas greater than 1,0 amplify the overall movements of the market whereas those with betas between 0 and 1,0 reduce the overall movements of the market. For example, a beta of 0,5 means that if the market declines by 1 %, one would expect the value of the investment to decline by 0,5 %. A beta of 1,5 means that if the market declines by 1 %, one would expect the value of the investment to decline by 1,5 %.

(21) Gearing is a measure of a company’s financial leverage. It compares the amount of debt financing to the amount of owner’s equity. The gearing determines the weight of cost of debt and cost of equity in the WACC. The gearing is explained further in section 5.3 of this Notice.

(22) For more details, see section 5.1.1 (in particular 5.1.1.4) of the Staff working document accompanying this Notice.

(23) See section 5.1.2 of the Staff working document accompanying this Notice.

(24) Residual maturity is the remaining time until the repayment of the bond. Throughout the Notice, bond maturity refers to residual bond maturity.

(25) See section 5.2.1.3 and section 5.2.1.4 of the Staff working document accompanying this Notice.

(26) Eurostat publishes the yields of 10-year government bonds for each Member State on a monthly basis (see https://ec.europa.eu/eurostat/tgm/table.do?tab=table&plugin=1&language=en&pcode=teimf050).

(27) See section 5.2.2 of the Staff working document accompanying this Notice.

(28) See section 5.2.1.3 and section 5.2.1.4 of the Staff working document accompanying this Notice.

(29) See section 5.2.2.3 of the Staff working document accompanying this Notice.

(30) Thinly traded shares may not reveal the correct and up-to-date value of the underlying companies because they cannot be easily sold without a significant change in price. They also tend to be more volatile than liquid ones.

(31) Investment grade credit rating depends on the specific credit rating agency: it corresponds to ratings from Aaa to Baa3 in the case of Moody’s and to ratings from AAA to BBB in the case of Standard and Poor’s.

(32) See section 5.3.3.2 of the Staff working document accompanying this Notice.

(33) https://www.stoxx.com/index-details?symbol=BKXP

(34) http://us.spindices.com/indices/equity/sp-europe-350

(35) https://www.stoxx.com/index-details?symbol=sx5e

(36) https://www.msci.com/europe

(37) Traditional adjustments to the equity beta are those proposed by Dimson (correcting distortions in the beta estimate when using daily returns due to mismatch between changes in market index and the time for the company stock to react to these); Blume (in the long term companies’ beta should tend towards a value of 1); or Vasicek (in the long term companies’ beta should tend towards an industry average).

(38) The conversion (unlevering) removes the impact of debt on the equity (or levered) beta. In other words, the asset beta abstracts away from the specific financing structure of a company, thereby allowing for comparisons between the betas of different companies (irrespective of their individual financing structures).

(39) See The Brattle Group, Review of approaches to estimate a reasonable rate of return for investments in electronic communications networks in regulatory proceedings and options for EU harmonization, final report published on 14 July 2016, p. 88, available at https://publications.europa.eu/en/publication-detail/-/publication/da1cbe44-4a4e-11e6-9c64-01aa75ed71a1/language-en

(40) See section 5.3.4 of the Staff working document accompanying this Notice.

(41) See section 5.3.5 of the Staff working document accompanying this Notice, and the literature cited therein.

(42) See section 5.4.1 of the Staff working document accompanying this Notice.

(43) See section 5.4 of the Staff working document accompanying this Notice.

(44) At the 38th Ordinary Plenary Meeting of the Board of Regulators of 7-8 March 2019, BEREC agreed to calculate the WACC parameters (BEREC envisaged calculating the risk-free rate and the equity risk premium already in 2019 and the beta, gearing and cost of debt starting from 2020). It has since emerged that BEREC expects to have published the full set of relevant parameters by mid-2020. The conclusions of this meeting (BoR (19) 45) are available at

https://berec.europa.eu/eng/document_register/subject_matter/berec/board_of_regulators_meetings/meeting_conclusions/8549-conclusions-of-the-38th-berec-board-of-regulators-plenary-meetings-on-7-8-march-2019-in-budapest-hungary

(45) See section 8 of the Staff working document accompanying this Notice.

(46) See section 9 of the Staff working document accompanying this Notice.

6.11.2019

Official Journal of the European Union

C 375/16

Final Report of the Hearing Officer (1)

Qualcomm (predation)

(AT.39711)

(2019/C 375/06)

1. Introduction

(1)

The draft decision to which this report relates (the ‘Draft Decision’) finds that Qualcomm Inc. (‘Qualcomm’), from 1 July 2009 to 30 June 2011 inclusive, supplied below cost certain quantities of three of its UMTS-compliant chipsets (2) to two of its key customers, Huawei and ZTE, with the intention of eliminating Icera Inc. (‘Icera’ (3)), its main competitor at that time in the market segment offering advanced data-rate performance. According to the Draft Decision, this infringed Article 102 TFEU and Article 54 of the Agreement on the European Economic Area.

2. Initial investigation phase

(2)

Case AT.39711 stems from an initial complaint made by Icera to the Commission on 30 June 2009, replaced by a revised and updated version dated 8 April 2010.

(3)

The Commission started its investigation on the basis of the allegations made in the latter replacement version. On 26 July 2010, Qualcomm submitted observations concerning that version.

(4)

By submissions of 1 May, 17 May and 31 July 2012, Nvidia, which had acquired Icera in May 2011, supplied further information, supplementing and bringing to the fore allegations concerning predatory pricing.

(5)

Between June 2010 and April 2015, the Commission sent several requests for information to Qualcomm, Icera/Nvidia and other actors in the baseband chipset sector. On 8 August 2013, the Hearing Officer (4), in accordance with Articles 3(7) and 4(2)(c) of Decision 2011/695/EU, extended the time limit for Qualcomm to respond to one such request, dated 10 July 2013. (5)

(6)

On 3 March 2014, following requests in 2013 from Qualcomm, the Commission Directorate-General for Competition (‘DG Competition’) provided Qualcomm with less redacted versions of the initial complaint of 2009 and of the amended complaint of 2010.

(7)

On 16 July 2015, the Commission initiated proceedings, for the purposes of Article 11(6) of Council Regulation (EC) No 1/2003 (6) and Article 2(1) of Commission Regulation (EC) No 773/2004 (7), in respect of Qualcomm in Case AT.39711.

(8)

Following a ‘state-of-play’ meeting held on 3 September 2015 between DG Competition and Qualcomm, DG Competition provided Qualcomm on 11 September 2015 with the submissions by Nvidia mentioned in (4) above.

3. First statement of objections

(9)

On 8 December 2015, the Commission issued, for the purposes of Article 27(1) of Regulation (EC) No 1/2003 and Article 10(1) of Regulation (EC) No 773/2004, a statement of objections addressed to Qualcomm (the ‘SO’).

3.1. Access to the file following the SO

(10)

On 21 December 2015, DG Competition provided Qualcomm with documents on an electronic storage device for the purposes of access to the investigation file in Case AT.39711 in accordance with Article 27(2) of Regulation (EC) No 1/2003. DG Competition subsequently provided Qualcomm with additional material, in particular less redacted versions of third-party submissions. However, not all access-to-file matters could be resolved between DG Competition and Qualcomm.

(11)

By letter to me dated 18 April 2016, Qualcomm requested, in accordance with Articles 3(7) and 7(1) of Decision 2011/695/EU, additional access to, or where appropriate clarification of, certain materials on the case file. My decision dated 13 July 2016 describes how I dealt with the numerous requests concerned, striking a balance between Qualcomm’s right to be heard and competing confidentiality interests. In essence, this involved the provision of less redacted versions of certain documents originating from Icera/Nvidia and various other third parties. Allowing the information providers concerned to be heard (8) and, where relevant, to draw up new versions of the documents concerned took some time, generally requiring several rounds of correspondence.

3.2. Time limit for Qualcomm’s written response to the SO

(12)

The cover letter enclosing the SO gave Qualcomm four months in which to reply in writing to the SO. This time limit was set taking into account, in particular, the size and nature of the file, the content of the SO and the fact that, on the adoption date of the SO, a statement of objections addressed to Qualcomm in Case AT.40220 – Qualcomm (exclusivity payments) had also been adopted. The time limit in Case AT.39711 was to start running from the date of, in substance, provision of access to the file. Following a request from Qualcomm, DG Competition moved the deadline for Qualcomm’s written response to 6 May 2016.

(13)

In the letter to me dated 18 April 2016 mentioned in (11) above, Qualcomm took issue with the revised deadline of 6 May 2016, seeking, in effect, additional time. By email to Qualcomm dated 29 April 2016, I suspended the running of the time limit for the submission of Qualcomm’s written response to the SO.

(14)

In my decision dated 13 July 2016 (see (11) above) I lifted that suspension and set a new deadline for Qualcomm’s written response to the SO at 15 August 2016. Qualcomm submitted its written response (the ‘SO Response’) on that date.

3.3. First oral hearing

(15)

On 10 November 2016, Qualcomm developed its arguments at an oral hearing that it had requested in the SO Response. (9)

4. Interested third person

(16)

In May 2016, I admitted an interested third person to proceedings in Case AT.39711 in accordance with Article 27(3) of Regulation (EC) No 1/2003, Article 13(1) of Regulation (EC) No 773/2004 and Article 5 of Decision 2011/695/EU. In July 2016, in accordance with Article 13(1) of Regulation (EC) No 773/2004, DG Competition informed that interested third person in writing of the nature and subject matter of the procedure (10) and set a time limit within which that person could make known its views in writing. In August 2016, the interested third person indicated by email to DG Competition that it had no comments to submit. It did not request to participate in an oral hearing and took no further part in proceedings in Case AT.39711.

5. Further investigative measures and related litigation

(17)

Following the first oral hearing, the Commission undertook further measures of investigation in the light of Qualcomm’s written and oral response to the SO.

(18)

In particular, on 30 January 2017, the Commission sent Qualcomm a request for information pursuant to Article 18(2) of Regulation (EC) No 1/2003. Qualcomm did not provide any response. On 31 March 2017, the Commission demanded this information by decision pursuant to Article 18(3) of that regulation (the ‘RFI Decision’). In 2017, the Commission also sent requests for information to Nvidia and two third parties.

(19)

By letter dated 10 April 2017, Qualcomm sought an extension until 28 July 2017 of the time limit for responding to the RFI Decision. On 26 April 2017, DG Competition rejected the arguments in that letter but, ‘as a matter of courtesy’ allowed Qualcomm until 26 May 2017 to respond to certain questions in the RFI Decision and until 9 June 2017 in relation to other specified questions in the RFI Decision.

(20)

On 8 May 2017, Qualcomm took issue with these deadlines in accordance with Article 4(2)(c) of Decision 2011/695/EU. By decision dated 15 May 2017, I revised them to 16 and 30 June 2017 respectively. Qualcomm’s substantive response to the RFI Decision respected these revised deadlines.

(21)

Subsequently, the Commission addressed clarification questions to Qualcomm. It also sent Qualcomm an additional request for information pursuant to Article 18(2) of Regulation (EC) No 1/2003 on 10 November 2017.

(22)

On 13 June 2017, Qualcomm lodged an application for the annulment of the RFI Decision and sought interim measures restraining the operation of the RFI Decision pending adjudication of this application. In July 2017, the President of the General Court dismissed the application for interim measures. (11) In April 2019, the General Court dismissed Qualcomm’s annulment action. (12)

6. Supplementary statement of objections

(23)

On 19 July 2018, the Commission adopted a supplementary statement of objections addressed to Qualcomm (the ‘SSO’). Compared with the SO, the SSO notably alleged a shorter duration of predation and used a revised methodology for conducting a comparison of Qualcomm’s prices and costs concerning the allegedly predatory sales concerned.

6.1. Access to the file following the SSO

(24)

On 31 July 2018, DG Competition provided Qualcomm with an electronic storage device containing copies of accessible documents that became part of the Commission’s investigation file after the adoption of the SO.

(25)

By letter to DG Competition dated 8 August 2018, Qualcomm sought, among other things, further access to documents that were part of the investigation file at the time of adoption of the SO but which, due to the passage of time, could, according to Qualcomm, no longer be considered as containing genuinely confidential information. By letter of 14 August 2018, DG Competition declined this request. It also addressed the other matters raised in Qualcomm’s letter of 8 August 2018.

(26)

On 19 September 2018, I received a reasoned request for further access to the file from Qualcomm. However, this request was materially different to the more general request made to DG Competition on 8 August 2018 (see (25) above). In the light of Articles 3(7) and 7(1) of Decision 2011/295/EU, I considered that Qualcomm’s request of 19 September 2018 was not ripe for my independent review. On 21 September 2018, I informed Qualcomm by email that I had transferred this request to DG Competition.

(27)

On 28 September 2018, DG Competition responded to Qualcomm’s letter of 19 September 2018. In particular, DG Competition granted further access to most of the documents listed by Qualcomm and explained that redactions featuring in the original versions of certain documents on the investigation file were attributable to legal professional privilege (‘LPP’) claims made by Nvidia.

(28)

By letter to DG Competition dated 4 October 2018, Qualcomm made a number of complaints inspired in large part by DG Competition’s explanations of 28 September 2018 regarding Nvidia’s LPP claims. DG Competition responded by letter dated 9 October 2018.

(29)

On 12 October 2018, Qualcomm wrote to me with, in essence, three requests: (i) that I direct DG Competition ‘to confirm whether it ha[d] ascertained that the … redactions contained in [specified documents were] based on sound assertions of LPP ....’; (ii) ‘access to unredacted versions’ of specified documents, of which Qualcomm already had redacted versions; and (iii) that I direct DG Competition ‘to address Qualcomm’s request to be provided with any documents which, contrary to what the [i]ndex [of the file] indicate[d], [were], in fact, accessible’.

(30)

In my decision dated 7 November 2018, I explained that Qualcomm’s first request described in (29) above did not concern access to the file and that the Hearing Officer does not have decisional powers to order such a course of action. As for the second request described in (29) above, I considered this to have been satisfied by the time of this decision, since the two third parties concerned had, by that stage, dropped their relevant confidentiality claims. I granted Qualcomm’s third request. Consequently, on 8 November 2018, DG Competition sent Qualcomm a follow-up letter, explaining in substance that the index of the investigation file was accurate.

6.2. Time limit for Qualcomm’s written response to the SSO

(31)

The Commission initially granted Qualcomm eight weeks for responding in writing to the SSO. On 2 August 2018, Qualcomm asked DG Competition for an additional period of ‘at a minimum … three additional months from the working day following the provision of effective access to file’. By letter of 9 August 2018, DG Competition granted Qualcomm, ‘as a matter of courtesy’, an extension of two weeks, bringing the deadline to 10 October 2018.

(32)

Dissatisfied, on 14 August 2018 Qualcomm sought, under Article 9 of Decision 2011/695/EU, ‘an extension until 23 December 2018’. By decision of 20 August 2018, the Hearing Officer (13) set 22 October 2018 as the revised deadline.

(33)

Qualcomm submitted its written response to the SSO (the ‘SSO Response’) on that revised deadline date.

6.3. Second oral hearing and related matters

(34)

In the SSO Response, Qualcomm requested an oral hearing.

(35)

This second oral hearing was initially scheduled to take place on Tuesday, 11 December 2018. On Friday, 7 December 2018, Qualcomm sought the postponement of this hearing until early January 2019. In the light of Qualcomm’s specific reasons, I rescheduled the hearing to 10 January 2019. On 19 December 2018, the Hearing Officer rejected Qualcomm’s subsequent request to have the hearing moved to the following week in January 2019. (14) Accordingly, the second oral hearing took place on 10 January 2019.

(36)

On 25 April 2019, Qualcomm, at its own initiative, sent the Commission a ‘follow-up submission to the [second] oral hearing’.

7. Letter of facts

(37)

On 22 February 2019, the Commission sent Qualcomm a ‘letter of facts’. This communication (the ‘Letter of Facts’) was designed: (i) to provide Qualcomm with clarifications regarding certain elements set out in the SSO with which Qualcomm took issue in the SSO Response; (ii) to inform Qualcomm about pre-existing evidence that was not expressly relied on in the SO or the SSO; and (iii) to bring to Qualcomm’s attention minor updates to the SSO’s price-cost analysis.

7.1. Access to the file in connection with the Letter of Facts

(38)

The Letter of Facts included an annex containing ‘the accessible versions of all documents reflecting contacts with third parties registered on the case file since the adoption of the SSO [that] were not previously provided to Qualcomm, as well as two other documents’.

(39)

One of these ‘two other documents’ was the original (non-redacted) version of a spreadsheet in relation to which its author had, in advance of my decision mentioned in (30) above, dropped confidentiality claims (the ‘Full-version Spreadsheet’) but which, owing to a ‘transmission error’ had been sent to Qualcomm on 19 October 2018 in its previous redacted version rather than its full original form (see further (53) and (54) below).

7.2. Deadline for responding to the Letter of Facts

(40)

The initial deadline for responding to the Letter of Facts was 12 March 2019. By letter to Qualcomm dated 8 March 2019, DG Competition changed this deadline to Friday, 22 March 2019‘as a matter of courtesy’, even though it considered the arguments in Qualcomm’s extension request, dated 1 March 2019, to be unmeritorious.

(41)

By email dated 12 March 2019, I rejected Qualcomm’s request of 8 March 2019 for an extension until 9 April 2019 of the time limit for responding to the Letter of Facts.

(42)

DG Competition subsequently acceded to Qualcomm’s request, dated 19 March 2019, to be allowed to provide its comments on the Letter of Facts by Monday, 25 March 2019. Qualcomm provided these comments on Sunday, 24 March 2019.

7.3. Rejection of Qualcomm’s request for a third oral hearing

(43)

In its comments on the Letter of Facts. Qualcomm contended, among other things, that the Letter of Facts was ‘not, in substance, a letter of facts’ but instead ‘an additional statement of objections’. In a footnote in these comments, Qualcomm accordingly requested another oral hearing.

(44)

By email to Qualcomm dated 26 April 2019, I rejected that contention and that request. (15)

8. Procedural issues raised by Qualcomm

8.1. Alleged inadequacy of time limits

8.1.1. Time limit for responding to the SO

(45)

The SO Response complained, among other things, that the extended time limit for responding to the SO in writing was inadequate to ensure that the SO Response was completed to Qualcomm’s satisfaction. In particular, Qualcomm argued that it was disproportionate and contrary to fundamental principles of EU law such as the rights of defence and equality of arms for Qualcomm to be required to prepare its defence in what the SO Response described as the ‘amount of additional time allowed by the Hearing Officer to respond to the SO once he ‘restarted the clock’ on 13 July 2016’.

(46)

As emerges from (9) to (14) above and contrary to what the SO Response sought to suggest, Qualcomm ultimately had over eight months in which to prepare the SO Response. This appears to have been ample time for Qualcomm to exercise effectively its right to be heard in writing, (16) bearing in mind the need to avoid undue delay in proceedings (17) and irrespective of the fact that, for over six of these months, Qualcomm was also preparing its response to a separate statement objections, adopted on the same day as the SO, in Case AT.40220 Qualcomm (exclusivity).

(47)

The SO Response’s suggestion of an improper and ‘disproportionate’ time limit for the SO Response is premised on a suggestion that Qualcomm could only begin work on the SO Response once I lifted the suspension of the time limit for the SO Response on 13 July 2016. However, from receipt of the SO in early December 2015, Qualcomm could begin and advance its preparation of the SO Response.

8.1.2. Time limit for responding to the SSO

(48)

In the SSO Response, Qualcomm complained that the total amount of time the Commission granted for the SSO Response was ‘insufficient and disproportionate’. However, the SSO Response does not specify in what respects Qualcomm considered that the revised time limit for responding to the SSO Response was ‘insufficient’ or its rights of defence had been compromised. In the extension decision of 20 August 2018 mentioned in (32) above, the Hearing Officer already took into account the degree of changes to Case AT.39711 brought about by the SSO. As regards Qualcomm’s suggestion that the period granted for the SSO Response was ‘disproportionate’, the mere juxtaposition in the SSO Response of Qualcomm’s description of the steps it took to comply with that time limit and the time between the first oral hearing and the SSO is incapable of demonstrating what Qualcomm seemed to consider to be an unlawful lack of proportion between the time period for responding to the SSO and the time the Commission took for all the steps leading to the issuance of the SSO.

8.2. Access to the file

8.2.1. Access to the file following the SO

(49)

In the SO Response, Qualcomm complained about the extent of confidentiality redactions applied to the documents initially provided to Qualcomm for the purposes of access to the file. However, the SO Response acknowledged that, ‘[f]ollowing the Hearing Officer’s review [leading to my decision dated 13 July 2016], Qualcomm received additional less redacted versions of documents on the case file’. The SO Response also acknowledged in substance that this review resolved the issue of the legitimacy of the confidentiality claims maintained by third parties in respect of those documents.

8.2.2. Access to the file following the SSO

(50)

The SSO Response claimed that a breach of Qualcomm’s rights of defence resulting from what Qualcomm considered to be an insufficient period for responding to the SSO (see (48) above) was ‘exacerbated by numerous access to file issues, some of which remain[ed] unresolved’.

(51)

It is unclear what access-to-file issues Qualcomm considered to ‘remain unsolved’. The SSO Response did not raise any such issue that was not already dealt with by the Commission before Qualcomm’s submission of the SSO Response. The SSO Response in essence simply contained the bald assertion that, in combination with what Qualcomm considered to be an inadequate period for responding to the SSO, the access-to-file issues described in the SSO Response seriously and irreparably hampered Qualcomm’s ability to defend itself.

8.2.3. Belated transmission of the Full-version Spreadsheet

(52)

The SSO Response did suggest that the full version of what it termed a ‘crucial’ document was improperly made available to Qualcomm a mere single working day before the deadline for the SSO Response. However, the SSO Response’s substantive argument ostensibly based on this version of that document could just as easily have been based on the previously accessible redacted version of the same document.

(53)

Indeed, it later transpired that that argument was in fact based on that previously accessible redacted version. This is because the document in question is the Full-version Spreadsheet, which, as indicated in (39) above, was sent in its original form to Qualcomm only after the submission of the SSO Response.

(54)

According to Qualcomm’s response to the Letter of Facts, the ‘belated’ transmission of the Full-version Spreadsheet compromised Qualcomm’s rights of defence. However, given that Qualcomm at no stage made submissions specifically related to the information that was redacted in the previously accessible version of this document, this claim is unsubstantiated.

8.3. Claim of a failure to conduct a proper and timely investigation

(55)

The SSO Response claims that the Commission’s investigation has been mismanaged and unusually long, thereby affecting Qualcomm’s ability properly to defend itself.

(56)

The reasonableness of the duration of an administrative proceeding falls to be appraised, if necessary, by the EU courts in the light of the particular circumstances of each case and, in particular, the background to the case, the various procedural stages followed, the complexity of the case, its importance for the various parties involved, as well as the conduct of the party concerned and the Commission. (18) However, this list of criteria is not exhaustive and the assessment of the reasonableness of the period in question does not require a systematic examination of the circumstances of the case in the light of each of them where the duration of the proceedings appears justified in the light of one of them. The purpose of those criteria is to determine whether the time taken in the handling of a case is justified. Thus, the complexity of the case or the dilatory conduct of the party concerned may be deemed to justify a duration which might at first sight appear too long. (19)

(57)

For present purposes, it does not appear to me that the length of proceedings has prejudiced the effective exercise of Qualcomm’s rights of defence. Among other things, the case has been technical and relatively complex from the start. As emerges from (2) to (4) above, the complaint made to the Commission in 2009 and revised in 2010 alleged abusive conduct by Qualcomm in various respects, but it was not until 2012 that suspected predatory pricing moved to the fore of the complainant’s allegations and of the Commission’s investigation. Qualcomm’s complaints that the passage of time has made its defence more difficult are put into perspective by the fact that, as the General Court has already noted in the context of Case AT.39711 (see (22) above), because Qualcomm had been in receipt of requests for information under Article 18(2) and (3) of Regulation (EC) No 1/2003 since 7 June 2010, it was incumbent on Qualcomm, at least from that date, to act diligently to take all appropriate measures in order to preserve such evidence as might reasonably be available to it. (20) In any event, the passage of time does not appear to have unduly hindered Qualcomm from running its defence. For example, the criticisms made by Qualcomm in exercise of its right to be heard in respect of the SO led the Commission to refine its case, as it is permitted to do by relevant case-law (see (59) below). In turn, drawing up the SSO and hearing Qualcomm’s response to it inevitably added to the duration of proceedings, without rendering that duration unreasonable.

8.4. Claims of prosecutorial bias and an unfair investigation

(58)

Asserting that Qualcomm had legitimate expectations that (i) the Commission would have completed its fact-finding by the time of the SO and (ii) the Commission would terminate its investigation after the first oral hearing, the SSO Response alleged that the SSO was ‘the product of manifest prosecutorial bias’. According to Qualcomm, the Commission engaged in a ‘fishing expedition’ after the first oral hearing when it should have ended its investigation after Qualcomm had undermined the SO’s case. In a similar vein, (i) Qualcomm’s response to the Letter of Facts asserts that the case against Qualcomm is the product of a biased and unfair investigation and (ii) Qualcomm’s ‘follow-up submission’ mentioned in (36) above casts aspersions on the objectivity of the second oral hearing and the investigation in Case AT.39711 more generally.

(59)

Qualcomm’s claims of legitimate expectations are unsubstantiated. In addition, the mere fact that the Commission, in the light in particular of matters raised by Qualcomm in response to the SO, conducted further investigations after the first oral hearing is not of itself such as to indicate prosecutorial bias. Indeed, it emerges from case-law that, since the Commission must take into account the factors emerging from the administrative procedure in order either to abandon such objections as have been shown to be unfounded or to amend and supplement its arguments, both in fact and in law, it is entitled to amend (aspects of) its case, if necessary by way of a supplementary statement of objections. (21) It should also be borne in mind that when a case is, like Case AT.39711, based on formal complaints, it cannot simply be dropped without rejecting these complaints.

8.5. Claims that the SSO is a misuse of the Commission’s powers and a breach of the duty of sound administration

(60)

According to the SSO Response, since the SSO ‘is the product of mismanagement and prosecutorial bias’, it is ‘a misuse of the Commission’s investigatory powers’. Similarly, Qualcomm alleged that the Commission has breached its duty of sound administration within the meaning of Article 41 of the Charter, since the SSO is evidence of maladministration and the result of prosecutorial bias.

(61)

It emerges from (45) to (59) above that Qualcomm’s claims of maladministration and prosecutorial bias are unconvincing. These subsidiary claims are thus likewise unconvincing.

Concluding remarks

(62)

In the light of Qualcomm’s submissions, the Draft Decision, among other things, holds Qualcomm liable for an infringement of shorter duration than that of the infringement alleged in the SO.

(63)

In accordance with Article 16 of Decision 2011/695/EU, I have examined whether the Draft Decision deals only with objections in respect of which Qualcomm has been afforded the opportunity of making known its views. I conclude that it does.

(64)

Overall, I consider that the effective exercise of procedural rights has been respected in this case.

Brussels, 16 July 2019.

Joos STRAGIER

(1) Pursuant to Articles 16 and 17 of Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings (OJ L 275, 20.10.2011, p. 29) (‘Decision 2011/695/EU’).

(2) These are ‘baseband chipsets’ (components contributing to mobile connectivity in cellular communications devices) that implement and comply with the ‘Universal Mobile Telecommunications System’ standard of cellular communications technology.

(3) In May 2011, Icera was acquired by and became a wholly owned subsidiary of Nvidia Inc. (‘Nvidia’), which consequently assumed Icera’s position as a complainant.

(4) Mr Wouter Wils acted as Hearing Officer in this case until 3 December 2015.

(5) The Hearing Officer extended the time limit concerned to 18 November 2013. On a subsequent request from Qualcomm, DG Competition further extended this time limit to 2 December 2013.

(6) Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty, OJ L 1, 4.1.2003, p. 1 (‘Regulation (EC) No 1/2003’).

(7) Commission Regulation (EC) No 773/2004 of 7 April 2004 relating to the conduct of proceedings by the Commission pursuant to Articles 81 and 82 of the EC Treaty (OJ L 123, 27.4.2004, p. 18) as amended by Commission Regulation (EU) 2015/1348 of 3 August 2015 (OJ L 208, 5.8.2015, p. 3) (‘Regulation (EC) No 773/2004’).

(8) As required by, in particular, the judgment of 24 June 1986 in Akzo Chemie v Commission, 53/85, EU:C:1986:256, paragraphs 30 and 31.

(9) Nvidia/Icera also took part in that hearing.

(10) By means of a non-confidential summary of the SO.

(11) Order of 12 July 2017, Qualcomm and Qualcomm Europe v Commission, T-371/17 R, EU:T:2017:485.

(12) Judgment of 9 April 2019, Qualcomm and Qualcomm Europe v Commission, T-371/17, EU:T:2019:232.

(13) In my absence, Mr Wils took this decision.

(14) In my absence, Mr Wils took this decision.

(15) Unlike a (supplementary) statement of objections, a letter of facts does not trigger for its addressees an entitlement to develop their written response arguments at an oral hearing.

(16) By 8 July 2016, Qualcomm had received in instalments access to all of the additional access-to-file material to which, in my decision dated 13 July 2016, I considered Qualcomm was entitled.

(17) This need is among the matters to which, according to Article 9(1) of Decision 2011/695/EU, the Hearing Officer is to have regard when deciding whether an extension of a time limit for responding in writing to a statement of objections is necessary.

(18) See to that effect, among others, judgments of 15 October 2002, Limburgse Vinyl Maatschappij and Others v Commission, C-238/99 P, C-244/99 P, C-245/99 P, C-247/99 P, C-250/99 P to C-252/99 P and C-254/99 P, EU:C:2002:582, paragraph 187, and of 8 September 2016, Merck v Commission, T-470/13, EU:T:2016:452, paragraph 471.

(19) See judgment of 15 October 2002, Limburgse Vinyl Maatschappij and Others v Commission, C-238/99 P, C-244/99 P, C-245/99 P, C-247/99 P, C-250/99 P to C-252/99 P and C-254/99 P, EU:C:2002:582, paragraph 188.

(20) See judgment of 9 April 2019, Qualcomm and Qualcomm Europe v Commission, T-371/17, EU:T:2019:232, paragraph 136.

(21) See to that effect, in particular, judgments of 14 July 1972, Bayer v Commission, 51/69, EU:C:1972:72, paragraph 11; of 11 November 1981, IBM v Commission, 60/81, EU:C:1981:264, paragraph 18; of 21 September 2006, JCB Service v Commission, C-167/04 P, EU:C:2006:594, paragraphs 100 and 101; and of 13 July 2011, Polimeri Europe v Commission, T-59/07, EU:T:2011:361, paragraph 73.

6.11.2019

Official Journal of the European Union

C 375/30

Publication of an application for approval of non-minor amendments to a product specification pursuant to Article 50(2)(a) of Regulation (EU) No 1151/2012 of the European Parliament and of the Council on quality schemes for agricultural products and foodstuffs

(2019/C 375/09)

This publication confers the right to oppose the amendment application pursuant to Article 51 of Regulation (EU) No 1151/2012 of the European Parliament and of the Council (1) within three months of the date of this publication.

APPLICATION FOR APPROVAL OF NON-MINOR AMENDMENTS TO THE PRODUCT SPECIFICATION FOR A PROTECTED DESIGNATION OF ORIGIN OR PROTECTED GEOGRAPHICAL INDICATION

Application for approval of amendments in accordance with the first subparagraph of Article 53(2) of Regulation (EU) No 1151/2012

‘Queso de Valdeón’

EU No: PGI-ES-0267-AM01 — 5.7.2018

PDO ( )

PGI (X)

1. Applicant group and legitimate interest

Name of the group: Consejo Regulador de la Indicación Geográfica Protegida ‘Queso de Valdeón’ (Regulatory Council for the Protected Geographical Indication ‘Queso de Valdeón’)

Address: C/El Cantón, s/n, 24915 Posada de Valdeón (León), Spain

Tel. +34 987740514

Email: quesosv@quesospicosdeeuropa.com

The applicant group represents the collective interests of the producers of ‘Queso de Valdeón’. It is the legitimate interested party in the amendment application for the Specification of the Protected Geographical Indication ‘Queso de Valdeón’ and is also responsible for its protection.

2. Member State or third country

Spain

3. Heading in the product specification affected by the amendment(s)

☐	Name of product

☒	Description of product

☐	Geographical area

☒	Proof of origin

☒	Method of production

☐

Link

☒

Labelling

☒	Other [inspection body]

4. Type of amendment(s)

☒	Amendments to the product specification of a registered PDO or PGI not to be qualified as minor in accordance with the third subparagraph of Article 53(2) of Regulation (EU) No 1151/2012

☐	Amendments to the product specification of a registered PDO or PGI for which a Single Document (or equivalent) has not been published and which cannot be qualified as minor in accordance with the third subparagraph of Article 53(2) of Regulation (EU) No 1151/2012

5. Amendment(s)

Amendment 1

‘Description of product’ section

Grounds for the amendment

In the first paragraph, a more detailed description of the cheese has been given in order to provide a better characterisation of the product.

This amendment affects Section 4.2 of the original Single Document or the equivalent thereof.

Current wording of the specification

The cheeses covered by the ‘Queso de Valdeón’ Protected Geographical Indication are full-fat blue cheeses made using cows’ milk or a mixture of cows’ milk and sheep’s and/or goats’ milk, marketed whole or creamed.

Proposed amendment

The cheeses covered by the ‘Queso de Valdeón’ Protected Geographical Indication are full-fat blue cheeses made from raw or pasteurised cows’ milk or a mixture of cows’ milk and sheep’s and/or goats’ milk, using lactic acid and enzymatic coagulation. The cheeses made from raw milk are ripened for at least two months, while the cheeses made from pasteurised milk are ripened for one month. They are marketed either whole or creamed.

Amendment 2

‘Description of product’ section

Grounds for the amendment

The weight range for the whole cheeses has been increased slightly (from between 0,5 kg and 3 kg to between 0,2 kg and 3 kg). This increase does not affect the characteristics of ‘Queso de Valdeón’. This amendment has been made because consumers’ eating habits have been changing as a result of sociodemographic shifts in the population, which have transformed consumption patterns. One pattern that has changed is the demand for food in smaller formats, due to the reduction in family sizes.

This paragraph also states that the cheese may be marketed in portions.

On the other hand, the rise in the use of ingredients with a quality mark by various establishments (such as hospitality establishments, retail chains and industrial plants) has led to an increase in demand for larger containers of non-solid ingredients such as the creamed cheese covered by this PGI. The permitted net content of the containers for creamed cheese has therefore increased to 6 kg.

Because of this demand, the group has to find new ways to market its products, adapting to market trends and requirements but without reducing the quality of the products.

These amendments affect Section 4.2 of the original Single Document or the equivalent thereof.

Current wording of the specification

Characteristics of the whole cheeses

—

Morphological characteristics

—	cylindrical, with slightly concave upper and lower surfaces,

—	maximum height of 15 cm and maximum diameter of 25 cm,

—	weight from 0,5 kg to 3 kg.

Characteristics of the creamed cheese

—	Morphological characteristics

It is sold in containers with a net content of between 20 g and 1,5 kg.

The creamed cheese is perfectly homogeneous, with no rind whatsoever.

Proposed amendment

Characteristics of the whole cheeses

—

Morphological characteristics

—	cylindrical, with slightly concave upper and lower surfaces,

—	maximum height of 15 cm and maximum diameter of 25 cm,

—	weight from 0,2 kg to 3 kg.

—	The cheese may also be marketed in portions.

Characteristics of the creamed cheese

—	Morphological characteristics

It is marketed in containers with a net content of between 20 g and 6 kg. Containers with a net content of between 1,5 kg and 6 kg are only intended for catering establishments, retail chains and industrial plants.

The creamed cheese is perfectly homogeneous, with no rind whatsoever.

Amendment 3

‘Description of product’ section, 1.— Characteristics of the whole cheeses

Grounds for the amendment

The last paragraph in the section, which indicates the various ways in which the cheese can be sold, has been deleted. It seems more appropriate to move this information to Section 3.2, where it is stated that the product may be marketed in portions, and to Section 3.6 on labelling, which describes the option to market the product in portions, provided that it is packaged and its origin can be identified.

The description of the process of cutting the product into portions and the minimum portion weight has been deleted, due to consumer demand for even smaller portions and because the minimum portion weight of 250 g and the cutting method do not affect the quality of ‘Queso de Valdeón’.

This amendment affects Section 4.2 of the original Single Document or the equivalent thereof.

Text deleted from the specification

The whole cheeses can also be marketed in portions obtained by making transverse cuts along the radius of the cylinder. These portions must retain their natural rind and weigh at least 250 g.

Amendment 4

‘Evidence that the product originates from the geographical area’ section

Grounds for the amendment

This section has been completely rewritten to improve its readability and adapt it to the current regulations.

References to authorisation or inspection by the Regulatory Council have been deleted so as not to restrict the free movement of goods and the freedom to provide services.

This amendment affects Section 4.4 of the original Single Document or the equivalent thereof.

Current wording of the specification

—	The cheese must come only from cheese dairies and ripening establishments registered with the Regulatory Council.

—	An initial evaluation must be carried out on cheese dairies and ripening establishments before they are registered with the Regulatory Council and evaluations must then be carried out on a regular basis to ensure that they still fulfil the requirements for registration.

—	The cheese must be produced in accordance with the method described in this specification.

—	Only cheese that has passed all the checks laid down may be marketed with the guarantee of origin certified by the Regulatory Council’s approval mark.

—	The number of secondary labels issued by the Regulatory Council to each cheese dairy must tally with the volumes of milk processed and the types and quantities of cheese produced.

—	The regular checks and evaluations must include analyses of the milk when it is placed in the curdling vats, inspection of the method of production, checks on documentation, checks on stocks, sampling and tests on the product.

—	Where appropriate, the Regulatory Council will apply the penalties laid down in its rules where irregularities are detected.

—	The Regulatory Council has a register showing the quantities of numbered labels, membership documents and certificates issued and to whom they are issued.

Proposed amendment

—	The cheese must come only from registered cheese dairies and ripening establishments which are subject to checks.

—	An initial evaluation must be carried out on cheese dairies and ripening establishments before they are registered and evaluations must then be carried out on a regular basis to ensure that they still fulfil the requirements for registration.

—	The cheese must be produced in accordance with the method described in Section E of this specification.

—	Only cheese that has passed all the checks laid down may be marketed with the guarantee of origin certified by the label identifying the protected geographical indication.

—	The regular checks and evaluations must include analyses of the milk when it is placed in the curdling vats, inspection of the method of production, checks on documentation, checks on stocks, sampling and tests on the product.

—

All operators must be able to identify:

(a)	the supplier, the quantity and the origin of all milk received;

(b)	the recipient, quantity and destination of all cheeses labelled;

(c)	the correlation between the milk received, as referred to in point (a), and each batch of cheese labelled, as referred to in point (b).

Amendment 5

‘Method of production’ section

Grounds for the amendment

The minimum ripening period for cheeses made with pasteurised milk has been changed from one and a half months to one month. This decrease, which is adapted to the reduction in weight of the whole cheeses, affects neither the natural ripening processes nor the final characteristics of the cheese.

This amendment affects Section 4.5 of the original Single Document or the equivalent thereof.

Current wording of the specification

The cheeses are ripened for a minimum of two months for cheeses made from raw milk and one and a half months for cheeses made from pasteurised milk.

Proposed amendment

The cheeses are ripened for a minimum of two months for cheeses made from raw milk and one month for cheeses made from pasteurised milk.

Amendment 6

‘Method of production’ section

Grounds for the amendment

The text concerning the labels and secondary labels for the cheeses has been deleted, as this information is better suited to the section on labelling. Please see the text included in the amendment to the section on labelling, which indicates the new location of that text in the specification.

This amendment does not affect any sections of the original Single Document or the equivalent thereof.

Text deleted from the ‘Method of production’ section of the product specification

Labels and secondary labels must be affixed to the cheeses and the containers of creamed cheese immediately before they are placed on the market.

Amendment 7

‘Labelling’ section

Grounds for the amendment

This section now contains all references to the possibility of marketing the cheese whole, creamed or in portions, provided that it is packaged and its origin can be identified.

The requirements regarding authorisation and checking by the Regulatory Council have been deleted, so as not to restrict freedom of movement.

The content of some of the mandatory terms has been laid down, and the logo of the protected geographical indication, which does not appear in the current specification, has been included.

The text removed from the ‘Method of production’ section referred to in amendment 6 has been included.

The text has been rearranged to make it easier to understand.

This amendment affects Section 4.8 of the original Single Document or the equivalent thereof.

Current wording of the specification

The words ‘Indicación Geográfica Protegida ‘Queso de Valdeón’’ (‘Queso de Valdeón’ protected geographical indication) and the logo of the Regulatory Council must be visible on each registered company’s commercial labels. These must be authorised by the Regulatory Council.

They must also be accompanied by the conformity mark of the protected geographical indication, which is the numbered secondary label issued by the Regulatory Council. This conformity mark guarantees the product’s traceability.

The numbered secondary label must be attached in such a way that it cannot be reused. It must include the words ‘Consejo Regulador de la Indicación Geográfica Protegida ‘Queso de Valdeón’’ (Regulatory Council for the Protected Geographical Indication ‘Queso de Valdeón’) and the logo of the protected geographical indication.

The Regulatory Council must set out in its quality manual the supplementary rules on the use of the protected geographical indication’s conformity mark with regard to its size and location on the product.

Proposed amendment

‘Queso de Valdeón’ may be marketed whole, in portions or creamed, provided that it is packaged and its origin can be identified.

The commercial labels used to market the protected product must bear the name and logo of the protected geographical indication.

Cheese marketed whole or creamed must bear a secondary label, each with its own serial number. This secondary label must display the name and logo of the protected geographical indication.

The labelling for cheese marketed in portions must bear the name and logo of the protected geographical indication, along with references that allow the origin of the cheese to be identified.

Labels and secondary labels must be affixed to the cheeses and the containers of creamed cheese before they are placed on the market.

The designation’s logo is:

Amendment 8

‘Inspection body’ section

Grounds for the amendment

The inspection body is the Instituto Tecnológico Agrario de Castilla y León (Castile and Leon Institute of Agricultural Technology), which is the competent authority.

This amendment affects Section 4.7 of the original Single Document or the equivalent thereof.

Current wording of the specification

Name: Consejo Regulador de la Indicación Geográfica Protegida ‘Queso de Valdeón’ [Regulatory Council for the Protected Geographical Indication ‘Queso de Valdeón’]

Address: C/El Cantón, s/n. 24915 – Posada de Valdeón (León)

Telephone: +34 987 740 514

Fax +34 987 740 568

The Regulatory Council for the Protected Geographical Indication ‘Queso de Valdeón’ must be able to comply with standard EN 45011 ‘General criteria for certification bodies carrying out product certification’.

Proposed amendment

Instituto Tecnológico Agrario de Castilla y León [Castile and Leon Institute of Agricultural Technology]

Ctra. de Burgos Km 119. Finca Zamadueñas

47071 Valladolid (Spain)

Telephone: +34 983 412 034

Fax +34 983 412 040

Email: controloficial@itacyl.es

Verification of compliance with the conditions described in the product specification for the Protected Geographical Indication ‘Queso de Valdeón’ is the responsibility of the Instituto Tecnológico Agrario de Castilla y León, through the Subdirectorate for Food Quality and Promotion, in accordance with Article 139 of Law No 1/2014 of 19 March 2014 on Castile and Leon Agriculture.

SINGLE DOCUMENT

‘Queso de Valdeón’

EU No: PGI-ES-0267-AM01 — 5.7.2018

PDO ( )

PGI (X)

1. Name

‘Queso de Valdeón’

2. Member State or third country

Spain

3. Description of the agricultural product or foodstuff

3.1. Type of product

Class 1.3. Cheeses

3.2. Description of product to which the name in (1) applies

Characteristics of the whole cheeses

Morphological characteristics

Cylindrical, with slightly concave upper and lower surfaces.

Maximum height of 15 cm and maximum diameter of 25 cm.

Weight from 0,2 kg to 3 kg.

The cheese may also be marketed in portions.

Physical and chemical characteristics

Fat content equal to or greater than 45 % in dry matter.

Minimum moisture content of 30 %.

Salt content of less than 3,5 %.

Organoleptic characteristics

The rind is natural, thin, tender and of yellowish colour with grey tones.

The cut is smooth, with numerous evenly distributed irregular eyes of various sizes and a greenish-blue appearance.

The body of the cheese is of ivory to cream colour and is shiny with a faint aureole depending on the degree of ripeness.

It has low elasticity, brittleness and firmness and medium flexibility, melts in the mouth, has good solubility and reduced adhesion.

Its aroma is lightly acidic or lactic when the cheese is semi-ripe, with a very specific scent of mould.

As the cheese continues to ripen, the lactic scent disappears, the aroma becomes more persistent and intense, and spicy, fruity and even aromatic scents develop, although the scent of mould remains.

The flavour is intense, salty and piquant, lightly ardent, and becomes more pronounced as ripening progresses and when goats’ or sheep’s milk is used.

The flavour is persistent, particularly in riper cheeses.

Characteristics of the creamed cheese

Morphological characteristics

It is sold in containers with a net content of between 20 g and 6 kg.

Containers with a net content of between 1,5 kg and 6 kg are only intended for catering establishments, retail chains and industrial plants.

The creamed cheese is perfectly homogeneous, with no rind whatsoever.

Physical and chemical characteristics

Fat content equal to or greater than 45 % in dry matter.

Minimum moisture content of 30 %.

Salt content of less than 3,5 %.

Organoleptic characteristics

The cream is ivory white with greenish-blue tones, and is not very shiny.

The cream is sticky, increasingly so as it gets warmer, with a medium, almost buttery consistency.

The aroma is lightly acidic or lactic, with a persistent and intense scent of mould.

The cream melts in the mouth, with good solubility and medium adhesion.

The flavour is intense and piquant, and intensifies when goats’ or sheep’s milk is used.

The cheese has an intense salty flavour, with a light ardent sensation.

The taste lingers in the mouth.

3.3. Feed (for products of animal origin only) and raw materials (for processed products only)

Once the milk from different species has been mixed together, the minimum analytical levels in the curdling vat are as follows:

—	fat: 3,5 %

—	protein: 3,1 %

—	dry matter: 12,0 %

3.4. Specific steps in production that must take place in the defined geographical area

The cheese’s production process, including ripening, must take place in the defined geographical area.

3.5. Specific rules concerning slicing, grating, packaging, etc. of the product the registered name refers to

—

3.6. Specific rules concerning labelling of the product the registered name refers to

‘Queso de Valdeón’ may be marketed whole, in portions or creamed, provided that it is packaged and its origin can be identified.

The commercial labels used to market the protected product must bear the name and logo of the protected geographical indication.

Cheese marketed whole or creamed must bear a secondary label, each with its own serial number. This secondary label must display the name and logo of the protected geographical indication.

The labelling for cheese marketed in portions must bear the name and logo of the protected geographical indication, along with references that allow the origin of the cheese to be identified.

Labels and secondary labels must be affixed to the cheeses and the containers of creamed cheese before they are placed on the market.

The designation’s logo is:

4. Concise definition of the geographical area

The municipality of Posada de Valdeón (León).

5. Link with the geographical area

The link between the geographical area and the product is based on various elements such as the area’s climate characteristics, the traditional production method and the product’s reputation.

‘Queso de Valdeón’ gets its name from the municipality of Posada de Valdeón, located entirely within the Picos de Europa national park. The area is surrounded by substantial natural barriers, which perfectly define it. These barriers are the Panderruedas Pass (1 450 metres), the Pontón Pass (1 311 metres) and the Pandetrave Pass (1 562 metres).

Between the high mountains lies a deep depression with an average altitude of 650 m.

The particular climate of the defined area can be classed as a high-mountain climate. Winters are long and cold, with frequent frosts (more than 100 days between November and April) and abundant precipitation, often in the form of snow. Summers are short and cool (minimum average temperatures around 5-6 °C, maximum 18 °C). Precipitation is abundant, although of shorter duration than the rest of the year. Relative humidity in summer remains high because of fog and mist.

Average annual precipitation is 1 100 mm in the valley and 1 800 mm on the mountain peaks.

Historically, these factors, together with the presence of caves in the area, have resulted in the development of certain special characteristics in relation to the ripening of the cheeses. These characteristics originate from the raw material available in the area due to its mountainous nature.

The specific microclimate in this area and the traditional production practices favour the development of the inoculated microbial flora that is characteristic of these cheeses.

The specific nature of the product is clear from the outset: it is a blue cheese, which gives it a different appearance to other cheeses produced in the area, but this characteristic is particularly noticeable when cutting the cheese, which contains greenish-blue eyes of variable size. It is matured in cold temperatures and high humidity, replicating the conditions in the ripening caves in the area. This is clear from the presence of visible mould and from the cheese’s aroma and flavour.

Various texts from the 19th and early 20th centuries attest to the artisanal production of the protected cheeses under these production conditions. The Valdeón valley is currently home both to production processes similar to those used for centuries and to processes that have incorporated new technology. However, the basic principles of the production process mentioned in the testimonies referred to here are still clearly applied.

Cheese production in the Valdeón valley goes back to pre-Roman times, when goats’ milk was used as the raw material.

The first written references to cheese production in the Valdeón valley date back to the mid-19th century. Pascual Madoz, in his Diccionario (1845-1859), refers to cheese production and the size of the goat population in the villages of the Valdeón valley.

During the second half of the 19th century, cheese production was an important activity in the Valdeón valley. When livestock was grazed in high pasture during the summer, milk was made into cheese directly in the shepherds’ huts or carried down to the valley. The Count of Saint-Saud testified to this in 1892: ‘The robust young women of Valdeón climb up there morning and evening, wearing their clogs with three wooden blocks attached to the sole, carrying a goat-skin bag in which they carry, on the way up, their food and, on the way down, wineskins full of milk from the herds on the mountain.’

There are numerous testimonies to the fact that the production of blue cheese in the Valdeón valley to be sold at regional markets was an important activity from the beginning of the 20th century. Of these testimonies, that of Marmeta Pérez Marcos is particularly relevant: ‘My entire family is from Caín, and I too was born there in 1922. My parents inherited the method from their ancestors, and until very recently I did it the same way. We would add kid goat rennet and curdle between 50 and 60 litres of milk a day, making cheeses weighing a little over a kilogram. We would take them by donkey to market in Puente Almuhey, which is more than 50 km away. Just after the War, the cheese would sell for 36 pesetas per kilo. Only after the War had ended did we start going to Arenas de Cabrales to sell cheeses. The cheeses were made from goats’ and sheep’s milk and a tiny amount of cows’ milk, to which we would add rennet. When a curd had formed, we would put it in the “arno” (cheese mould), then place it in a “duerna” (trough), where the whey would be released over a couple of days. We would then remove them and put them on boards in the farm huts to dry. After 15 days they were ready. However, in order to obtain good, blue cheeses, we would take them to the caves. After a month, the cheese was excellent.’

Reference to publication of the specification

(the second subparagraph of Article 6(1) of this Regulation)

http://www.itacyl.es/documents/20143/342640/2019_02_8+Pliego+Q+Valde%C3%B3n+.pdf/fbc379cc-8473-8829-744b-6baf02ed5ba7

(1) OJ L 343, 14.12.2012, p. 1.

		ISSN 1977-091X
Official Journal of the European Union		C 375

English edition	Information and Notices	Volume 62 6 November 2019

Contents		page
	II Information
	INFORMATION FROM EUROPEAN UNION INSTITUTIONS, BODIES, OFFICES AND AGENCIES
	European Commission
2019/C 375/01	Communication from the Commission Commission Notice on the calculation of the cost of capital for legacy infrastructure in the context of the Commission’s review of national notifications in the EU electronic communications sector ( 1 )	1
2019/C 375/02	Non-opposition to a notified concentration (Case M.9544 — Brookfield/Johnson Controls Autobatterie) ( 1 )	12

	IV Notices
	NOTICES FROM EUROPEAN UNION INSTITUTIONS, BODIES, OFFICES AND AGENCIES
	European Commission
2019/C 375/03	Euro exchange rates — 5 November 2019	13
2019/C 375/04	Opinion of the Advisory Committee on restrictive practices and dominant positions at the first meeting on 5 July 2019 concerning a draft decision in Case AT.39711 — Qualcomm (predation) Rapporteur: Spain	14
2019/C 375/05	Opinion of the Advisory Committee on restrictive practices and dominant positions at the second meeting on 15 July 2019 concerning a draft decision in Case AT.39711 — Qualcomm (predation) Rapporteur: Spain	15
2019/C 375/06	Final Report of the Hearing Officer Qualcomm (predation) (AT.39711)	16
2019/C 375/07	Summary of Commission Decision of 18 July 2019 relating to a proceeding under Article 102 of the Treaty on the Functioning of the European Union and Article 54 of the EEA Agreement (Case AT.39711 — Qualcomm (predation)) (notified under document C(2019) 5361) ( 1 )	25

	Currency	Exchange rate
USD	US dollar	1,1109
JPY	Japanese yen	120,93
DKK	Danish krone	7,4712
GBP	Pound sterling	0,86113
SEK	Swedish krona	10,6943
CHF	Swiss franc	1,1009
ISK	Iceland króna	137,70
NOK	Norwegian krone	10,1525
BGN	Bulgarian lev	1,9558
CZK	Czech koruna	25,536
HUF	Hungarian forint	329,84
PLN	Polish zloty	4,2643
RON	Romanian leu	4,7583
TRY	Turkish lira	6,3807
AUD	Australian dollar	1,6079
CAD	Canadian dollar	1,4584
HKD	Hong Kong dollar	8,7040
NZD	New Zealand dollar	1,7339
SGD	Singapore dollar	1,5078
KRW	South Korean won	1 284,34
ZAR	South African rand	16,3876
CNY	Chinese yuan renminbi	7,7734
HRK	Croatian kuna	7,4486
IDR	Indonesian rupiah	15 516,50
MYR	Malaysian ringgit	4,5886
PHP	Philippine peso	56,012
RUB	Russian rouble	70,4334
THB	Thai baht	33,544
BRL	Brazilian real	4,4655
MXN	Mexican peso	21,3072
INR	Indian rupee	78,5290

	V Announcements
	PROCEDURES RELATING TO THE IMPLEMENTATION OF COMPETITION POLICY
	European Commission
2019/C 375/08	Prior notification of a concentration (Case M.9331 — Danaher/GE Healthcare Life Sciences Biopharma) ( 1 )	28
	OTHER ACTS
	European Commission
2019/C 375/09	Publication of an application for approval of non-minor amendments to a product specification pursuant to Article 50(2)(a) of Regulation (EU) No 1151/2012 of the European Parliament and of the Council on quality schemes for agricultural products and foodstuffs	30