Common Ownership under EU Competition Law: Current State and Possible Developments

(1)

Faculty of Law

LLM European Competition Law and Regulation Master Thesis – 12 ECTS

2017/2018

Common Ownership under EU Competition Law: Current State and Possible Developments

Yannis Schlüter

(2)

1 Table of Contents

Abstract ... 2

Introduction ... 3

Part 1: The implications of common ownership for competition ... 4

I. Defining minority shareholding and common ownership ... 4

II. The reasons for common ownership ... 5

III. The theory of harm ... 6

IV. The role for competition law ... 12

Part 2: The current regulation of common ownership ... 12

I. Minority shareholding and common ownership ... 12

II. The EU Merger Regulation (EUMR) ... 13

1. Current Situation ... 13

2. Shortcomings and Issues ... 14

III. Article 101 TFEU ... 15

IV. Article 102 TFEU ... 18

Part 3: Possible Alternatives ... 21

I. Policy proposals ... 21

1. The Commission White Paper ... 21

2. Other proposals ... 23

3. Approaches in other jurisdictions ... 25

II. Possible regulatory approaches ... 26

1. Macro-regulatory approach ... 26

2. Micro-regulatory approach ... 27

3. Mixed approach ... 29

Conclusion ... 31

(3)

2

Abstract

A recent discussion in competition policy concerns the treatment of common ownership. Common ownership refers to situations in which a range of institutional investors own minority shareholdings in every company in a given market. These situations are very common in concentrated industries and recent economic evidence has alleged that they lead to anti-competitive effects. This thesis considers these effects and attempts to devise a regulatory system that could prevent them. To that end, it considers the current legislative framework as well as policy proposals that have been put forward. It also introduces possible other systems that could work towards resolving the issue. The challenge is to create a system that adequately addresses the problem, without however leading to disproportionate regulation. Three different systems will be considered, with differing degrees of intrusiveness. The first considers an outright ban of common ownership. The second requires competition authorities to take into account common ownership in the merger clearance process. The final approach provides for a jurisdictional rule that leads to an effects assessment. The advantages and shortcomings of the different regulatory approaches will be considered.

(4)

3

Introduction

Competition policy is a field that is in constant motion. It is inextricably linked to micro-economic developments in individual industries but also to macro-micro-economic developments such as the overall state of the economy. On a macro-economic level, competition law must consider not only the overall state of the economy but also new trends that range across industries and threaten the competitive arena that competition authorities seek to protect.

Discovering new threats and addressing them in a timely manner is important to safeguard competition. It is however equally important to prevent overregulation and limit oneself to what is necessary to address the threat. Unnecessary regulation could result in a disproportionate burden for market participants and could lead to higher prices or less innovation for consumers. Competition policy should address these issues, not create them. It is therefore important to ensure that competition policy limits itself to what is necessary when considering regulating a new trend.

These considerations should be kept in mind when discussing a development in competition policy that has recently been picked up by several economic studies. The focus of these studies lies on the amount of common ownership that is present in certain industries. In short, common ownership refers to situations where one investor has minority ownership interests in all companies in a given market. The assumption is that this has anti-competitive consequences, because firms with a common owner have less incentive to compete against each other. The situation is particularly difficult because evidence suggests that the common owners are big financial institutions or asset managers whose very business model of offering passive investment services is the source of the problem. Passive investing however can be very beneficial to consumers by lowering the costs of investing and enabling easy access to capital markets, ultimately benefiting companies and consumers equally.

The evidence of a new threat to competition means that there is a need to decide whether a revised regulatory approach is necessary and if so, how such a regulatory approach should be crafted. This thesis aims to do exactly that. However, before any possibility to regulate common ownership is introduced, a theory of harm must be provided, outlining why common ownership should be regulated in the first place. Such a theory of harm will be provided in the first part of this thesis. As a rule, additional regulation should only be used as a measure of last resort to prevent imposing unnecessary burden on companies and harm their economic prospects. Therefore, once a harm to competition has been established, this thesis will in a second part

(5)

4

consider whether the current regulation is fit to address the concerns. The current regulatory framework will be explained, and its effectiveness tested. This thesis takes the assumption that there is indeed a harm to competition and that it cannot be properly addressed by current regulatory instruments. In its third and final part, this thesis will thus provide possible alternatives to the current system, explaining the advantages and shortcomings of each of them. In this way, this thesis hopes to provide a clear and coherent approach to the issue of common ownership. The aim is to provide an overview of the issue and to provide a list of possible solutions to the problem, should it be established that there is a need for regulation. Such an overview is relevant, especially now that competition authorities around the world are picking up the issue.1 In the EU for example, Commissioner Vestager, in charge of Competition Policy, has in a recent speech expressed the need to investigate common ownership to establish possible anti-competitive effects.2_{In the light of such recent political interest, it is important to}

examine the possible choices that policy makers could take.

Part 1: The implications of common ownership for competition

I. Defining minority shareholding and common ownership

Before explaining the theory of harm behind common ownership, it is important to define the concept. Common ownership can be defined as a situation in which one or more owners of a company also own shares of one or more other companies in the same industry, which is usually characterised by a limited amount of market players.3_{In other words, common ownership refers}

to situations in which investors, usually big assets managing companies (such as BlackRock, Vanguard, State Street, Berkshire Hathaway…), own minority stakes in all competing companies in concentrated markets, such as airlines or banking. To take the example of airlines operating in the Unites States, each of the asset managers has an ownership interest of 2-5% in each of the airlines.4

It is important to distinguish common ownership from horizontal minority shareholding (hereinafter ‘minority shareholding’). The latter concerns a situation in which a company

1_{OECD Competition Committee, Hearing on Common ownership by institutional investors and its impact on}

competition - Summary of contributions (DAF/COMP/WD(2017)83, 2017)

2_{Margrete Vestager, ‘Competition in changing times’, FIW Symposium, Innsbruck, 16 February 2018}

3_{OECD Competition Committee, ‘The Competitive Effects of Common Ownership: Ten Points on the Current} State of Play - Note by Daniel P. O'Brien’ (DAF/COMP/WD(2017)97, 2017), page 2

4_{OECD Competition Committee, ‘Common Ownership and Competition: Facts, Misconceptions, and What to Do} About It - Note by Martin C. Schmalz’ (DAF/COMP/WD(2017)93, 2017), page 4

(6)

5

acquires a minority shareholding in a competitor. This should be distinguished from common ownership, where an investor acquires minority shareholdings in two or more competing companies. While both transactions concern the acquisition of minority shareholding, the circumstances and the theory of harm are very different. In short, minority shareholding could lead to coordination between companies, whereas common ownership is rather about reduced pressure from shareholders on companies’ management to act in a competitive manner.

Despite the differences, the two concepts should not be viewed in complete isolation. While the theory of harm might be different, the underlying mechanism remains the same as in both cases the problematic transaction is an acquisition of a minority shareholding. Since the transactional structure, in other words the creation, of both issues is similar, this thesis will also consider the solutions that have previously been put forward by academics and policy makers to address the issue of minority shareholding and see whether they could be fit, with certain adaptations, to regulate common ownership.

II. The reasons for common ownership

To understand the issues with regulating common ownership, it is important to understand what the underlying reason for the creation of common ownership is. Academic literature has linked the development of common ownership to the rise in so-called ‘index funds’ (also called Exchange Traded Funds’ or ‘ETFs’). The aim of such funds is to enable investors to invest passively. Simply put, ETFs are financial derivatives that attempt to represent the development of a benchmark index (such as the S&P 500, the DAX, or even an industry related index). ETFs are baskets, compiled by asset managers, such as BlackRock, Vanguard or State Street. These baskets are composed of stocks of every company represented in the index that the ETF wishes to represent. Investors can buy into these baskets, and with that money, the asset managers will buy new stocks of the companies represented in the benchmark index.

Certain benchmark indexes represent a certain industry.5 It follows that, in case there was an ETF representing such an index, the asset manager offering the ETF would have to buy stocks of every firm represented in the benchmark, which would mean that it would eventually hold a minority stake in every firm in this industry. This is especially so in concentrated industries, where, due to the limited number of competitors, levels of shareholding increase more readily as investors put money into the ETF.

(7)

6

ETFs have considerably gained in importance. Around 25% of all funds worldwide are now passive funds, totalling in almost US$ 5 trillion invested in ETFs.6 In fact, every day, US$ 1.8 billion is invested in ETFs.7 The reason for this impressive growth is that ETFs allow investors to invest in a diversified way at a very low cost, as transaction fees are much lower than for active funds, in which fund managers decide which investments are made by the fund. These are two strong advantages for investors, in fact the argument could be made that ETFs are a way for consumers, who would have never thought of investing due to the complicated nature of it, to have access to capital markets. At the same time, ETFs are also an advantage for companies, who have easier access to new capital through issuing equity without having to resort to debt.

The advantages that ETFs carry must be balanced against the possible threat to competition that will be addressed in the next section. It is however important to keep in mind the benefits and the importance of passive investing for consumers and companies alike when discussing the need for regulation of common ownership, as such regulation might have a direct impact on the way index funds function.

III. The theory of harm

Now that the term ‘common ownership’ has been defined and its origins have been discussed, it is time to turn to the actual topic of this thesis, namely the competitive threat of common ownership. From the outset, it should be noted that the theory of harm in relation to common ownership is still being developed. The economic evidence supporting it is still debated and further research is certainly necessary. The aim of this thesis is not to contribute to the development of the economic theory of harm. Instead, the current state of research will briefly be described, and points of contradiction will be pointed out. There are both empirical and theoretical arguments in favour of, or against, a competitive threat of common ownership.

The main theoretical argument in favour of the recognition of anti-competitive effects of common ownership is the idea that a common owner wants to increase his total rent, and not the individual profit of a single firm. For the common owner, it is better if all the companies it owns stakes in do well and competition between those firms is not too fierce. Common owners are not interested in lower prices, since in case one firm undercuts another firm, prices in the

6_{http://www.ey.com/Publication/vwLUAssets/ey-global-etf-survey-2017/$FILE/ey-global-etf-survey-2017.pdf} 7_{https://www.ft.com/content/09cb4a5e-e4dc-11e7-a685-5634466a6915}

(8)

7

market fall and the total rent of the common owner falls.8 In general, companies and their owners have to be careful about raising prices, as it might lead them to lose customers. However, in case of common ownership, the owner is less scared of higher prices due to customer recapture. In fact, the loss of customers of the firm raising prices will benefit the competitors of this firm who will capture these customers. Since the competitors are also owned by the common owner, this means that raised prices will lead to a higher market price and no loss of customers, and thus an increase in total rent.9 For example, when a passenger was forcibly removed from a United Airlines flight in April 2017 and this was widely publicized, there was no significant shareholder reaction. In fact, these shareholders actually profited from the incident, as they also owned stock of United Airline’s competitors, which went up more than United Airline’s stock went down.10

The above mentioned are convincing arguments why the common owners have an interest in an anti-competitive market. However, an incentive of the common owners is not enough proof in this case. Since the owners are all minority shareholders, they do not exercise actual control over the firms they own.11_{Therefore, it is necessary to prove that the anti-competitive}

incentive of the owners was passed on to the firms owned. The way in which this incentive is passed on could of course be collusive. However, this is not what economic literature focuses on, especially because there is no clear evidence of such collusion. This is not to say that collusion is impossible, there are many meetings between institutional investors and companies that are happening behind closed doors that could be used for anti-competitive agreements.12 Instead, three things are said to be the reason why the firms act in a manner that benefits the common owner. The first reason is less about the things that common owners do but more about what they fail to do. In fact, several studies allege that institutional investors simply fail to encourage competition.13 The idea is that firms only compete if they have an incentive to do so,

8_{José Azar, Martin C. Schmalz, and Isabel Tecu, ‘Anti-Competitive Effect of Common Ownership’ (2018), 73(4)} Journal of Finance, page 7

9_{Steven C. Salop and Daniel P. O’Brien ‘Competitive Effects of Partial Ownership: Financial Interest and} Corporate Control’ (2000) 67 Antitrust L.J., page 573

10_{OECD Competition Committee, ‘Common Ownership and Competition: Facts, Misconceptions, and What to} Do About It - Note by Martin C. Schmalz’ (DAF/COMP/WD(2017)93, 2017), page 5

11_{For the purpose of this thesis, it is assumed that common owners don’t have direct control over the companies} they own. This mirrors the actual situation and is also the situation that is assumed in economic literature. It will therefore not be considered what the situation would be if a common owner actually exercised control over the firms it owns.

(9)

8

and that common owners, through omission, fail to give these firms an incentive.14 The second reason alleges that common owners not only fail to promote competition, but that they purposely overpower activist investors who are driving for more competition.15 Since every common owner has an interest in less competition, institutional investors can, despite their individual small shareholdings, jointly block decisions of activist investors, even if these might have bigger shareholdings.16 This is especially so because many shareholders do not actually go to shareholder meetings, meaning that the total number of votes needed for a blocking minority is lower than one might expect.17 The last reason is that, rather than being linked to firm performance, executive compensation is linked to industry performance or not linked to performance at all.18 Institutional investors support this type of compensation, even though it is known to be inefficient.19 Already in 1982, Holmström’s Nobel prizewinning work has shown that it would be more efficient to base compensation of executives on the performance of the firm relative to other firms in the industry, instead of basing it on industry performance generally.20_{There is in fact statistical evidence that in cases of common ownership, executive}

compensation is less linked to performance of the firm.21_{These three reasons show that, even}

without actual collusion, there are ways in which common owners could influence conduct of the companies.

There is not only theoretical, but also empirical evidence of anti-competitive conduct. Two economic studies conducted in the banking and the airline sector have found that there is a link between common ownership and higher prices for consumers. Concerning airlines, José Azar et al. found that airline tickets are 3-7% more expensive than without common ownership.22 The same holds true for banking products, which are more expensive than they would be without common ownership.23 In addition, taking the example of the corporate governance of

14_{OECD Competition Committee, ‘Common Ownership and Competition: Facts, Misconceptions, and What to} Do About It - Note by Martin C. Schmalz’ (DAF/COMP/WD(2017)93, 2017), page 5

16_{Monopolkommission ‘Wettbewerb 2016’ 21}st_{Report of the German monopoly commission in accordance with} § 44(1) GWB, pages 228-230

17_{Ibid, page 230}

18_{José Azar, Martin C. Schmalz, and Isabel Tecu, ‘Anti-Competitive Effect of Common Ownership’ (2018), 73(4)} Journal of Finance

19_Ibid.

20_{Holmström, ‘Moral Hazard in Teams’ (1982), 13(2) Bell Journal of Economics, pages 324-40}

21_{OECD Competition Committee, ‘Antitrust and Institutional Investor Involvement in Corporate Governance -} Note by Edward B. Rock and Daniel L. Rubinfeld’ (DAF/COMP/WD(2017)94, 2017)

23_{José Azar, Sahil Raina and Martin Schmalz ‘Ultimate Ownership and Bank Competition’ (2016), available at} SSRN: https://ssrn.com/abstract=2710252

(10)

9

the chemical firm DuPont, the top shareholders of the company are also the top shareholders of the main competitor, Monsanto.24 A proxy fight erupted between an activist shareholder, the Trian Fund, and DuPont’s management, with Trian alleging that DuPont did not attempt to gain market share and generally did not compete aggressively enough with Monsanto. These allegations were strongly rejected by the common owners.25 While this is not proof that common ownership lowers competitive incentives, it is a concrete example of a situation in which common ownership and lessened competition coexist.

The evidence for anti-competitive effects has been disputed however, both theoretically and empirically. One of the most common arguments against anti-competitive effects is that institutional investors have very diverse portfolios that often change over time.26 Indeed, it is said that it would be very difficult for a company to behave in a manner that would please all its minority shareholders.27_{A certain course action that would benefit one shareholder, i.e.}

BlackRock, would not necessarily be beneficial to another shareholder, i.e. Vanguard, as the portfolios of both firms are different. However, this criticism is based on the premise that one institutional investor owns a different set of firms in a certain industry than another investor. This is not the case however. Both BlackRock and Vanguard are in fact amongst the top five shareholders in 70% of publicly traded US firms and in Europe, institutional investors also own considerable amounts of equity.28 Therefore, investor portfolios do not actually vary that much. In fact, institutional investors own every firm in a given industry and not only a selection that differs from investor to investor. Another argument that is often raised is that common owners own not only horizontally related companies but also vertically related companies, both upstream and downstream. The anti-competitive conduct on one level of the market would have harmful effects on the upstream or downstream market, this being a disadvantage for the common owner who has invested in companies active on those markets. In other words, the advantages that anti-competitive conduct of firms would have for the common owner would be neutralised by the harmful effects on suppliers or customers of these firms, which the common owner owns as well.29 This argument has also been rebutted however. Since common owners

24_{Einer Elhauge ‘Thesis - Horizontal Shareholding’ (2016) 129 Harv. L. Rev, page 1270} 25_{Ibid. page 1271}

26_{OECD Competition Committee, ‘Antitrust and Institutional Investor Involvement in Corporate Governance -} Note by Edward B. Rock and Daniel L. Rubinfeld’ (DAF/COMP/WD(2017)94, 2017), page 4

27_{OECD Competition Committee, The Competitive Effects of Common Ownership: Ten Points on the Current} State of Play - Note by Daniel P. O’Brien (DAF/COMP/WD(2017)97, 2017), page 4

28_{Eric Gonnard, Eun Jung Kim and Isabelle Ynesta ‘Recent trends in Institutional Investors statistics’ (2008)} OECD Financial Market Trends

(11)

10

own most of the suppliers and customers of the firms in question, it is not out of the question that there is anti-competitive conduct also on those levels of the market.30 Another theoretical argument raised is that in general, management compensation at least partly depends on firm performance as it is paid in equity (for example in stock options).31 However, this argument disregards that stock prices of a company rise not only when the company itself is doing well but also when the industry in general is doing well. Anti-competitive conduct and the higher overall industry profits therefore also benefit executives partially remunerated in equity. The only way to ensure the right incentives through compensation would be to link executive compensation to actual firm performance, which is not common practice however.32 Other than these main arguments against anti-competitive conduct, it has also been held that the alleged anti-competitive conduct of firms benefits not only one common owner but also the other common owners with whom this investor competes, this being an undesirable situation for him.33_{This is also not entirely correct. In fact, common owners have an interest in the success}

of their competitors. This is because there are cross-shareholdings between institutional investors and because they need one another as they are providing each other with the liquidity required for their operations.34 Therefore, conduct benefitting competitors is not always undesirable for institutional investors. Furthermore, it has even been explored whether common ownership could create efficiencies, for example by mitigating friction between competitors, reducing information asymmetry and making possible the exploring of new business opportunities.35 The literature on this topic is not yet developed however and will not be considered further in depth here, even if it is an interesting field for further research.

From an empirical point of view, the evidence of anti-competitive effects has also been contested. There are two main criticisms. First, while there is evidence of correlation between higher prices and common ownership, it has not been proven that there is also causality.36 Both

30_{Einer Elhauge ‘The Growing Problem of Horizontal Shareholding’ (2017) Antitrust Chronicle, Vol. 3,} Competition Policy International, page 12

32_{Miguel Antón, Florian Ederer, Mireia Giné, Martin Schmalz ‘Common Ownership, Competition and Top} Management Incentives’ ECGI Finance Working Paper N. 511/2017, page 35

34_{Monopolkommission ‘Wettbewerb 2016’ 21st Report of the German monopoly commission in accordance with} §44(1) GWB, pages 228 and 230

35_{Jie He and Jiekun Huang ‘Product Market Competition in a World of Cross-Ownership: Evidence from} Institutional Blockholdings’ (2016) Review of Financial Studies, page 2. Available at SSRN:

https://ssrn.com/abstract=2380426

(12)

11

the airline and the banking sector, in which correlation has been found, are concentrated sectors with few competitors. Since there is endogeneity of concentration in prices, the fact that the prices are higher in the banking and the airline sectors is not necessarily due to common ownership but could also be due to higher concentration in these sectors.37 The assumed anti-competitive conduct could in that case be an example of tacit collusion.38 Second, another empirical criticism is that the evidence is limited and marginal. In fact, the findings are limited to only two sectors in the United States, and it is not proven whether the findings apply to other sectors or regions, although it has been alleged that this is the case.39 In fact, the same systemic issue, that institutional investors own most of publicly traded stocks, exists in other regions too (in Germany for example, institutional investors own 60% of publicly traded stock).40 Next to that, the effect is marginal. The most quoted effect is the 3-7% price increase in airline ticket prices. This already small amount has furthermore been challenged, with other studies, notably by the US Federal Reserve Board, reporting much lower effects.41

It follows that the effects of common ownership are not clear yet. Theoretically, the anti-competitive effects are discussed extensively; there is developed and convincing evidence that common owners have the possibility and incentive to adopt anti-competitive conduct. The arguments trying to prove the opposite can and have been rebutted. However, empirically, there is still a lack of decisive evidence. The main shortcomings have been described above and they cast a shadow over the theoretical evidence. This thesis acknowledges the need for further research into the actual empirical effects of common ownership. Nonetheless, the convincing theoretical evidence, as well as the, albeit limited, empirical proof, justify a discussion of potential approaches to regulation. Such a discussion is interesting, be it just to see the potential options and to see whether any regulation would be grossly disproportionate. This is why, after having considered existing options in part two, the third part will consider the different possibilities to regulate common ownership.

37_{OECD Competition Committee, ‘Antitrust and Institutional Investor Involvement in Corporate Governance -} Note by Edward B. Rock and Daniel L. Rubinfeld’ (DAF/COMP/WD(2017)94, 2017), page 5

38_{OECD Competition Committee, ‘Antitrust and Institutional Investor Involvement in Corporate Governance -} Note by Edward B. Rock and Daniel L. Rubinfeld’ (DAF/COMP/WD(2017)94, 2017), page 5-6

40_{Monopolkommission ‘Wettbewerb 2016’ 21st Report of the German monopoly commission in accordance with} § 44(1) GWB, page 223

41_{Gramlich, Jacob and Serafin Grundl ‘Testing for Competitive Effects of Common Ownership’ (2017) Finance} and Economics Discussion Series 2017-029, Washington: Board of Governors of the Federal Reserve System, page 12

(13)

12

IV. The role for competition law

Before considering whether or how competition law regulates common ownerships, it should at least briefly be considered whether competition law should be the instrument regulating common ownership in the first place. In fact, it would be possible to address the theoretical causes of common ownership through corporate law. For example, compensation plans for executives could be imposed by law to make sure that they are adequately linked to firm performance. Other options would be to limit voting rights for asset managers or imposing limits on the levels of shares of a company that can be held by asset managers. However; there are several practical shortcomings to the use of corporate law instruments. First, not all causes can be addressed through corporate law. For example, it is difficult to force investors to promote competition. There would also be obvious interpretation issues, such as deciding which conduct exactly promotes competition. Second, it would be very intrusive to force a certain conduct on investors or companies and enforcing these rules would be burdensome. Therefore, this thesis takes the view that traditional corporate law instruments are not suited to address the causes of common ownership in a proportionate manner.

Part 2: The current regulation of common ownership

I. Minority shareholding and common ownership

The first part of this thesis concerned the theory of harm behind common ownership. While it was admitted that there is still uncertainty as to economic evidence of harmful effects, the conclusion was still that it would be at least of academic interest to consider possibilities to regulate common ownership. For this reason, this part will now consider whether current instruments available under EU competition law are sufficient to counter anti-competitive effects.

Before the different instruments are addressed, the distinction between minority shareholding and common ownership must be addressed again. As previously said, the distinction between both concepts is important but viewing them in complete isolation is not desirable. In fact, while the theory of harm is different, the underlying transaction is similar as it concerns the acquisition of a minority shareholding. It is true that there is a significant difference with regard to which actor acquires this shareholding. However, from a transactional perspective, both situations concern the acquisition of a minority shareholding. For this reason,

(14)

13

this thesis also considers the way that competition law applies to minority shareholding, as it would apply in the same manner to the transaction underlying common ownership. In addition, the assumption is that the theory of harm is as described above. Therefore, it is not assumed that there is any actual collusion in form of common owners prescribing certain behaviour to the companies they own.

II. The EU Merger Regulation (EUMR)

1. Current Situation

To fall under the Regulation 139/2004 (‘the Merger Regulation’), there must be a concentration. A concentration is deemed present in case there is a change of control over an undertaking.42

Not every acquisition of shareholding will confer control to the shareholder, so in order to fall under the Merger Regulation there must be an acquisition of de facto or de jure control.43_In

case there are special rights granted to a certain shareholder, granting him control over the undertaking, this would be an example of de jure control.44 The amount of shareholding does not matter; around 20% can be enough.45 On the other hand, in case there are no special rights, but the situation is such that one company has effective control, there can be de facto control. The existence of such control depends on the constellation of other shareholders, for example how fragmented they are and whether they have structural links with each other. De facto control can arise as of very low levels, the Commission expressed concerns in respect of shareholdings as low as 15%.46

Control, either de facto or de jure, can arise either on a sole basis or a joint basis.47 Joint control is present if two or more companies must reach an agreement concerning decisions of the controlled company.48 For joint control to arise, each of the controlling companies must generally have veto rights over important decisions.49 It is possible in exceptional cases that even without veto rights, commonality of interests is enough for joint control.50 It is important

42_{Article 3(2) EUMR}

43_{Commission Consolidated Jurisdictional Notice under Council Regulation (EC) No 139/2004 on the control of} concentrations between undertakings (2008/C 95/01) (‘Commission Jurisdictional Notice’), para 55

44_{Consolidated Jurisdictional Notice, para 57}

45_{See for example Case IV/M.258 of 25.09.1992 CCIE/GTE, where CCIE only owned 19% of shares but had} special rights.

46_{Case IV/M.967 of 22.09.1997 KLM/Air UK, the shareholding in this case was together with other financial} arrangements.

47_{Consolidated Jurisdictional Notice para 63} 48_Ibid.

49_{Ibid. Para 65} 50_{Ibid. Para 75}

(15)

14

to note that the Commission explicitly state that financial interests of investors do not qualify for joint control, because coalitions of minority shareholders often change.51

2. Shortcomings and Issues

The main issue with the EUMR is that there must be a change of control in an undertaking. In case of an acquisition of a minority shareholding by a common owner, there is, at least under the theory of harm employed, no acquisition of sole control, neither on a de facto nor on a de

jure basis. Even the lowest levels of shareholding the Commission admitted for sole control are

still almost three times as high as the shareholdings that common owners generally have (around 5%).52 This is next to the fact that there are no indications that common owners enjoy any special rights, possibly granting them de jure control. It follows that sole control is certainly not present in situations of common ownership. This is different with joint control, since most companies in the affected sectors are owned by a few large investors, a case could be made for joint control. However, as stated above, the Consolidated Jurisdictional Notice explicitly excludes the possibility that financial interests could result in a commonality of interests that is strong enough to amount to joint control. Another point is that most of the investments of the common owners are passive investments. This means that the common owners will not take an activist stand and it is unlikely that they would actively cooperate to take control of a company. Their business model, at least in theory, is not to control the company’s management but simply to hold the shares on behalf of their customers. For this reason, it is unlikely that common owners would have a common interest in a company, meaning that joint control is unlikely to be the case.

Generally, when minority shareholdings are evaluated under the EUMR, this usually happens in the context of a Merger filing in a separate, unrelated transaction.53 In the clearance process of that transaction, the Commission will look at the minority shareholdings that the parties to the transaction have in the relevant markets.54 This could lead to the Commission

51_{Ibid. Para 79-80}

52_{Monopolkommission ‘Wettbewerb 2016’ 21st Report of the German monopoly commission in accordance with} §44(1) GWB, page 227

53_{Tommy Staahl Gabrielsen, Erling Hjelmeng and Lars Sørgard, ‘Rethinking Minority Share Ownership and} Interlocking Directorships: The Scope for Competition Law Intervention’ (2011) 6 European Law Review, page 845

54_{Ioannis Platis ‘Competition Law Implications of Minority Shareholdings: The E.U. and U.S. Perspectives’} (2013) Hellenic Review of European Law (HREL), page 11

(16)

15

asking for the sale of these minority shareholdings as a commitment to clear the merger.55 However, it is important to note that the Commission could not order the divestment of an already existing minority shareholding in the framework of a merger filing, unless it is considered part of the transaction at hand. In other words, where a transaction is considered under the EUMR and the acquirer already has minority shareholdings, the acquirer cannot be ordered to divest these shareholdings if the merger is prohibited.56

For these reasons, the EUMR is not a suitable instrument to regulate common ownership. The actual transaction whereby the minority shareholding is acquired does not fall in the scope of the Regulation because there is no change of control. In addition, evaluation of this shareholding in the framework of another merger case is dependent on the existence of a notifiable merger of the investor in a related market. Moreover, even in that case, divestiture of the minority shareholding could most likely not be ordered.

III. Article 101 TFEU

In order to be caught by Article 101 of the Treaty on the Functioning of the European Union, it is first necessary that there is an agreement. In the case of common ownership, the theory of harm does not go from the premise that there is actual collusion between the common owner and the companies in which it holds minority stakes. Therefore, there is no traditional agreement, even oral. In this case, the only act of the common owner is the acquisition of a minority shareholding. This does not preclude the existence of an agreement however. In its

Philip Morris judgment, the Court of Justice held that the acquisition of a minority stake

amounted to an agreement, but in this case, the acquisition was executed in the context of a share transfer agreement.57

In case an agreement is present, this agreement also has to have as its effect or object the restriction of competition. In its Philip Morris judgment, the Court clearly stated that the

56_{Tommy Staahl Gabrielsen, Erling Hjelmeng and Lars Sørgard, ‘Rethinking Minority Share Ownership and} Interlocking Directorships: The Scope for Competition Law Intervention’ (2011) 6 European Law Review, 847. See also T-411/07 Aer Lingus v European Commission (2010) ECR II-03691

57_{Joined Cases 142 and 156/84 British-American Tobacco e.a. v European Commission (1987) ECR-04487} (‘Philip Morris case’), para 31

(17)

16

acquisition of a minority shareholding could never in itself be restrictive of competition.58 The Court therefore seems to exclude that there is a restriction by object, meaning that the concrete impact of the transaction must be analysed.59 The Court continues and outlines four situations in which minority shareholdings can have anti-competitive effects:60

- If the shareholding leads to de facto or de jure control.

- The agreement leaves open the possibility of reinforcing its position at a later stage. - The agreement creates structures that could be used for commercial cooperation. - The shareholding requires firms to take into consideration each other’s interests when

determining their commercial policy.

Every situation, and particularly the two last criteria, providing for anti-competitive effects in case of undue influence, have to be read in the light of the market structure however.61_In

oligopolies characterised by little price competition and high barriers to entry, it is very likely that minority shareholdings are used to take over a competitor.62_{However, it is more difficult}

to prove cooperation, as the market structure has a natural tendency to lead to parallel pricing.63

The first issue is the requirement of an agreement. As stated above, situations involving minority shareholdings have been considered agreements previously, but these situations concerned acquisitions by share acquisition agreement. In case of common ownership, there are generally no share transfer agreements but simple stock market transactions. This is because the common owner buys and sells shares of a company depending on the amount of money transferred in or out of the ETFs it offers. Thus, shareholdings are not acquired as a block through an agreement but in smaller batches through stock market transactions. Situations where shares are purchased without share transfer agreements are much less likely to be considered agreements, since the counterparty to the transaction is usually not even known.64 However, an alternative would be to consider the articles of association of the company an

58_{Ibid., para 37}

59_{See for example Case IV/M.856 BT/MCI, para 44} 60_{Philip Morris case, paras 38-39 and 47-48} 61_{Philip Morris case, para 40}

62_{Philipp Morris case, para 44}

(18)

17

agreement, in fact a shareholder’s control over another company depends on these articles of association, it acts as an agreement between the company and the shareholder as to the rights that the shareholder has.65 Certain national company laws even expressly state that articles of association are in fact agreements between shareholders.66 However, while this interpretation might be logically sound, it has not been confirmed by the Commission or the Court of Justice. To the opposite, the Commission has even expressed doubts that articles of association could amount to agreements, without elaborating on it too much however.67

Even in case an agreement could be established, this agreement would still need to have anti-competitive effects. The Court in Philipp Morris held that anti-competitive effects could be assumed in case the acquisition of the minority shareholding confers de facto or de jure control, the share acquisition agreement gives the possibility to increase the shareholding at a later stage or the shareholding grants undue influence over the target. In case of common ownership, it has been previously established that the shareholding does not grant de facto or

de jure control.68_{Since no share acquisition agreements are used by common owners, they}

cannot contain a provision providing for acquisition of control at a later stage. Therefore, it must be seen whether common ownership creates structures that enable cooperation or requires firms to take each other’s interest into account. It is as such possible to imagine that common ownership facilitates information exchange and could possibly create a so-called ‘hub-and-spoke’ cartel. In such a scenario, the institutional investor would function as a proxy through which the companies would exchange commercially sensitive information. However, this is not what is alleged by the theory of harm behind common ownership. The most relevant question is therefore whether the common ownership requires parties to take into consideration each other’s interests. The Court in Philip Morris explicitly considers whether passive shareholdings could have such an effect.69 This shows that the Court accepts the possibility that passive minority shareholding grants influence on a competitor’s conduct.70 Whether undue influence is present must be assessed on a case-by-case basis, taking into account size of shareholding,

65_{Alec J. Burnside ‘Minority Shareholdings: An overview of EU and national case law’ (2013) e-competitions N.} 56676, page 2-3

66_{Ibid page 3. See also section 31(1) Irish Companies Act 2014 and section 33(1) UK Companies Act 2006} 67_{European Commission White Paper Towards more effective EU merger control, COM(2014) 449 final, para 40} 68_{See Part 2, II.2 above}

70_{OECD Competition Committee, ‘Tackling Horizontal Shareholding: An Update and Extension to the Sherman} Act and EU Competition Law - Note by Einer Elhauge’ (DAF/COMP/WD(2017)95, 2017), page 20

(19)

18

special rights of the shareholder, overall market structure, etc.71 Accordingly, shareholdings of common owners could be anti-competitive, if companies take into account the interests of the common owner as a consequence. This takes us to the heart of the debate about anti-competitive effects of common ownership. As explained in Part 1 of this thesis, it has been alleged that common owners can influence companies, notably through compensation plans. However, this has not been proven conclusively yet and it is unclear whether the Commission would accept the economic evidence.

For these reasons, Article 101 TFEU is not a reliable instrument to regulate common ownership. First, it is unclear whether acquisitions of stocks on the stock market could be considered agreements, even if arguments for this view can be made. Second, even if an agreement could be found, anti-competitive effects would have to be proven. This would be possible in case common ownership forces companies to consider interests of common owners. This is problematic because economic theory has not yet been able to prove beyond reasonable doubt that common ownership really leads companies to act in the interest of common owners.

IV. Article 102 TFEU

The final part of the current legislative framework is the prohibition of abuses of dominance, laid down in Article 102 TFEU. Acquisitions of minority shareholdings can be caught by this prohibition, under the condition that one of the two companies involved in the transaction has a dominant position.72 Two situations are conceivable. First, the acquired company could have a dominant position. In that case the shareholding must give the acquirer effective control of the company for Article 102 TFEU issues to arise.73 The second possibility is that the acquiring company is dominant. The Court dealt with that possibility in the Gillette case.74 In that case, the Court changed its approach and held that in case the acquiring company is dominant, there is no need for the shareholding to result in ‘effective control’. In such cases, ‘some influence’ is enough for Article 102 TFEU being applicable, as this already results in the structure of the

71_{Tommy Staahl Gabrielsen, Erling Hjelmeng and Lars Sørgard, ‘Rethinking Minority Share Ownership and} Interlocking Directorships: The Scope for Competition Law Intervention’ (2011) 6 European Law Review page 851

73_{Philip Morris, para 65}

(20)

19

market being influenced, thereby hindering the maintenance of the competition still present in the market.75 The Court therefore seems to lower the boundaries of application of Article 102 TFEU in case the acquiring company is dominant.

There is one more hypothetical situation. This concerns the situation in which none of the companies is initially dominant but the share acquisition results in collective dominance. The topic of collective dominance is disputed in EU competition law due to it lying on the borderline between Articles 101 TFEU and 102 TFEU. The Court of Justice has defined collective dominance in the Compagnie Maritime Belge case, requiring that two independent economic entities had to present themselves together or act together as a collective entity.76 The assessment whether a collective entity exists has to be made from the customer’s point of view77

and economic links between the undertakings have to exist.78 Such links must not necessarily be in the form of an agreement, they must simply flow from economic assessment of the structure of the market.79_{In case there are structural links leading to collective dominance, the}

creation of additional links, such as minority shareholdings or common ownership, could be seen as an illegal strengthening of that dominant position.80_{Thus, strengthening of collective}

dominance through acquisitions of additional minority shareholdings could be a violation of Article 102 TFEU.

Concerning Article 102 TFEU, the biggest obstacle to an application to common ownership is the requirement of a dominant position. As explained above, one of the parties involved in a transaction has to be dominant in order for acquisitions of minority shareholdings, and thus for common ownership, to fall under the prohibition. In case of common ownership, this is problematic. While the common owners in question (BlackRock, Vanguard and State Street) are the biggest asset managers, the market is very fragmented, even if one only considers the total number of institutional investors engaging in asset management, i.e. the practice of managing investments for the account or the interest of others.81 If one were to consider

75_{Gillette case, paras 23-24}

76_{Joined cases C-395/96 P and C-396/96 P Compagnie Maritime Belge (2010) ECR I-01365 case para 36} 77_{Ibid para 39}

78_{Ibid para 41} 79_{Ibid para 45}

81_{Monopolkommission ‘Wettbewerb 2016’ 21st Report of the German monopoly commission in accordance with} § 44(1) GWB, page 225

(21)

20

institutions engaged in proprietary trading, i.e. managing investments for the institution’s own account and in its own interests, as being active on the same market as asset managers, the market shares of these three common owners would further decrease. Therefore, it cannot be said that institutional investors are dominant. It follows that the rule stemming from the Gillette case, concerning situations in which the acquirer is dominant, cannot be used. Turning to the second possibility, that the acquired company is dominant, this depends very much on the industry. Since common ownership is most problematic in concentrated markets, it is possible that one of the undertakings could be found dominant. However, this is not enough, as the Court in the Philip Morris case requires that effective control must be acquired over the dominant undertaking. In case of common ownership, this is not the case, meaning that this condition would not be fulfilled.

Therefore, the only way in which Article 102 TFEU could apply to common owners is through collective dominance. In this case, the idea would be that two competing companies, that are already collectively dominant, strengthen the economic links between them due to the phenomenon of common ownership.82_{This requires two things: pre-existing collective}

dominance and a strengthening of economic links. Concerning pre-existing collective dominance, it is unlikely that this is the case, at least in the sectors in which common ownership is alleged, namely banking and airlines. For one, it does not seem as if customers consider independent airlines or banks one collective entity. However, an argument could still be made that, for airlines, airline alliances could create this perception. For banks, the perception of a clubby and interlinked industry could also be argued to exist, especially since the financial crisis. Concerning a strengthening of economic links, once again, it must be said that it has not yet been conclusively proven that common ownership creates links between companies.

For these reasons, it is unlikely that Article 102 TFEU would apply to common ownership. First, neither the acquired company, nor the acquiring company, have dominant positions in their relevant markets. Second, collective dominance is also not present because pre-existing dominance is difficult to prove and even if it could be established, it would still be necessary to prove that common ownership actually creates or reinforces economic links between competing companies.

(22)

21

Part 3: Possible Alternatives

In part 2, the current legal framework was analysed and its applicability to common ownership was tested. The conclusion was that none of the currently available tools in EU Competition law, namely the EU Merger Regulation, Article 101 TFEU and Article 102 TFEU, are applicable to the issue of common ownership. For this reason, this part will now consider how the system could be changed in order to be applicable to common ownership. First, this thesis will consider policy proposals that have been made and that could be used to regulate common ownership. Second, it will be discussed how a regulatory approach, designed to be applicable to common ownership, could look like. Inspiration will be drawn from the policy proposals that have been made. Three possible approaches to the regulation of common ownership will be introduced: macro regulation, micro regulation and mixed regulation.

I. Policy proposals

1. The Commission White Paper

In 2014, the European Commission made a proposal to close a perceived “enforcement gap” in EU Merger Control.83 The proposal, called the “White Paper - Towards more effective EU merger control” (hereinafter ‘the White Paper’)84_{aims at bringing acquisitions of minority}

shareholdings in the scope of EU merger control. In fact, under the current system of EU merger control, acquisitions of minority shareholdings, insofar as they do not result in a change of control, cannot be reviewed by the European Commission.85

The assumption of the Commission is that minority shareholdings can raise concerns if they are horizontal because they give a financial interest in a competitor’s profits, thereby increasing incentives to increase prices.86 In addition, minority shareholdings could give the acquirer influence over the target,87 or facilitate collusion between the competitors.88 It is further said that minority shareholdings cannot be addressed through current competition law instruments.89

83_{Ulrich von Koppenfels, ‘A Fresh Look at the EU Merger Regulation? The European Commission’s White Paper} “Towards More Effective EU Merger control” (2015) 36 Liverpool Law Review, page 12

84_{European Commission White Paper Towards more effective EU merger control, COM(2014) 449 final} 85_{See above, Part 2, section II.1}

86_{European Commission White Paper Towards more effective EU merger control, COM(2014) 449 final, para 29} 87_{Ibid, para 32}

88_{Ibid, para 35} 89_{Ibid, para 39-41}

(23)

22

The proposal of the Commission revolves around a “targeted transparency system”.90

Under this proposal, only transactions creating a competitively significant link should fall under the scope of review of the Commission.91 A competitively significant link is deemed present in case: 92

- The shareholding is acquired in a competitor or a vertically related company.

- The acquired shareholding is ‘significant’, meaning that it exceeds 20% or is between 5-20% and accompanied by special rights (veto powers, seat on the board, access to sensitive information…).

The idea of the White Paper is that parties self-assess whether a transaction creates a competitively significant link. If they believe that this is the case, they should file a short information notice.93_{The information notice should contain basic information such as turnover,}

a description of the transaction, levels of shareholding before and after the transaction, any special rights attached to the shareholding, and finally some market share information.94

Finally, a waiting period of 15 days is proposed, starting with the filing of the notice.95 Within that time, the Commission could consider whether it requires parties to submit a full notification through a Form CO.96 This system is supposed to fit within the existing system of division of competences, the allocation of competence between Commission and national competent authorities (NCAs) would be based on the existing turnover thresholds. The referral system in the EUMR would also be applicable.97

The system proposed by the Commission could in theory also be used for the control of common ownership. It catches the acquisition of minority shareholdings, meaning that common ownership would be scrutinized the moment it is acquired. The obvious shortcoming would be the low level of shareholding in common ownership. Since common owners rarely have shareholdings exceeding 5%, and no special rights are attached to that shareholding, it is unlikely that the White Paper would be able to regulate common ownership. However, the basic

90_{Ibid, para 45}

91_{Ulrich von Koppenfels, ‘A Fresh Look at the EU Merger Regulation? The European Commission’s White Paper} “Towards More Effective EU Merger control” (2015) 36 Liverpool Law Review, page 20

92_{European Commission White Paper Towards more effective EU merger control, COM(2014) 449 final, para 47} 93_{Ibid, para 48}

94_{Ibid, para 49} 95_{Ibid para 50} 96_{Ibid para 49}

97_{Commission Staff Working Document Towards more effective EU merger control, SWD(2013) 239 final, page} 9

(24)

23

architecture of the system could be the basis for a regulatory approach, specifically designed for common ownership, which will be discussed below.

2. Other proposals

There have also been economic and legal papers that have proposed alternative systems of regulation, targeted specifically at common ownership. Two examples will be discussed below.

The first proposal, elaborated by Chicago Law School professor Eric A. Posner, is of an economic nature. It aims at reducing the MHHI98 in concentrated industries. The proposal states that:

“No institutional investor or individual holding shares of more than a single effective firm in an oligopoly may ultimately own more than 1% of the market share unless the entity holding shares is a free-standing index fund that commits to being purely passive. “99

This policy can be best described as the ‘one firm per shareholder’ rule. The idea is that institutional investors should invest only in one firm in industries that are deemed oligopolies. There would be a safe harbour of 1%, below this threshold, shareholdings are allowed even if another firm in the same industry is already owned. Finally, there is an exception for purely passive index funds. For that purpose, purely passive means that no communication to top management is allowed and that votes are allocated in proportion to existing votes (no matter how low the turnout) so that effectively no influence is exercised.100 This policy would indeed solve the issue of common ownership, as only one company per industry could be owned. However, there is an entire range of issues that this policy would entail, two of which are even identified by the authors themselves. First, in case an institutional investor wants to switch his investment from one firm to another, he would have to sell off all its stock in the company 1 and only then invest in company 2. This would exert immense pressure on the stock price and liquidity of company 1 and the price of the competitor’s stock would rise. This would make such a switch of investment extremely costly and lengthy for the company and its investors, without even taking into consideration possible consequences for the industry affected.101 This

98_{MHHI (‘modified HHI‘) is an adapted version of the HHI that takes into account market concentration due to} common ownership.

99_{Eric A. Posner, Morton Scott, Fiona M. and Glen E. Weyl, ‘A Proposal to Limit the Anti-Competitive Power of} Institutional Investors’ (2017) Antitrust Law Journal, Forthcoming. Available at SSRN:

https://ssrn.com/abstract=2872754, page 33 100_{Ibid. Page 34}

(25)

24

issue is not remedied by the author’s solution to offer a grace period for investors wanting to switch investments, during which an investor would be allowed to hold stocks of two companies.102 Such a grace period does not change the need to sell off all stock, meaning that, unless a very long period would be granted, it would at most stretch out the effects for the first company. A related concern that the authors identify is that such a policy would need a complete change of current market structures. The ETF market would have to be reformed almost in its entirety, meaning that there would be a massive sale of stocks as ETFs are closed. Market turmoil would be the consequence.103 Once again, the authors propose a delay between announcement and actual implementation,104 but considering the amount of stock held by institutional investors, effects would be catastrophic nonetheless. Other issues that the authors don’t consider are the difficulty to define industries or assess whether they are characterised by oligopoly, as well as monitor changes in the market that affect these concepts. In addition, it would raise issues in case two companies merge or create joint ventures, meaning that a company could unexpectedly enter or exit a market. Finally, there are systemic issues, such a lack of possibilities to diversify or problems arising in case every investor chooses to invest in the same company.105 It follows that the ‘one firm per shareholder’ rule could maybe solve the issue of common ownership, but at a cost that would be way beyond the possible benefits.

The second proposal is of a legal nature and has been elaborated by Harvard Law School professor Einer Elhauge. The core of the proposal is that competition authorities should investigate stock acquisitions that create a ΔMHHI of over 200 in a market with an MHHI over 2500.106 In case these thresholds are fulfilled, the authority should determine whether the acquisition would be likely to increase prices. The advantage of this approach would be that the effects analysis that is required would mean that there would be no blanket prohibition creating market turmoil, as would be the case in the ‘One firm per shareholder’ rule. This proposal will be discussed more in depth below.

102_Ibid.

103_{Eric A. Posner, Morton Scott, Fiona M. and Glen E. Weyl, ‘A Proposal to Limit the Anti-Competitive Power} of Institutional Investors’ (2017) Antitrust Law Journal, Forthcoming. Available at SSRN:

https://ssrn.com/abstract=2872754 , page 39 104_Ibid.

105_Ibid.

(26)

25

3. Approaches in other jurisdictions

Another interesting point is the question how other jurisdictions approach the issue of minority shareholdings or common ownership. The approach in Germany and in the United Kingdom will be considered in depth.

In the United Kingdom, minority shareholdings are addressed through merger control. In fact, merger control in the UK employs a wide concept of control, going beyond what would typically be considered de jure or de facto control.107 Instead, the UK Enterprise Act employs the concept of material influence over the acquired business; in case material influence is acquired, the transaction is subject to scrutiny.108 The analysis whether a minority shareholding confers material influence rests on two pillars, namely voting rights and board representation. However, other factors will be taken into account, and the final decision will be taken based on all relevant facts of the case.109_{While the UK system readily applies to the acquisition of}

minority shareholdings, it is doubtful whether it would also apply to common ownership. In most cases, common owners do not reach the ‘material influence’ standard, at least if the criteria of voting rights and board representation are considered. Therefore, the UK system would not be a possible alternative to control common ownership.

In Germany, a slightly different approach is taken. Under German law, a transaction is subject to notification “if the shares, either separately or in combination with other shares

already held by the undertaking, reach (…) 25% of the capital or the voting rights of the other undertaking”.110 A notification is also required in case the acquisition gives the holder the possibility to “directly or indirectly exercise a competitively significant influence”.111_{There are}

therefore two jurisdictional thresholds. The first one is unambiguous, but also not relevant to the situation of common owners, who never reach a 25% shareholding. The second threshold is more ambiguous. It is important to note that significant influence must be established under corporate law, economic dependency is not sufficient.112 This means that there must be certain

de jure or de facto circumstances that provide the minority shareholder with a blocking minority

on a lasting basis.113 It follows that such an approach is similar to the one of the United

108_{Alec J. Burnside ‘Minority Shareholdings: An overview of EU and national case law’ (2013) e-competitions} N. 56676, page 5

109_{Ibid. See also Competition Commission Report, Ryanair Holdings plc/Aer Lingus Group plc, 28 August 2013,} para. 4.9.

110_{Section 37(1) No.3(b) GWB} 111_{Section 37(1) No.4 GWB}

112_{Bundesgerichtshof, II ZR 171/83, BGHZ 90, 381, BuM/WestLB, 26.3.1984.}

113_{Jens Peter Schmidt, ‘Germany: Merger control analysis of minority shareholdings - A model for the EU?’} (2013) Concurrences N. 2-2013 page 208-209