9.1 Introduction

In this article, prepared for a special JCLE symposium, I revisit my initial findings regarding the prevalence of ‘horizontal directors’ in the United States.Footnote ¹ ‘Horizontal directors’ serve on the board of multiple companies operating within the same industry. I have previously spotlighted the prevalence of horizontal directors in the US, despite a prima facie prohibition on director service among competitors.

Despite that spotlight and increasing attention to common ownership and to director interlocks, this Article explores six additional data years to further demonstrate the prevalence of horizontal directors as recently as the end of 2019.

In fact, in this project, the original dataset has been enhanced with bookend data from 2007 to 2009 and 2017 to 2019 for the director-level analysis. These expanded data confirm the rise of horizontal directors previously discussed and highlight their continued prevalence even against a backdrop of increased attention to the effects of common ownership and directors’ interlocks.

Horizontal directors are significant because they stand at a unique intersection of antitrust law and corporate governance. They offer many legitimate governance and operational benefits to companies and shareholders but at the same time pose significant concerns both to the governance of the corporation and to antitrust policy. Despite that significance, horizontal directors have yet to receive the proper attention from regulators, stock exchanges, and investors.

This lack of attention to the horizontal aspect of director interlocks is particularly surprising for two reasons. First, existing US antitrust regulation specifically prohibits directors from serving on the boards of two competitors,Footnote ² so sharing a director across companies in the same industry may violate these laws. Second, antitrust law has re-emerged in recent yearsFootnote ³ commanding increased attention to market concentration and consumer welfare (specifically in merger decisionsFootnote ⁴ and ‘horizontal shareholding’Footnote ⁵ by institutional investors). This emerging literature has sparked a vivid academic and public debate regarding the effects of shareholder concentration on antitrust policy.Footnote ⁶ Specifically, scholars have raised concerns regarding the incentives of companies to compete where major institutional shareholders hold large equity positions in all competitors. While market concentration by large investors has been widely acknowledged, a vivid debate has ensued on whether such concentration materialises in ways that promote anticompetitive behaviour.Footnote ⁷ As a by-product of this debate, recent focus has been directed to the question of which channels common owners use to effect anticompetitive behaviour.

For instance, one channel prominently discussed in recent years is executive compensation. Some suggest that common owners may either actively discourage performance-sensitive compensation or not actively push for a particular compensation plan and instead promote the status quo which lets executives ‘get away with high performance-insensitive pay’.Footnote ⁸ While less plausible, another channel is through a targeted strategy of specific actions that is communicated to management that promotes portfolio value even at the expense of firm value.Footnote ⁹ In this channel, a common owner obtains leverage over management with its voting power and its ability to sell shares and decrease the market price of firm stock in order to promote its strategy.Footnote ¹⁰ Interestingly, and surprisingly, the role of directors as potential conduits of common ownership has been left underexplored. Specifically, horizontal directors might create this exact channel for common owners’ influence, therefore facilitating anticompetitive effects on the market.Footnote ¹¹ Alternately, horizontal directors may alone serve as the channel through which anticompetitive practices are achieved, without the need to pin such results on common owners.Footnote ¹²

Equally important, horizontal directors are not a rarity: in fact, as shown below, empirical data reveal hundreds of directors concurrently serving on boards of companies operating in the same or similar industries. More so, the prevalence of horizontal directors is on the rise. Yet, despite the rise in horizontal directorships over time, proactive corporate disclosure of horizontal directors remains sparse.Footnote ¹³ Notably, the presence of horizontal directors across industry lines is not equal. Some industries are more prone to having horizontal directors than others. For example, while the construction industry has on average 10.6% industry-level horizontal directors serving on two or more companies’ boards, the manufacturing industry has 33.7%.Footnote ¹⁴ This variance across industries invites for further research trying to connect industry levels of anticompetitiveness and horizontal directorships.

This Article proceeds as follows. Part II starts by providing an overview of the unique case of horizontal directors. Part III then provides a refreshed empirical analysis of the S&P 1500 director dataset and also revisits the company-level analysis of disclosure practices. Part IV then provides an overview of the current US legal framework governing and regulating horizontal directors. Part V discusses and analyses the implications of those results in more detail in light of the potential benefits and concerns horizontal directors may bring.

9.2 From Interlocks to Horizontalness

Directors’ service on multiple boards has drawn both investor and academic attention,Footnote ¹⁵ mostly focusing on one of two areas: (1) the number of board seats a director holds and whether directors who hold several board positions have an impact on company performance or other governance metricsFootnote ¹⁶ or (2) the ‘interlocks’, or the connections and bridges, created between two (or more) companies by having a director that serves on both (or multiple) boards.Footnote ¹⁷

Busy directors, and the interlocks they create, are a natural and inevitable by-product of a corporate culture that taps directors to serve on multiple boards at once.Footnote ¹⁸ The benefits of serving on multiple boards are tangible. Busy directors have more experience, provide more connections and develop more industry expertise at a rate that exponentially increases with the number of boards they serve.Footnote ¹⁹ Conversely, less-busy directors provide fewer tangential benefits derived from busyness to firms.Footnote ²⁰

While there is no shortage in attention to busy directors,Footnote ²¹ missing from current discourse is a more nuanced account of the boards on which busy directors serve. While by definition directors serving on more than one board fall into the definition of a ‘busy director’, some of these busy directors also serve on more than one board in the same industry – these are what I termed as horizontal directors.Footnote ²²

Horizontal directors are particularly important because their prevalence, as discussed below, raises both governance and antitrust concerns. Horizontal directors raise antitrust concerns since their concurrent board seats provide a channel between companies in the same industry. These channels can lead to either collaboration and/or collusion.Footnote ²³ Additionally, horizontal directors also raise corporate governance concerns, such as reduced independence and increased conformity in governance practices that could lead to systemic governance risk.Footnote ²⁴ Moreover, the rise in horizontal directors comes against a backdrop of heightened concentration in the US markets. More than 75% of US industries experienced a rise in concentration levelsFootnote ²⁵ in recent years.Footnote ²⁶ As the distribution of a given market among participating companies becomes less spread out, the anticompetitive effects of collusion and price-fixing intensify. Therefore, the potential impact of horizontal directors is also amplified.

A final factor contributing to the profound effect horizontal directors can have in the corporate landscape is a product of the fact that demarcation lines among industries are becoming increasingly more difficult to ascertain.Footnote ²⁷ Because horizontal directors are identified based on industry, the murkier industry lines become, the harder it will be to identify and monitor horizontal directors for anticompetitive behaviour.

9.3 Revisiting Horizontal Directors: Empirical Findings

This Part provides augmented data on the prevalence of horizontal directors in the US as well as information disclosed to investors on horizontal directors. The data presented herein extend prior analysis,Footnote ²⁸ expanding the analysis to include data from 2007 to 2009 and 2017 to 2019, therefore providing an even more broad view of the rise of director horizontalness and the persistence of it even as recently as January 2020.

9.3.1 Horizontal Directors in S&P 1500 Companies

9.3.1.1 Methodology

This Part examines the prevalence of horizontal directors on boards in the same industry for companies within the S&P 1500 from 2007 and through 2019. The data for this sample were originally compiled from Equilar’s BoardEdge dataset.Footnote ²⁹ Both director-level data and company-level data were obtained for each company within the S&P 1500 for each year previously mentioned. These separate datasets were merged to create one panel dataset at the director-company-year level where each individual case describes a director’s service on a specific S&P 1500 board as well as any additional boards that specific director served on and that were outside of the S&P 1500. For the purposes of analysis all directors were included, whether they were designated as independent or not. This consolidated dataset was subsequently supplemented with company-level data from FactSet and the North American Industry Classification System (NAICS) Association to add NAICS codesFootnote ³⁰ and industry classifications for groups of SIC and NAICS codes.

Directors were coded as ‘horizontal’ using four classifications: whether a director served on more than one board in the same (1) SIC code, (2) SIC industry, (3) NAICS code, or (4) NAICS industry. Because an ‘industry’ contains multiple SIC codes or NAICS codes, the industry-horizontal classifications are a broader measure than the classifications based on specific SIC or NAICS codes. However, using both SIC and NAIC classifications allows for a more robust analysis. Directors were given a binary variable (0 or 1) for each classification indicating their horizontal status. These variables were used to calculate the number of horizontal boards on which the director served in a given year. Directors were also coded as ‘busy’ – serving on more than one board at a time – regardless of whether those boards are horizontal. Each busy director was individually coded with an indicator variable as well as a count of the number of boards on which they served each year.

9.3.1.2 Director-Level Analysis

As depicted in Table 9.1, the expanded data show the number and percentage of directors in the S&P 1500 who served on more than one board. From 2007 to 2019, more than 30% of all directors sat on more than one board, and a substantial number of directors (around 12%) held three or four board positions. The data covering the 2010–2016 years (‘The Original Data’) pointed out that the percent of busy directors did not vary more than 2% from 2010 to 2016. However, the supplemented data from 2007 to 2009 and 2017 to 2019 show a more prominent incline in busy directors over time. For example, the number of directors that served on two boards increased by 17% from 2007 to 2019, and the number of directors that served on three boards increased by 39% for this same time period.

Table 9.1Number of boards a director sits on

	1	2	3	4	5	6	7+	N
2019	61.93%	24.99%	9.21%	3.11%	0.63%	0.09%	0.04%	100.00%
	6,159	2,485	916	309	63	9	4	9,945
2018	60.81%	24.91%	10.14%	3.22%	0.75%	0.12%	0.05%	100.00%
	6,175	2,530	1,030	327	76	12	5	10.155
2017	60.42%	25.08%	10.43%	3.16%	0.78%	0.09%	0.03%	100.00%
	5,943	2,467	1,026	311	77	9	3	9,836
2016	65.14%	22.44%	9.08%	2.66%	0.52%	0.10%	0.04%	100.00%
	8,191	2,822	1,142	335	66	13	5	12,574
2015	63.75%	23.49%	9.05%	2.92%	0.59%	0.13%	0.07%	100.00%
	7,937	2,925	1,127	363	74	16	9	12,451
2014	63.55%	23.37%	9.22%	2.83%	0.79%	0.18%	0.07%	100.00%
	7,960	2,927	1,155	354	99	22	9	12,526
2013	63.31%	23.28%	9.78%	2.70%	0.70%	0.17%	0.07%	100.00%
	7,822	2,877	1,208	333	86	21	9	12,356
2012	63.92%	23.24%	9.40%	2.44%	0.78%	0.14%	0.07%	100.00%
	7,755	2,820	1,141	296	95	17	8	12,132
2011	64.23%	23.25%	9.11%	2.44%	0.75%	0.19%	0.03%	100.00%
	7,650	2,769	1,085	291	89	23	3	11,910
2010	63.63%	23.41%	9.20%	2.68%	0.74%	0.30%	0.03%	100.00%
	7,564	27,83	1,094	318	88	36	4	11,887
2009	72.88%	18.76%	6.03%	1.71%	0.50%	0.12%	0.01%	100.00%
	10,090	2,597	835	237	69	16	1	13,845
2008	71.60%	19.88%	6.35%	1.56%	0.49%	0.11%	0.02%	100.00%
	10,164	2,822	902	221	69	15	3	14,196
2007	69.60%	21.35%	6.64%	1.80%	0.47%	0.11%	0.02%	100.00%
	9,675	2,968	923	250	66	15	3	13,899

Additionally, the percent of directors that serve on two or more boards has increased despite the fact that the total number of directors in the sample has decreased by 3% on average each year (total decline of 28% from 2007 to 2019).

As I have previously underscored, horizontal directors are not outliers among directors of public companies – the opposite is true. Table 9.2 shows that the trends previously highlighted with respect to the Original Data are amplified by the additional years included. Although the data show that the percent of busy directors that share an industry within their boards served does not vary more than 8% in this 13-year period, a closer look at the data shows more nuanced patterns. For example, the number of directors that served on boards of at least two companies within the same industry slowly increased from 2007 to 2009. However, from 2009 to 2010 there was a 20% decrease in the number of directors that served on two or more boards within the same industry, showing some year-to-year volatility in the appointment of horizontal directors.

Table 9.2Number and percentage of busy directors sharing an industry within boards served

# of boards	2007	2008	2009	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
2	895 (34%)	994 (35%)	1473 (46%)	952 (34%)	1002 (36%)	1056 (37%)	1077 (37%)	1094 (37%)	1075 (37%)	1051 (37%)	936 (38%)	960 (38%)	954 (38%)
3	519 (62%)	571 (63%)	618 (65%)	676 (62%)	687 (63%)	724 (63%)	800 (66%)	769 (67%)	751 (67%)	767 (67%)	675 (65%)	674 (65%)	596 (65%)
4	197 (79%)	184 (78%)	266 (83%)	258 (81%)	237 (81%)	237 (81%)	267 (80%)	289 (82%)	300 (83%)	282 (84%)	262 (84%)	281 (85%)	269 (86%)
5	63 (90%)	62 (90%)	97 (92%)	80 (91%)	81 (91%)	89 (94%)	79 (92%)	93 (94%)	67 (91%)	62 (94%)	75 (95%)	73 (95%)	56 (90%)
6	16 (100%)	15 (100%)	25 (93%)	33 (92%)	22 (96%)	16 (94%)	21 (100%)	22 (100%)	16 (100%)	13 (100%)	9 (100%)	12 (100%)	10 (100%)
7+	1 (100%)	4 (100%)	10 (100%)	4 (100%)	3 (100%)	8 (100%)	9 (100%)	9 (100%)	9 (100%)	5 (100%)	2 (67%)	4 (80%)	3 (75%)

Although industry classification is broader than a single SIC or NAICS classification, combining multiple codes, the number of horizontal directors is substantial under both metrics. There were 1,888 directors that served on the board of more than one company within the same industry in 2019. On a more granular level, there were 412 directors (10.8% of directors serving on more than one board) who served on at least two companies in the same industry per four-digit SIC code. Similarly, there were 250 directors (9.2% of the directors serving on more than one board) that served on at least two companies’ boards within the same NAICS code.

The number of directors that serve on more than one company with the same NAICS or SIC code has decreased by only 3% from 2016 to 2019 (Table 9.3). Therefore, there has not been a significant change in the striking presence of horizontal directors since the Original Data.

Table 9.3Number and percentage of busy directors sharing SIC/NAICS within boards served

# of boards	SIC/NAICS	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
2	SIC	140 (5%)	149 (5%)	180 (6%)	183 (6%)	195 (7%)	208 (8%)	209 (8%)	170 (7%)	169 (7%)	184 (7%)
	NAICS	104 (4%)	120 (4%)	144 (5%)	155 (5%)	165 (6%)	159 (5%)	147 (5%)	143 (6%)	151 (6%)	156 (6%)
3	SIC	101 (9%)	103 (9%)	123 (11%)	139 (12%)	129 (11%)	148 (13%)	150 (13%)	146 (14%)	154 (15%)	129 (14%)
	NAICS	80 (7%)	81 (7%)	99 (9%)	118 (10%)	115 (10%)	129 (11%)	132 (12%)	121 (12%)	123 (12%)	107 (12%)
4	SIC	45 (14%)	43 (15%)	53 (18%)	70 (21%)	68 (19%)	85 (25%)	102 (29%)	73 (23%)	77 (23%)	67 (21%)
	NAICS	36 (11%)	40 (14%)	41 (14%)	57 (17%)	65 (18%)	62 (17%)	56 (17%)	65 (21%)	65 (20%)	63 (20%)
5	SIC	21 (24%)	19 (21%)	28 (29%)	28 (33%)	31 (31%)	24 (32%)	26 (39%)	38 (48%)	29 (38%)	22 (35%)
	NAICS	14 (16%)	14 (16%)	28 (29%)	23 (27%)	23 (23%)	19 (26%)	24 (36%)	35 (44%)	23 (30%)	14 (23%)
6	SIC	13 (36%)	8 (35%)	9 (53%)	8 (38%)	13 (59%)	9 (56%)	10 (77%)	7 (78%)	7 (58%)	9 (90%)
	NAICS	13 (36%)	10 (43%)	7 (41%)	5 (24%)	11 (50%)	7 (44%)	8 (62%)	5 (56%)	6 (50%)	9 (90%)
7+	SIC	1 (25%)	0 (0%)	5 (62%)	7 (78%)	6 (67%)	4 (44%)	2 (40%)	2 67%)	3 (60%)	1 (25%)
	NAICS	1 (25%)	0 (0%)	5 (62%)	6 (67%)	5 (56%)	3 (33%)	2 (40%)	2 (67%)	3 (60%)	1 (25%)
Total	SIC	321 (7.4%)	322 (7.5%)	398 (9.2%)	435 (9.5%)	442 (9.7%)	478 (10.5%)	499 (11.3%)	436 (11.1%)	449 (11.2%)	412 (10.8%)
	NAICS	248 (5.7%)	265 (6.2%)	324 (7.5%)	364 (8%)	384 (8.4%)	379 (8.3%)	369 (8.4%)	371 (9.5%)	371 (9.2%)	350 (9.2%)

Table 9.4 further shows that the percentage of horizontal directors as a percent of busy and total directors has been trending upwards over time with a merely a slight decline seen in the last two years. For example, in 2010, 7.4% of all horizontal directors sat on at least two boards of within the same SIC classification. This number increased by 7% on average each year from 2010 to 2016 but decreased by 2% in both 2018 and 2019. A similar trend was observed under the NAICS classification with an average increase in that metric of 8% each year with a decline of 3% in the last two years.

Table 9.4Time trend of horizontal directors

Year	2010	2011	2012	2013	2014	2015	2016	2017	2018	2019
% of Industry-Horizontal Directors out of All Directors	16.9	17.1	17.6	18.2	18.2	17.8	17.3	19.7	19.5	18.8
% of Industry-Horizontal Directors out of Busy Directors	46.3	47.7	48.7	49.7	49.8	49.1	49.7	50.2	50.2	49.6
% of SIC-Horizontal Directors out of All Directors	2.7	2.7	3.3	3.5	3.5	3.8	3.9	4.4	4.3	4.2
% of SIC-Horizontal Directors out of Busy Directors	7.4	7.6	9.1	9.6	9.7	10.6	10.9	11.2	11	10.8
% of NAICS-Horizontal Directors out of All Directors	2.1	2.2	2.7	2.9	3.1	3	2.9	3.7	3.6	3.6
% of NAICS-Horizontal Directors out of Busy Directors	5.7	6.2	7.4	8	8.4	8.4	8.4	9.5	9.3	9.2

The pattern of steady growth followed by a slight decline is magnified in Figure 9.1. As depicted below, there has been a significant amount of growth from 2007 to 2016, particularly among directors serving on three or more boards. However, the expanded data show that the number of busy directors who sit on boards for companies within the same SIC code has slightly decreased in the last couple of years especially for directors serving on four or more boards, though it is still at significantly higher levels compared to 2010.

Figure 9.1

Percentage of busy directors sitting on at least two boards with the same SIC code

9.4 The Peculiar Presence of Horizontal Directors

9.5 Horizontal Directors: Zero-Sum Proposition?

Some level of collaboration between companies within the same industry can be beneficial to consumers; therefore, antitrust laws only target efforts that lead to anticompetitive outcomes or collusion.Footnote ⁶⁹ Horizontal directors may provide value to the company and investors, such as contributing to the diffusion of beneficial corporate governance practices, networking, and expertise. By sitting on boards of multiple companies in the same industry, horizontal directors gain intimate knowledge can be a valuable asset to a director’s ability to advise and monitor the management team.Footnote ⁷⁰ However, the presence of horizontal directors also presents concerns, such as an increased risk of antitrust collaboration, an increased risk of systemic governance risk,Footnote ⁷¹ and decreased director independence.Footnote ⁷² Furthermore, horizontal directors may facilitate anticompetitive practices that could further insulate management from market pressures which may lead to a loss of shareholder value in the long term.

As previously explored, this Article re-emphasises the need to shine a spotlight on horizontal directors and to address the accompanying concerns. Even though there has been a slight decline in the number of directors serving on companies within the same industry in the last few years, the overall prevalence of horizontal directors remains a concern.

Yet, horizontal directors are not necessarily a zero-sum proposition. Companies could still tap the valuable aspects of horizontal directors while at the same time minimising the concerns that they may present.

First, legislation that targets higher risk companies will be more effective at mitigating antitrust risks and will be easier to enforce uniformly, which will mitigate some of the current Section 8 underenforcement concerns. As I previously discussed,Footnote ⁷³ some industries and SIC codes are more likely to have horizontal directors, and some of these horizontal-director-saturated industries also exhibit strong levels of industry concentration, perhaps making them a key starting point for evaluation.

Focusing the prohibition on concentrated industries might strike a desired balance. The right balance would allow companies to enjoy the benefits these directors provide while prohibiting their presence in cases where the costs to competition are more likely to outweigh these benefits. For instance, Section 8 could be revised to exempt from the prohibition industries with an HHI that is below a certain threshold. Aggressively enforcing Section 8 for that subset of public companies would reduce significant antitrust risk. Italy took a similar approach, in focusing on the banking sector. The main provision dealing with interlocking directors prohibits members of boards and any top manager of a company operating in the banking, insurance, and financial sectors from holding one of those offices in a competing company.Footnote ⁷⁴ However, the Italian legislation was criticised for its inflexibility and lack of clarity.Footnote ⁷⁵ For example, the Act did not include a de minimis exemption for very small firms that did not prompt the same concerns as larger companies.Footnote ⁷⁶ If policymakers choose to amend the Clayton Act, they should take care to incorporate flexibility to strike the desired balance.

Second, regulators could consider an ex-ante design to Section 8 of the Clayton Act that would allow directors to apply for a waiver before accepting a horizontal directorship. By obtaining an ex-ante ‘no action’ waiver,Footnote ⁷⁷ companies would be more certain about nominating potential directors. Furthermore, companies would be able to justify the nomination of directors that would technically violate the Act. Giving the FTC a veto right ex-ante would also reduce the need for costly ex-post enforcement and may lead to more consistent enforcement.

In fact, a similar arrangement is already employed in the context of interlocking bank directorships. The Federal Reserve’s (the Fed) Regulation L is similar to the Clayton Act in that it prohibits an officer or director of a bank from serving as an officer or director for more than one of any bank’s holding companies with over $10 billion in assets.Footnote ⁷⁸ However, Regulation L allows the Fed to grant waivers when it determines that an interlock would not substantially lessen competition.Footnote ⁷⁹ While the banking industry is more regulated than other industries, one can easily find ways to implement this rule in a cost effective way across other industries. For instance, the FTC may require a public notice to be filed, with a presumption of approval unless otherwise indicated within 20 days. The notice in turn will also allow shareholders, stock exchanges, and proxy advisors to apply private ordering restrictions if they so desire.

Third, and as previously mentioned, horizontal directors toe the line between antitrust and corporate governance, and a comprehensive reform should highlight the benefits of these directors as well as address the corporate governance risks. As Part II highlighted, many companies currently keep disclosures to the bare minimum required.Footnote ⁸⁰ Thus, it is often difficult to even identify the industry of the other boards on which horizontal directors serve. Regardless of whether shareholders see horizontal directors as positively or negatively impacting the company, improved transparency via more comprehensive disclosures would enable shareholders to more effectively participate in corporate governance by making more informed director nominations and board recommendations.

An alternative approach to consider for reform would be updating stock exchanges’ self-regulation to better incorporate horizontal directorship concerns. One concern of horizontal directorships is the ability of directors to serve as independent directors when they have extensive experience in one industry. From a shareholder perspective, including directors with deep industry experience may add significant value to the company. Accordingly, stock exchanges may consider revising their independence definitions to exclude directors from being considered independent if they serve on the boards of two companies in the same SIC code, whether or not they are considered competitors. This restriction could strike a better balance between enabling companies from benefitting from horizontal directorship and preventing boards and directors from becoming too dependent on their specific industry connection.

Finally, state law and fiduciary law can also evolve to increase restrictions to mitigate the concerns that arise from the prominence of horizontal directors. Additionally, tightening judicial review of non-compete agreements, corporate opportunity waivers, and board fiduciary duties may position common law jurisprudence to more effectively address potential governance issues that stem from the presence of horizontal directors. For example, courts may examine corporate opportunity waivers more skeptically where a horizontal director is involved and where the opportunity is given to a horizontal company. The common law route can provide the flexibility and adaptability that regulatory intervention often lacks; however, it will depend on litigants voicing concerns.

9.6 Conclusion

In many ways, horizontal directors epitomise the push and pull of our corporate governance system. Directors are expected to monitor management, to provide expertise and networking, and to make the corporation’s most important decisions.Footnote ⁸¹ Yet, we lean on outsiders to serve as directors, and we allow, and even encourage, their service on other boards, potentially undermining their ability to appropriately serve their director role. Indeed, many directors serve on more than one board when given the opportunity. Director is a desired position due to the relatively limited time commitment to each board, significant salary and perks, and limited exposure to legal risk.Footnote ⁸² Additionally, serving on several boards across industries, within the same industry, and even within the same SIC code can benefit not only the director but also the companies she serves – at least under certain conditions.Footnote ⁸³

Yet, there is an open question as to how horizontal directors should fit within our current antitrust regulatory framework and corporate governance regime. Moreover, how to appropriately balance the competing interests remains unresolved. Similarly, it remains unclear how we should view horizontal directorships given increased industry concentration and vivid discourse regarding horizontal mergers and horizontal shareholdings.

This Article demonstrates that horizontal directors remain a prevalent feature (and bug) of the US corporate landscape. Future research into horizontal directorships is still needed, given the increased reliance on the board as an institution,Footnote ⁸⁴ and the emergence of contemporary antitrust discourse regarding horizontal ties between companies through common shareholders. Understanding that not all companies are created equal, investors may be better situated than regulators to account for the rise in horizontal directorships and offer market-based solutions to the inherent tension that these directors present.

10.1 Introduction

Interlocking directorates refer to situations in which one or more companies have one or more members of their boards in common. In the US, under Section 8 of the Clayton Act, competing companies are prohibited from having common board members.Footnote ¹ In application of this prohibition, Eric Schmidt, CEO of Google, stepped down from the board of Apple in 2009. In the EU, however, such links between companies’ boards are not uncommon. In the EU, as well as in the different Member States,Footnote ² there is no such express prohibition of interlocking directorates between competitors. Some economies have even been characterised by very dense networks of companies owing to multiple links among their boardsFootnote ³.

This chapter highlights the potential anticompetitive risks raised by interlocking directorates between competitors. The anticompetitive effects stem both from the increased ability to collude enabled by interlocks, as well as the reduced incentive to compete fiercely on markets characterised with numerous social and corporate links. In addition, this chapter touches upon the questions of conflict of interest and problems of directors’ independence that are inherent when a board member sits on the boards of two competing companies.

The main claim of this chapter is that there may be an enforcement gap around anticompetitive effects of interlocking directorates in Europe. Although Article 101 TFEU and EU Merger Control regulation theoretically apply to the coordinated and unilateral effects of interlocks, these provisions are of very limited use in practice. Company law in some Member States, such as France, limits the number of board appointments a director may hold, but such solutions are specific to certain types of companies and are largely insufficient to address the anticompetitive effects of interlocking directorates. Principles of corporate governance, such as fiduciary duties, are not binding and seem inappropriate to prevent anticompetitive effects and issues of conflict of interests.

Issues raised by interlocking directorates do not attract the attention they deserve and are notably absent from discussions on possible issues raised by financial links at the EU level.Footnote ⁴ In the US, the discussion about anticompetitive effects of common ownership should also grant more attention to interlocking directorates, particularly in the light of recent findings on the prevalence of interlocking directorates in the US.Footnote ⁵ This is because financial ownership links and interlocking directorates raise similar concerns, critically at the edge of competition law and corporate governance.Footnote ⁶ Therefore, this chapter draws attention on the need to tackle potentially significant issues that are currently largely undebated in Europe.

This chapter demonstrates that the anticompetitive effects of interlocking directorates (Section 10.1) may fall short of EU competition law (Section 10.2). Section 10.3 explains how interlocking directorates may be regulated in other jurisdictions, including in the US, and discusses whether tools of corporate governance may remedy the identified anticompetitive concerns (Section 10.4).

10.2 Anticompetitive Effects of Interlocking Directorates

Various studies highlighted that corporate networks across industries, based on interlocking directorates, have been particularly dense in continental Europe, although networks have tended to become less dense and more diffuse over the past decades.Footnote ⁷ These studies also demonstrate that corporate networks based on interlocks have been comparatively less dense in the US and in the UK.Footnote ⁸ Germany has long been characterised by dense networks of companies where banks play a central role, leading to the qualification of ‘cooperative capitalism’ as a feature of the German economy.Footnote ⁹ In France, large companies were typically connected through their boards with a high number of CEOs sitting as independent board members of competitors.Footnote ¹⁰ A recent study analysed the structure and evolution of corporate networks of the top 100 French, British, and German companies over a 14-year period (2006–2019). It found that although networks are composed of weaker links (individuals hold less appointments on average), they are wider in scope (more companies are now part of the networks).Footnote ¹¹ While it does not specifically provide intra-sectoral information, this study demonstrates the existence of traditionally dense corporate networks in Europe comprising companies from the same industry. As explored by various corporate governance scholars, the existence of interlocking directorates may enhance the firm performance owing to the cognitive input of an experienced board member, and resource potential the link may create.Footnote ¹²

When held among competing companies, interlocking directorates may give rise to unilateral and coordinated effects.Footnote ¹³ The first impact on competition stems from the information and communication flows facilitated by interlocking directorates. Board members have access to strategic, accounting, and commercial information as well as information regarding the appointment and compensation of executives.Footnote ¹⁴ Information and communication between competitors have been shown to facilitate collusion, even when not specifically related to prices and quantities. Information flows may help in reaching a collusive agreement and also provide monitoring tools for competitors to prevent deviation from the collusive agreement.Footnote ¹⁵ As an example, a network of interlocking directorates has helped stabilise a number of cartels, including the international uranium and diamond cartels.Footnote ¹⁶ Accordingly, the purpose of the US prohibition of interlocking directorates is expressly to ‘avoid the opportunity for the coordination of business decisions by competitors and to prevent the exchange of commercially sensitive information by competitors’.Footnote ¹⁷ Anticompetitive agreements can also be facilitated by indirect interlocks where competitors sit on the board of a third party. Information exchanges can be more discrete with indirect rather than direct interlocks.Footnote ¹⁸

Interlocks may also affect unilateral incentives to compete. Social ties created by the attendance of common board meetings may discourage aggressive commercial strategies towards rivals. If interlocks are widespread within industries, this may reduce the overall intensity of competition.Footnote ¹⁹ When attached to financial interests, interlocking directorates may provide the ability to influence a competitor’s conduct. The remuneration schemes in place may also affect the incentive to compete, especially if closely tied to the firm’s performance.Footnote ²⁰

Nevertheless, economic efficiencies are more likely to exist in the area of interlocking directorates than in the situation of minority shareholdings (e.g. where used to align incentives in joint ventures).Footnote ²¹ Information exchange, enabled by such links, may reduce strategic uncertainty which may under certain circumstances be pro-competitive if it improves business decision-making. The presence of the board member of a competitor offers the benefit of his expertise and experience which may improve decision-making. Moreover, the exchange of information can create synergies in the control and management of companies facing similar technical and economic issues. A business can also benefit from the reputation of an independent board member and use it in situations where the asymmetry of information may be an obstacle in negotiations to obtain financing from banks or investors. Similarly, the expertise and reputation of the board member of a competitor can facilitate contractual negotiations with suppliers and customers – especially in small businesses.Footnote ²²

The anticompetitive effects of interlocking directorates are exacerbated if the corporate governance of the competing companies is weak. Directors sitting on several boards may influence the decision process in one company, as a way of favouring another company of which they are a board member. Directors may also be tempted to disclose confidential information of a company at another company’s board meeting. These issues may be mitigated by the quality of the fiduciary duty. A strong fiduciary duty, which indicates good corporate governance, may prevent the director from engaging in these types of practices. A director’s fiduciary duty to one company, however, may naturally conflict with their fiduciary duty in another company.Footnote ²³ Overall, bad quality of corporate governance is more likely to induce directors with shared directorship to compete less aggressively.Footnote ²⁴

10.3 The Reach of EU Competition Law Over Interlocking Directorates

In the EU, a structural link is scrutinised under the Merger Regulation if it is part of an acquisition that confers a ‘lasting change in the control of the undertaking’.Footnote ³⁷ Interlocking directorates which are not part of an acquisition conferring control can be captured by Article 101TFEU only to the extent there is an agreement or concerted practice between undertakings, or by Article 102 TFEU if there is dominance.

10.4 Interlocking Directorates in Other Jurisdictions

10.5 Principles of Corporate Governance

Structural links are at the heart of corporate governance systems. This section discusses whether principles of corporate governance can set constraints over the anticompetitive effects of interlocking directorates. Competition law may adjust its boundaries to address common issues that corporate laws and corporate governance fail to address. This overall shows that the discussion at the EU level require a multi-disciplinary approach to the issue of interlocking directorates.

While protection of minority shareholders and the freedom to appoint board members are essential to corporate governance, these corporate arrangements can also hinder rivalry between companies. In addition, policies regarding corporate governance encourage an active role by institutional investors in the corporate governance, which seems to conflict with the competitive concerns raised by common ownership.Footnote ⁸¹

Yet, corporate governance and competition law seem to converge on other issues. A core principle of corporate governance is the fiduciary duty of management to shareholders.Footnote ⁸² Fiduciary duty may mitigate the anticompetitive effects of structural links. In the context of interlocking directorates, a strong fiduciary duty may prevent a common board member from disclosing information from one company to the other. However, a director’s fiduciary duty to one company may naturally conflict with their fiduciary duty to another company.Footnote ⁸³

Independence of decision-making is another important principle of corporate governance.Footnote ⁸⁴ Accordingly, decisions should be made in the company’s best interest, without consideration of other companies.Footnote ⁸⁵ The French Asset Management Association warns against the risk of interlocking directorates as undermining transparency and independence of decision-making, unless associated with a strategic economic alliance.Footnote ⁸⁶ In practice, however, an increase in price taken in the interest of a competitor may be difficult to identify. Collecting evidence and taking action, such as voting to remove a director in breach of a fiduciary obligation, could be difficult and risky for the shareholders. Further exploration of corporate governance mechanisms is therefore critical to understanding the practical ability of a board to raise prices unfavourably for the company, for the financial benefits of a competitor.Footnote ⁸⁷

As an example, the French Court of Cassation reaffirms the legal requirement of fiduciary duty for top executives, which then applies to those sitting on the board of a competing company. In addition, the court clarified that this duty forbids the chief executive from commercial negotiation in his capacity as manager of another company within the same industry.Footnote ⁸⁸ However, such requirement, rather limited to apprehend the whole spectrum of anticompetitive effects, only applies to executives (and not to all directors) of French companies. In addition, the code of corporate governance recommends that as an ethical rule, a board member should be bound to report to the board any actual or potential conflict of interest, and refrain from voting on the related resolution.Footnote ⁸⁹ Although no express mention is there made of conflicts of interests arising from individuals sitting within multiple board meetings, generic rules on conflicts of interest are likely to encompass such instances.

Interlocking directorates may pose additional problems both for corporate governance and competition law, if top managers favour the selection (or exclusion) of board members based on how passive (or active), they are on other boards, in an effort to retain control over the board.Footnote ⁹⁰ In addition, mutual interlocks can reflect and contribute to CEO entrenchment, resulting in higher compensation and lower turnover.Footnote ⁹¹

11.1 Introduction

In this chapter, we provide an overview of the Italian legislation on interlocking directorates and its enforcement in the last decade. Italy has introduced an anti-interlocking provision to promote competition in the banking, insurance, and financial sectors. Even if it is not easy to make comparisons with other EU Member States, many studies claimed that the number and relative dimension of the Italian financial companies linked by interlocking directorates were greater than in other Member States.Footnote ¹ This is why, in 2011, in the aftermath of a very harsh financial crisis, Italy enacted a statutory provision forbidding the simultaneous appointment of the same person to the board of directors (or to other corporate bodies)Footnote ² of two or more competing financial companies.

After explaining why, without regulation, these personal ties may facilitate or reinforce the achievement of a collusive or quiet life equilibrium among competitors, we provide a brief description of the main features and scope of the 2011 Italian interlocking ban. We then attempt to evaluate its effectiveness and limits. Using the banking sector as a case study, we gathered data on the number of interlocking directorates that persisted among the 25 largest banks and banking groups in Italy at the end of 2018. The result of our study is that interlocking directorates among major Italian banks seem to have disappeared. This is at odds with the prevailing empirical literature which has claimed that interlocking directorates are still a widespread reality of Italian capitalism, with possible persisting anticompetitive effects in many markets. To counter this claim, we also considered some empirical studies showing that, in the period following the entry into force of the interlocking ban, bank lending rates fell, which suggests more vigorous competition.

We conclude the chapter by questioning whether the 2011 interlocking ban has had any effect on the ownership structure of the relevant market players – for instance, contributing to the disposal of minority and cross-shareholdings held by companies operating in the sectors concerned – as well as on the composition of their governing bodies.

11.2 Interlocking Directorates, Ownership Ties, and Their Effect on Competition

While there are different justifications to establish interlocking directorates,Footnote ³ these personal links may produce anticompetitive effects when the linked firms operate on the same markets. The conclusion is straightforward when the personal tie is one of the elements of a wider collusive scheme in which a director sitting on the board of two horizontal competitors helps make sure that each firm sticks to the terms of an agreed-upon behaviour.

However, even when there is no anticompetitive agreement in place, interlocking directorates can still limit competition by facilitating the establishment of a quiet life equilibrium.Footnote ⁴ This often happens when interlocking directorships are coupled with ownership ties. To be sure, not all ownership ties are relevant in an antitrust perspective. For instance, common majority shareholdings are usually not relevant, since companies belonging to the same group generally constitute a single economic entity and are thus not considered independent competitors.Footnote ⁵ Other ownership ties may, however, pose antitrust concerns. This is the case for minority shareholdings (e.g., firm A holds a minority stake in competitor firm B),Footnote ⁶ cross-shareholdings (e.g., competitors A and B hold shares in one another), and horizontal shareholdings (e.g., horizontal competitors A and B have, in all or in part, the same minority shareholders).Footnote ⁷ In all these cases, the horizontal competitors might have an incentive to forego or limit competition, as the higher profits that each firm would earn from competition could be, in all or in part, offset or even outweighed by a loss in the value of the stake held in the competitor (directly by the company or by its common minority shareholders).

The incentives are mostly the same when the ownership ties between the competing firms are only indirect and include companies operating in other markets or sectors, as it would happen if the cross-shareholding involved one or more interposed companies. This is probably why, in the Italian experience, interlocking directorates and ownership ties have often gone hand in hand. It is also why, considering that Bank of Italy regulations restrict ownership ties between financial firms,Footnote ⁸ they have typically involved not only these companies but listed companies more generally.Footnote ⁹ Empirical studies based on data collected before the 2011 statutory ban on interlocking directorates showed, for instance, that the ownership patterns of large Italian listed companies, including major financial firms, closely resembled the interlocking directorates among them.Footnote ¹⁰ At least until 2011, many listed companies that shared one or more directors also directly or indirectly shared some stockholders and/or held shares in one another. Interlocking directorates have hence likely enhanced or strengthened the anticompetitive pressure created by the ownership links of Italian capitalism.

However, interlocking directorates may have anticompetitive effects also in other circumstancesFootnote ¹¹ and even if the competing firms are not linked by ownership ties. For instance, before and during board meetings, strategic information is shared among the directors and officers of the firm. The interlocked directors receive this information together with the other members of the board and may use it to the benefit of the competing firm in which they also serve.

Even when the interlocked directors do not actively use the information to the advantage of the competitor, the information may influence their opinions and votes at board meetings.Footnote ¹² This is because a director sitting on the boards of two competing companies owes fiduciary duties to both of them and must try to maximise the profits of both. That may per se lead to a quiet life equilibrium, since the director will vote for the solution that maximises the joint profits of the two companies or that increases the profit of one without damaging the other.Footnote ¹³ Indeed, if ‘X is a director of both Corporation A and Corporation B’, ‘X could hardly vote for a policy by A that would injure B without violating his duty of loyalty to B; at the same time he could hardly abstain from voting without depriving A of his best judgment’.Footnote ¹⁴ If the director is an executive officer in at least one of the interlocked companies, the impact on the company’s management and day-to-day decisions is straightforward.Footnote ¹⁵ In other cases he/she may still signal to both boards the choice that would maximise joint profits.Footnote ¹⁶

Holding a directorship position in two competing firms should thus be prohibited not only when it helps competitors reach a common understanding about future behaviour or detect each other’s deviations from that common understandingFootnote ¹⁷ but also when it enables them to establish a quiet life equilibrium. The question remains, however, as to what is the best way to address this potential information-sharing mechanism.

Absolute prohibitions may be too strict.Footnote ¹⁸ Indeed, based on the circumstances of the relevant markets, lawmakers typically choose between ex-post antitrust enforcement and ex-ante limits or bans. In the first case, personal ties are eradicated if a reduction in competition has been observed, but there is a chance that they escape regulation and enforcement. The pervasiveness of the phenomenon and a history of collusive behaviour in the relevant market might thus suggest the need for an ex-ante ban on interlocking directorates, eventually accompanied by exceptions for situations that, due to the size of the firms involved or to other circumstances, do not pose actual dangers to competition.

11.3 The Uncommon Spread of Interlocking Directorates in the Italian Banking, Insurance, and Financial Sectors

Interlocking directorates have always been a common feature of Italian capitalism. In 1928, statistician Luzzatto Fegiz wrote that

Interlocking directorates and ownership ties have been used in Italy, typically by families, as a mechanism to secure control of the biggest privately owned corporations across different sectors (conglomerate interlocks), as well as a defensive tool against hostile takeovers.Footnote ²⁰ They reinforced the ability of coalitions and weak owners to maintain control and extract private benefits.Footnote ²¹

However, interlocking directorates were a common feature even within industrial and financial sectors, thus affecting the relationship between horizontal competitors. Empirical studies suggest that, after World War II, interlocking directorates were a persistent and endemic phenomenon especially within the Italian banking, insurance, and financial sectors.Footnote ²² The Mediobanca-Generali conglomerate group was at the centre of a galaxy of interlocking directorates, passive investments, and active minority shareholdings among directly or indirectly competing firms and groups. Some authors referred to this characteristic as the circular ownership of Italian listed companies:Footnote ²³ a ‘petrified forest’ where structural links represented an objective limit to the competitive dynamic that these markets could have otherwise exhibited.Footnote ²⁴

Some tables and figures illustrate the impressive thickness and density of the web of personal links between Italian banks, insurance companies, and financial firms before 2011.

According to the Italian Competition Authority, at the beginning of 2008 many banks (roughly 40% of the sample) had no members of their governing bodies interlocked with other banks. However, some had ten or even more interlocked persons. In terms of firm dimension, banks having seven or eight interlocked directors represented more than 47% of the total asset value in the sample. Similar results were found in the insurance and financial sectors.Footnote ²⁵

Table 11.1 gives an idea of the widespread use of interlocking directorates in the Italian financial sector more generally.

Figure 11.1 instead shows the degree of concentration in the Italian banking sector on 31 December 2010. The lines represent interlocking directorates between the 25 largest banking groups in Italy (controlling at the time a total of 130 banks). In a perfectly competitive market, one should expect a picture only made up of isolated points.Footnote ²⁶ Figure 11.1 instead displays only a few isolated firms.

Figure 11.1

Interlocking directorates between the 25 largest banking groups in Italy on 31 December 2010

Source: Federico Ghezzi and Chiara Picciau, ‘Il divieto di interlocking nel settore finanziario: spunti da un’analisi empirica sui principali 25 gruppi bancari italiani’ (2020) 65 Rivista delle società 1659, 1680

11.4 Reasons for the Introduction of an ex-ante Ban on Interlocking Directorates Between Financial Firms

Despite the prevalence of interlocking directorates, the Italian antitrust legislation, which is largely modelled on the EU legislation, had the same limits and gaps of the latter, including the absence of specific tools and remedies to counteract the potential anticompetitive effects of interlocking directorates.Footnote ²⁷ While the Italian Competition Authority could and did intervene on several occasions to sever interlocking directorates,Footnote ²⁸ its enforcement tools were not designed to deal specifically with these inter-company links, nor were they always effective and timely for that purpose.Footnote ²⁹ Yet, the financial sector exhibited at that time all the characteristics that, according to economic theory, would justify the introduction of an ex-ante interlocking prohibition: (i) significant market power of the main interlocked companies; (ii) a history of underdeveloped competition in the affected markets (due to state ownership, past regulation on premiums, interest rates, and credit allocation, as well as evidence of collusive behaviour and information exchanges);Footnote ³⁰ and (iii) stability and pervasiveness of the interlocking directorates’ network, which may have functioned as a reciprocal trust creator in a context of tacit collusion. For this reason, the Italian Competition Authority, in its sector inquiries,Footnote ³¹ public speeches, advocacy activity,Footnote ³² and annual reports,Footnote ³³ strongly favoured creating an ad hoc provision prohibiting interlocking directorates between banking, insurance, and financial companies.

An additional reason for introducing an interlocking ban came at the end of 2008, when the global financial crisis hit Italy with particular severity. While every sector of the Italian economy was affected by a harsh recession, the financial sector experienced the most severe effects. The weakness of the financial sector and its inability to react promptly to the crisis were ascribed, among other factors, to a lack of competition and to the presence of a net of personal and ownership links between banks and other financial firms.Footnote ³⁴

These circumstances led a new ‘government of experts’ – appointed at the end of 2011 to fight the crisis and headed by former EU Competition Commissioner Mario Monti – to introduce a statutory provision dealing with interlocking directorates in the financial sector.Footnote ³⁵

11.5 The Main Features of the 2011 Provision Dealing with Interlocking Directorates in the Banking, Insurance, and Financial Sectors

Article 36 of the so-called ‘Save-Italy’ DecreeFootnote ³⁶ provides that the members of boards of directors, supervisory boards or boards of statutory auditors, as well as the ‘top managers’ of a company or group of companies operating in the banking, insurance, and financial sectorsFootnote ³⁷ shall not accept or hold any such offices in a competing company or group of companies.Footnote ³⁸ These offices are deemed to be incompatible with one another.

Persons holding two or more incompatible offices must choose which one to terminate within 90 days of the appointment. If the term has expired without a decision being made by the interested person, the interlocker is dismissed from all offices by a decision of the corporate bodies in which he/she serves or, if they do not take action, by the regulatory agency supervising the company (the Bank of Italy for banks; Consob for financial firms; and IVASS for insurance companies).

With the entry into force of the prohibition in 2012, hundreds of people had to step down from one or more incompatible offices,Footnote ³⁹ with some of the largest Italian banks and insurance companies, including Intesa Sanpaolo, UniCredit, Mediobanca, and Generali, deeply affected by the measure.Footnote ⁴⁰

Yet, the interlocking ban raised many interpretive issues and was harshly criticised for its inflexibility and lack of clarity.Footnote ⁴¹ For this reason, the regulatory agencies decided to issue a joint set of interpretive guidelines.Footnote ⁴² An initial complication stemmed from the fact that Article 36 does not establish any de minimis exemption for situations in which the same person holds office in two very small firms that have no market power and thus are no threat to competition. The guidelines hence introduced a threshold based on the turnover of the companies involved.Footnote ⁴³ The interlocking directorates’ prohibition applies when at least two of the companies (or groups of companies) involved achieve a net (group) turnover exceeding 30 million euros.Footnote ⁴⁴

According to Article 36, the competitive relationship must be evaluated at the group level. It is reasonable to apply the ban when the interlocked companies are directly competing in the same markets or when one of them controls one or more companies that are in direct competition with the interlocked company.Footnote ⁴⁵ In the latter case, even though there is no direct competition between the interlocked companies, the controlling company may influence the decisions of the subsidiaries that are direct competitors of the interlocked company. However, as the guidelines specify, once a competitive relationship between two financial groups has been ascertained, any interlocking between them is prohibited under Article 36. This applies regardless of the competitive relationship between the interlocked companies or the possibility to (indirectly) affect competition. Imagine two groups that are deemed to compete on the life insurance market because each has a subsidiary operating in that market. Since the two groups are considered to compete in a relevant market, any interlocking directorate between any company belonging to each of the two groups – even that between a bank subsidiary of the first group and a leasing company of the second group, neither of which controls a company in the life insurance market – is prohibited.

The wide reach of the prohibition is partly mitigated by the fact that the guidelines introduced a second de minimis threshold. If the personal tie concerns two non-competing companies belonging to different groups and the turnover of each company is less than 3% of the total turnover of the group, the interlocking ban does not apply.Footnote ⁴⁶ However, it remains true that the scope of Article 36, as applied in the context of competing financial groups, may include interlocking directorates that do not and cannot restrict competition, not even on a theoretical level. This result is at odds with the purpose of the provision and may unduly limit the freedom of a financial company to select the members of the board of directors (or of other corporate bodies) that it deems fittest for the office.

Article 36 applies not only to executive and non-executive members of the board of directors and to top managers but also to members of the board of statutory auditors and, in two-tier governance systems, to members of the supervisory board, whose main task is to monitor compliance with sectoral laws and regulations that are often complex. In order to efficiently perform this activity, many Italian banks and insurance companies used to ‘share’ the same skilled professionals. The broadening of the interlocking ban to these other bodies was thus strongly opposed by both the companies and the professionals involved. While these objections seem understandable,Footnote ⁴⁷ they are not founded in an antitrust perspective. Members of the board of statutory auditors (as well as members of supervisory boards in two-tier systems) participate in the meetings of the board of directors and have an ongoing relationship with the directors and officers of the company. They may hence influence board decision-making and management and are in a position to obtain, share, or release strategic information about the company.

On a final note, the ‘sanction’ set forth in Article 36 is rather peculiar. The burden of non-compliance is on individuals, not on companies. In fact, if the director does not opt in due time for one of the incompatible offices, he/she must step down from all of them. The purpose is, clearly, to provide the interlocker with a strong incentive to promptly put an end to the forbidden multiple appointments. The role given to supervisory agencies is thus a residual one. They must intervene only if (i) the director didn’t opt in a timely manner for one of the incompatible offices and (ii) the interlocked companies didn’t declare the director’s removal.Footnote ⁴⁸

11.6 No More Interlocking Directorates? The Disputed Effects of the Italian Anti-Interlocking Provision

One of the main open questions regarding Article 36 concerns its effectiveness at eliminating interlocking directorates in the financial sector. To be sure, the sanction for non-compliance might not deter interlockings, as it may not always be administered in practice. If the people appointed to multiple incompatible offices do not point out the existence of a prohibited interlocking, it is for the firms involved (and for their boards) to uncover the situation and ask the person to choose between the incompatible offices. This may not be easy. Therefore, a prohibited interlocking may go undetected and unsanctioned because of the silence of the interested party and the inaction of the firm.

It is true that the competent supervisory authorities should also be aware of the prohibited interlocking, given the availability of databases on the composition of the corporate bodies of the supervised firms, and could thus remove the interlocker from all incompatible offices. However, these proceedings are not public. We therefore do not know whether the authorities have actually intervened or instead tolerated violations of the interlocking prohibition that did not raise antitrust concerns in specific cases.

In the absence of official data, a first indication on the possible effect of the interlocking ban is provided by studies that have examined the density of the interlocking directorates’ network in general. Some studies seem to confirm that the interlocking ban has led to a reduction in personal ties among listed companies,Footnote ⁴⁹ even though companies at the centre of the network, which often include financial firms, appear to have maintained some relevant connections.Footnote ⁵⁰ This is probably due to the fact that the interlocking ban applies only within the banking, insurance, and financial sectors. Personal ties between financial and non-financial firms are therefore still allowed. It is also possible that direct personal (and ownership) links between financial firms have been substituted by indirect links, that is, by interlocking directorates involving a non-financial firm in between the previously interlocked financial firms.Footnote ⁵¹

A more specific study on the effectiveness of the interlocking ban, which considered 95 financial firms and adopted a broad definition of ‘competition’, compared the interlocking directorates detected in the financial sector in 2008 with those observed in 2015.Footnote ⁵² According to the study, the number of interlockings in the financial industry appears to have decreased after the introduction of the ban, but not by much, at least with respect to personal ties across different sectors (for example, between a bank and an insurance company or between the latter and a financial firm). Even within sectors interlocking directorates seemed to still be widespread. For instance, the study claims that, in 2008, 60% of banks had personal links with other banks, compared with a slightly lower 50% in 2015, while interlocking directorates among insurance firms only declined from 71% to 45% in the same period.Footnote ⁵³ In other words, the study suggests that the Italian interlocking ban cannot reach all existing interlockings and/or that its enforcement has been vastly ineffective.Footnote ⁵⁴

However, the implications of the study seem to be limited, considering that it adopted a very broad definition of competitor in the financial sector, even broader than the already extensive notion employed in Article 36.Footnote ⁵⁵ Moreover, it does not account for the de minimis exemptions introduced by the supervisory authorities, nor does it explicitly assess whether the interlocked firms were horizontal competitors within each sector.Footnote ⁵⁶

11.7 Interlocking Directorates in the Italian Banking Sector: A Case Study

We have therefore gathered our own data on the number of interlocking directorates after the introduction of the interlocking ban, focusing for the time being only on personal ties between banks.Footnote ⁵⁷ Specifically, we determined whether, at the end of 2018, there were any interlocking directorates between the 25 largest banking groups in Italy.Footnote ⁵⁸ This is a representative sample (consisting of more than 90% of the market both in terms of assets and bank branches).Footnote ⁵⁹ Given the size of the banks and banking groups involved, we can assume that they are competitors in at least some geographic or product markets and that they do not fall within the exemption thresholds.

For each bank, we identified the members of the board of directors, the members of the internal control body, and the general manager. In the case of banks controlled by other firms, we extended the analysis to the parent company (with the exclusion of foreign parent companies, to which the ban does not apply). This process ensured that we captured in our data the potential cases of a director, general manager, or member of the internal control body of a bank or of a bank controlling company holding an analogous office in a rival bank, which are equally prohibited by Article 36.

The results of our analysis, carried out using 31 December 2018 as a reference point, are clear. Among the banks and banking groups considered, there is not even a single relevant interlocking directorate.Footnote ⁶⁰ More precisely, using a concept of competition closer to traditional antitrust principles (comprising companies that are active in the same geographic or product banking markets and their controlling firms), we can conclude that the anti-interlocking provision was meticulously followed, at least as of 31 December 2018 and by the Italian banking sector’s largest players.Footnote ⁶¹

This conclusion helps assess whether the interlocking ban affected the degree of competition in banking markets. A second, more general issue is, in fact, whether the reduction in the number of interlocking directorates has in any way enhanced competition in the relevant markets.

If interlocking directorates can have anticompetitive effects, then the introduction of an interlocking ban should have a positive impact on price (e.g., the interest rates in the banking sector or insurance premiums in the insurance market). Unfortunately, it is very difficult to find any data on the causal relationship between the interlocking ban and prices (or quantities). However, a recent study focusing on the effects of the interlocking ban in the Italian banking sector tried to fill the gap by using a difference-in-differences framework. The study takes advantage of the fact that the legislative reform took place almost unexpectedly, which allows a comparison between the situation before and after the introduction of the ban as a quasi-natural experiment.Footnote ⁶² The results seem to support the procompetitive effects of the interlocking prohibition. In fact, the major finding of the study is that the severance of interlocking directorates resulted in a drop in the interest rates applied to ‘treated’ relationships: that is, credit relationships between a firm and a bank interlocked with another bank that also granted credit to the same firm. The drop was between 10 and 30 basis points vis-à-vis all other credit relationships, defined as ‘controls’.Footnote ⁶³ The study also showed that interest rates became more dispersed after the introduction of the ban, and provided some evidence of a slight increase in the quantity of credit used for treated relationships.Footnote ⁶⁴

11.8 The Ban on Interlocking Directorates and Its Consequences on Ownership Ties

A related issue is whether the reduction in interlocking directorates has caused a corresponding decrease in the strength or number of ownership links that previously connected Italian listed firms, including banking, insurance, and financial companies. The available empirical evidence does not specifically address this question, and economists have actually suggested that further research be undertaken to investigate if and how ownership ties between these firms have changed since the enactment of the ban.Footnote ⁶⁵

Indeed, it could be expected that Italian ownership patterns have been influenced by the anti-interlocking provision. Recall that some ownership links might have been established, in whole or in part, for collusive or anticompetitive purposes. If the interlocking ban makes it more difficult to obtain this result, by severing an important communication channel among previously interlocked companies, the existing ownership connections between those companies could also lose importance.

According to the Italian financial market authority (Consob), concentrated ownership still largely prevailed among Italian listed companies in 2019 (50 of the 228 companies listed on the Italian stock exchange at the end of 2019 were financial companies).Footnote ⁶⁶ However, the number of companies not belonging to any group has continued to increase in the recent past.Footnote ⁶⁷ Moreover, the ‘[d]ata confirm the decline in the number of major holdings owned by banks and insurance companies, especially by Italian ones’.Footnote ⁶⁸ As Table 11.2 shows, the mean stake held by Italian banks and insurance firms in Italian listed companies has remained more or less stable in the period between 2010 and 2019, with only a slight increase after 2015. However, the number of holdings that these financial firms have in Italian listed companies has sharply decreased from a total of 44 in 2010 to six in 2019.

These data do not allow us to draw any sure conclusion with respect to the effect of the interlocking ban on the ownership patterns in the Italian financial sector. The data merely show that banks and insurance firms have curtailed their investments in listed companies, but do not enable us to establish the cause of this reduction. More specifically, on the basis of the available data it is not possible to ascertain whether the shareholdings that have been sold by banks and insurance firms were previously held in interlocked companies. Nor is it possible to determine whether the severance of the ownership connections corresponds in any way to a reduction in personal ties. Nevertheless, and despite a significant degree of stability in the ownership of Italian listed firms,Footnote ⁶⁹ the data show that transformations have been underway precisely with respect to some of the financial firms (Italian banks and insurance companies) to which the interlocking prohibition applies.

Identifying what propelled these changes remains in any case difficult. The task is made even more daunting by the fact that at least some of the aforementioned trends seem to have a more distant origin. For instance, even before the introduction of the 2011 interlocking ban some commentators had pointed out a decline in ownership links,Footnote ⁷⁰ and the same is true with respect to the decrease in the number of firms belonging to pyramidal or other groups and to the reduction of the average stake held by controlling shareholders in listed companies.Footnote ⁷¹ As a matter of fact, these and other transformations might also be explained by a number of other reforms that have been enacted in Italy in recent decades: from the privatisations of the early 1990s to the introduction of new legislation on issuers and public offerings, among others.Footnote ⁷² Therefore, while one might have a hunch that the interlocking ban did have some effect on ownership ties, there is yet no sure way to tell.

11.9 Interlocking Directorates and the Competence of Financial Firms’ Boards

Unambiguous conclusions cannot even be reached with respect to the impact of the interlocking ban on the composition and competence of corporate boards in Italy. Some commentators have pointed out that interlocking directorates may serve the positive function of enabling companies to attract scarce expertiseFootnote ⁷³ that might be valuable especially when firms undergo complex operations (such as a listing or a takeover);Footnote ⁷⁴ in highly regulated businesses (such as the banking, insurance, and financial sectors);Footnote ⁷⁵ or for small firms.Footnote ⁷⁶ One may thus fear that an interlocking ban would prevent companies from selecting and sharing the most competent and talented professionals, in a context where expertise, competence, and talent are rare.

While this argument is surely appealing and has some foundation, it is impossible to determine whether the Italian anti-interlocking provision has actually diminished the overall competence of the governing bodies of companies operating in the financial sector. The interlocking ban is just one of the many provisions that, over the last decades, have affected the composition and diversity of corporate boards.

For instance, the Italian Consolidated Law on Finance,Footnote ⁷⁷ which was introduced in 1998 and amended several times afterwards, limits the number of directorships and control offices that members of the internal control bodies of issuers, including many financial firms, may hold at the same time.Footnote ⁷⁸ It also provides that issuers must have at least one independent director or two independent directors when the board is composed of more than seven membersFootnote ⁷⁹ (the fraction raises to one-third of the board for firms adopting the one-tier corporate governance model).Footnote ⁸⁰ For certain large listed companies, an even higher number of independent directors is recommended, under a comply-or-explain approach, by the Italian Corporate Governance Code.Footnote ⁸¹

Italian issuers must also make sure that at least two-fifths of the members of the board of directors and of the board of statutory auditors (or the supervisory board in two-tier governance systems) consist of the less-represented gender (typically, women).Footnote ⁸² Other specific personal requirements apply to board members of financial firms.Footnote ⁸³

Clearly, these provisions, which have only been briefly sketched here, may already limit the freedom of financial firms to select which people to appoint to their governing bodies.Footnote ⁸⁴ As a result, it may be hard, if not impossible, to discern whether any observed change or decline in the competence of the boards of financial firms is due to the interlocking ban or to these other requirements.

However, upon closer inspection, several reasons indicate that the interlocking prohibition does not (significantly) prevent financial firms from acquiring and sharing valuable talent and competence.

First, if competence and skill are hard to find, the solution should be investing in training and education, inside and outside the firm, not enabling the same person to serve on a large number of boards at the same time. Indeed, nudging firms to go beyond the usual panel of candidates for a job may foster investment in managerial and technical skills by aspiring candidates. Such investments are particularly important because even the most talented and competent professionals may not do their job well when overworked by holding offices in many different firms.Footnote ⁸⁵

Second, one may very well doubt that competence and expertise are so hard to find. Education rates have improved across the developed world,Footnote ⁸⁶ and financial firms increasingly draw from a wider set of candidates, including women and foreign professionals.Footnote ⁸⁷ With respect to the Italian case, one must also consider that interlocking directorates are prohibited only between competing banking, insurance, and financial companies or groups. The prohibition thus does not prevent financial firms from sharing skilled and talented board members with non-financial companies or groups.Footnote ⁸⁸

To be sure, one might argue that financial firms do not simply need competent and talented individuals, but people with very specialised knowledge and expertise, and with the additional personal requirements set forth in prudential regulation. Finding the right people may therefore be particularly difficult. Yet, consider that, according to the guidelines on the application of Article 36, the notion of ‘top managers’ – to whom the interlocking ban applies, in addition to board members and members of the internal control body of financial firms – only comprises the general manager and the manager responsible for the corporate financial reporting of the firm.Footnote ⁸⁹ This leaves out a variety of highly specialised, high-ranking officers and managers who could serve as directors or members of the internal control body of other financial companies. If these people do not hold multiple offices is because financial firms generally do not allow it (consider that top managers typically are employees). In short, the Italian interlocking ban, by itself, does not seem to significantly limit the ability of financial firms to acquire skills that are already fruitfully employed elsewhere.

Third, and most importantly, there is yet no empirical support for the proposition that limiting interlockings has a negative impact on the competence of the board or on firm value. Similar concerns have been raised in the past with respect to gender quotas and have been disproven by empirical studies that show that increasing board diversity, also with respect to gender, generally improves performance.Footnote ⁹⁰ Indeed, even artificially intelligent algorithms, programmed to select the best performing directors for a given firm, suggest that diversity is the right way to go.Footnote ⁹¹

In any case, the impact of these reforms is still uncertain and needs to be evaluated over time, in light of the perhaps unexpected effects of many of these new measures. For instance, while the introduction of gender quotas in listed firms has increased the number of women sitting on corporate boards, it has perhaps also led to the spread of female interlockings across issuers.Footnote ⁹² Issuers have started to appoint women on company boards, but they continue to draw from a limited set of candidates who sit on the boards of many listed firms at the same time. It remains to be seen whether this is just an adjustment phase or a new future trend.

11.10 Conclusions

Interlocking directorates among competitors may facilitate collusion and endanger competition. In Italy, they typically accompanied direct and indirect ownership ties and were a systemic and widespread phenomenon particularly in the banking, insurance, and financial sectors. This is why, in 2011, the Italian lawmaker introduced an ex-ante prohibition on interlocking directorates in the financial sector. Despite the breadth of the prohibition, which arguably even covers situations in which a threat to competition is not likely or lacking, some empirical studies questioned the effectiveness of the ban, claiming that relevant personal links persisted, together with possible anticompetitive effects.

Using the banking sector as a case study, we have argued instead that there is some indication that the Italian interlocking ban has met its goal. The data we collected on interlocking directorates among the 25 largest banking groups in Italy on 31 December 2018 show that, on that date, there were no relevant personal ties among the largest banks and banking groups. This result is in line with other studies that claim that, after the introduction of the prohibition, bank lending rates in Italy fell, which should indicate that the interlocking ban had a procompetitive effect.

However, further research is needed not only to ascertain the effects of the ban but also to determine whether it had any impact on the ownership of Italian firms and on the competence of their governing bodies. While it is highly unlikely that board competence was significantly affected by the ban, ownership ties might have changed as a result. Indeed, it is curious that the Italian lawmaker decided to ban interlocking directorates without also intervening on ownership ties, such as minority shareholdings. However, at least in some cases the interlocking ban might have killed two birds with one stone.

12.1 Introduction

Common ownership is the phenomenon of ownership of natural competitors in an industry by an overlapping set of institutional investors. Ignited by an empirical study of US airline competition under common ownershipFootnote ¹, a recent literature exploring the possibility of anticompetitive effects caused by common ownership has not only raised long dormant conceptual questions in corporate finance, organizational and labour economics, and industrial organizationFootnote ² but also raised questions in antitrust law,Footnote ³ corporate law,Footnote ⁴ and securities lawFootnote ⁵. Moreover, the original common ownership paper has triggered a policy discussionFootnote ⁶ that arguably challenges the business practices of the asset management industry as we know itFootnote ⁷.

In a review of the economics and finance literatureFootnote ⁸ on common ownership in 2018, I laid out the conceptual challenges this line of inquiry brings to the surface, and suggested directions for future research. In a second review,Footnote ⁹ I reported on the exploding empirical literature that responded to some of these directions, but mainly documented robustness of the basic premise: the effects of common ownership on corporate behaviour and market outcomes have now been documented in a host of different industries and settings and using a variety of different methodologies. Further, many early criticisms of the literature were proven incorrect and have been withdrawn.Footnote ¹⁰

In this paper, I analyse conceptual challenges that the original ‘airlines’ paper left for future research to address, and which many believed would need to be addressed before a consideration of policy changes would be appropriate. These questions include: how would estimates or (anti-)competitive effects of common ownership of horizontal competitors be affected if agency problems, informational frictions, and organizational complexities were considered? Are estimates of horizontal common ownership links on product market outcomes affected by considering the effect of common ownership links between vertically related firms as well? Are there methods to ensure that effects are truly causal? And perhaps most importantly: how does the lifting of data limitations change researchers’ ability to measure effects of common ownership on firm behaviour and market outcomes? Given the data limitations, is there a consensus now on how much common ownership there is? The first part of this paper brings the reader up to speed by addressing these questions.

In the second part of the paper, I develop a framework and use it to evaluate policy proposals others have made to address the antitrust and governance challenges posed by the rise of ‘common ownership’ and the related rise of ‘passive investing’. The objective of that discussion is primarily to provide a clear framework to help organize and analyse the various proposals, and to view their conceptual commonalities and differences from an economic perspective. The framework is meant to clarify which directions have which likely costs and benefits, and which uncertainties come with which approach. The objective is not to engage with detailed – though important – issues of implementation, or to provide a final answer to what should or shouldn’t be done, but to help shape a more informed and perspicuous debate. As such, the present paper is not meant to substitute for or contain the state of debate between the authors of various proposals. Instead, I focus on overlooked dimensions relevant to the policy debate that arise from the recent economics and finance literature’s application to law and policy.

12.2 New Conceptual Breakthroughs

In this part, I explore how our conceptual understanding of important features of common ownership has advanced by recent research responding to questions that arose with the publication of Azar, Schmalz, and Tecu’s paperFootnote ¹¹.

12.3 Discussion of Policy Proposals

Legal scholars have made a number of proposals, taking at face value that there are competitive effects of common ownership and/or governance problems caused by large institutional investors, and that those problems are as widespread as the phenomenon of common ownership itself – i.e. across all firms and industries held by institutional investors. An interesting and worthwhile discussion of these proposals’ strengths and weaknesses followed in recent years. What I find to be missing in the discussion is a clear and simple economic framework to organize that discussion. I aim to provide such a framework. I will then illustrate how some of the more prominent proposals fit in the framework, and which arguments and questions naturally arise that are thus far missing from the debate.

12.4 Conclusion

As this review showed, essentially all dimensions regulators have viewed as roadblocks to regulation have been addressed by recent research: we are assured the measured correlations between common ownership and higher product prices are most likely to have a causal interpretation and that they are, in particular, not driven by endogenous market shares. We know realistic mechanisms exist, and that agency conflicts are most likely a key driver of the empirical facts the literature has uncovered. Similarly, organizational complexities are a feature of models that can make nuanced predictions that are verified in the data. When vertical links are considered, the measured effects of common ownership on prices get stronger. The same is true when more complete and more accurate ownership data are used.

Which of the proposals should be implemented? The bottom line of this paper is that this is not a question that can be answered with purely economic analysis. As this review illustrated, economic research has made progress – but not nearly enough to be able to fully predict the effect of policy proposals that aim to fundamentally change the asset market equilibrium, and hence macroeconomic performance as well. Rather than a question that can and should be decided by economic technocrats, a broader question is whether we want an economic system featuring a ‘Problem of Twelve’Footnote ⁵⁹ in which a very small number of players effectively controls large swathes of American industry. In the past, politicians, such as Pecora, saw no need for structural estimates of competitive effects across all industries before taking action. Instead, it was clear to them that increasingly centralized control over business was not conducive to a thriving capitalist economy. Whether the resulting laws were overly restrictive (including the 10% limits for single funds) appears to be more a question of political convictions and personal incentives than rigorous economic analysis. A further concern with calls for more research is that when we insist on having ‘quasi-experimental’ evidence to evaluate whether a given policy is good, we are limiting the scope of analysis and allowable policies to those we have tried before. This restriction is obviously self-defeating for a phenomenon that has not occurred previously. In other words, I believe that we are at risk of relying on economic analyses that miss the forest for the trees. The danger is that economists can have the effect of hindering regulation by pointing to uncertainties, while omitting the fact that anticompetitive harm continues, at ever increasing scale, while the debate continues. In sum, I positively view it at this stage as determined rather by political processes than further economic analysis – and to the extent these political processes reflect the interests of ordinary citizens I normatively agree with that notion to some extent. The unarguable part of the debate appears to me that transparency on who controls corporate America should be fostered. The only grounds to obscure the facts to me appear to be a desire to deal with the ‘investor-protection challenge’ of the twenty-first century rather than with a concern about competition and governance.

13.1 Introduction

There is compelling evidence that both concentration and profitability in oligopolistic industries have increased over the past two decades. Over roughly the same time period, the concentration of shareholding in the hands of the largest institutional investors has dramatically increased, with a corresponding increase in the degree to which investors (such as Vanguard, State Street, and BlackRock) own large equity stakes in competing portfolio companies. A number of authors have argued that the growth in this ‘common ownership’ has caused the increase in oligopoly profits and have proposed a variety of policy responses.

In this paper, we review the available evidence. We argue that as of now (a) the evidence that common ownership is the driving force behind the increasing oligopoly profits is unconvincing, (b) there are plausible competing explanations for the correlation between profitability and common ownership. As a result, (c) regulatory intervention directed against common ownership is not currently warranted, given the significant costs of such intervention.

This paper proceeds as follows. In Section 13.2 we provide an overview of the evidence that concentration and profitability have increased. In Section 13.3, we consider the evidence that increased common ownership is the cause of the increase in profitability. Section 13.4 considers alternative explanations for the correlation between increasing concentration, increasing profitability, and increasing common ownership, along with the available evidence in support of these alternative hypotheses. Section 13.5 considers the policy implications of the current state of play.

13.2 The Evidence: Concentration and Profitability

Over the past two decades, major industries in the US and worldwide have become more concentrated and, over the same time period, more profitable. The US Council of Economic Advisors 2016 ‘Issue Brief’ documents that between 1997 and 2012 changes in the revenue share of the largest firms in a variety of industries have ranged between −2% and 11.4%, with the majority of those firms showing substantial increases.Footnote ¹ More recently, a study by Bajgar et al. used firm-level concentration measures and found that the share of industry sales due to the 10 largest companies in 10 European economies increased on average by 2% in manufacturing and 3% in non-financial services from 2001 to 2012.Footnote ² The authors conclude that there has been a clear increase in industry concentration in both Europe and North America (from 2000 to 2014) by between 4% and 8%, with the absolute increase being somewhat greater in North America.Footnote ³

Industry-specific studies support and augment this broad picture. To mention just a few, a 2010 Congressional Research Service study by Shields found that between 1971 and 2002 dairy industry concentration increased in eight of the nine agricultural industries studied.Footnote ⁴ Gaynor, Ho, and Town found that between the early 1990s and 2006, the average HHI for hospital markets increased by about 50% to approximately 3,200, substantially above the DOJ/FTC Guidelines’ 2,500 cutoff measure of high concentration.Footnote ⁵ With respect to mobile wireless, concentration has steadily increased over time, highlighted by the recent successful acquisition of Sprint by T-Mobile, that left the US with only three facilities-based wireless carriers having a national footprint.Footnote ⁶

Interestingly, there are significant exceptions to this overall pattern in some high-profile markets. Froeb and Werden found that US airline route-level HHIs slightly decreased over their broad period of study, ranging from 1984 through 2012.Footnote ⁷ Likewise, a study by the UK Social Market Foundation found little indication of increasing concentration in most UK consumer markets over the period of 2000 to 2017.Footnote ⁸

The link between concentration and profitability has been more contested. Antitrust law, scholarship, and policy have all been based on a link between the two.Footnote ⁹ Indeed, that link has been one of the foundations for the DOJ-FTC Horizontal Merger Guidelines. The Guidelines, in turn, were highly influenced by the ‘Structure-Conduct-Performance paradigm’ in Industrial Organization.Footnote ¹⁰ Convincing empirical analysis has historically been sparse for a variety of reasons that modern industrial organisation scholarship describes. For one thing, concentration is not an exogenous force; as a result, we cannot be certain of the direction of the concentration-profitability relationship. For another, published concentration measures are often not coincident with the relevant economic markets that underlie industrial organisation economics.

Studies looking for evidence from earlier periods have found weak or no correlation between these variables.Footnote ¹¹ Indeed, the uncertain connection between industry concentration and anticompetitive outcomes is one of the reasons why, in the Guidelines, HHI levels are the starting point for further investigation and do not, on their own, trigger challenges.Footnote ¹² Post-2000, however, the evidence of this link is more robust. A Swiss finance Institute Research Paper by Grullon et al, supports the correlation between concentration and profit margins.Footnote ¹³ The authors find that more than three-fourths of US industry have experienced an increase in concentration levels over the past two decades and that those industries with the highest increases in concentration have seen higher profit margins.Footnote ¹⁴ The authors credit these changes in part to reduced antitrust enforcement and increased technological barriers to entry.Footnote ¹⁵

Along the same lines, an informative paper by Gutierrez and Philippon shows clearly that there has been an increase in the HHI in most US industries and correspondingly an increase in profit margins as measured by the Lerner index.Footnote ¹⁶ The evidence that, since 2000, increased concentration is correlated with increased profitability suggests that an adequate theory must explain two things: why is there a correlation between concentration and profitability? And why has there been a strong(er) correlation post-2000 than pre-2000? Jonathan Baker tells a compelling story. According to Baker, large businesses have profited by using sophisticated pricing algorithms and customer data to secure substantial, persistent advantages over smaller players.Footnote ¹⁷

13.3 Has Common Ownership Led to Higher Profit Margins? The Claims and a Critique of the Evidence

Much of the current debate results from the extraordinary attention attracted by Jose Azar, Martin C. Schmalz and Isabel Tecu’s (AST) widely read working paper (now published in the Journal of Finance) that claims the increased common ownership by diversified investors has caused a significant increase in the price of airline tickets.Footnote ¹⁸ A related paper claims that the same effect is found in commercial banking.Footnote ¹⁹ In response to the dramatic claims of these papers, there has been a huge outpouring of theoretical and empirical research and analysis. In this section, we briefly review that research.

13.4 Alternative Explanations for the Correlation Between Concentrated Markets, Higher Profits, and Concentrated Common Ownership

If the increase in common ownership is not the cause of the increase in concentration or profitability, what might that cause be? In this section, we consider some alternative explanations.

13.5 Current Policy Implications

The rise of the large institutional investors over the last 30 years has been the biggest ‘story’ in corporate governance.Footnote ⁴¹ With AST’s pathbreaking work on the competitive effects of common ownership, we are at the beginning of what promises to be a fascinating investigation of the competitive effects of common ownership. In this section, we consider some of the policy implications of this new debate.

Some who are convinced by the AST analysis have proposed systemic solutions to what they believe is a systemic problem. Einer Elhauge argues that the current common ownership by the largest institutional investors constitutes a continuing violation of Section 7 of the Clayton Act and possibly Section 1 of the Sherman Act and advocates for government enforcement actions and private class actions.Footnote ⁴² As we have discussed at length elsewhere, we disagree with Elhauge’s legal analysis.Footnote ⁴³ For what it is worth, we are not aware of any enforcement actions or private class actions embracing Elhauge’s legal theory.

Posner, Scott Morton, and Weyl are, likewise, convinced by the AST analysis and have proposed an alternative to complete divestiture. In their view, the danger posed by common owners is so severe that they should be put to a choice: divest all but one firm in each oligopoly; or limit holdings to less than 1% and pre-commit to governance passivity by sterilising votes.Footnote ⁴⁴ Given the huge benefits of index investing for ordinary investors, and what we view as the generally positive role that the largest institutional investors play in corporate governance, we think that the Posner et al policy change is not warranted by the evidence gathered to date, and would cause significant harm.

For both Elhauge and Posner et al.’s proposals, the difficulty of replicating the AST results in other industries, discussed above, undermines the case for a global/systemic reform. Rather, any intervention addressing the anticompetitive effects of common ownership should require a specific showing of such effects, based on particularised industry findings. Although common ownership is a market wide phenomenon, there is no evidence that the supposed anticompetitive effects of common ownership obtain in every concentrated market.

Although unconvinced that common ownership undermines competition systemically, common ownership does raise significant antitrust issues that enforcement authorities should investigate. First, in oligopolies, shareholders – whether they are common owners or undiversified owners – can indisputably play an anticompetitive role. They can, for example, organise competitors into a ‘hub and spokes’ conspiracy and, if they do so, will violate Section 1 of the Sherman Act and be subject to criminal sanctions and treble damages.Footnote ⁴⁵ Likewise, there are a variety of other plausible coordinated scenarios in which shareholders can cause competitive harm, such as if shareholders act as a trustworthy conduit for communication among competitors, or advocate an industry-wide anticompetitive compensation structure and even possibly as the spreader of anticompetitive practices.Footnote ⁴⁶ In each of these cases, depending on the factual context, shareholder conduct may violate existing antitrust law and be subject to sanctions.

Finally, there may be implications for merger policy. The European Commission, in the Dow/DuPont matter, suggested that, in light of the AST analysis, the treatment of traditional measures of market concentration, such as the HHI, should be supplemented by the MHHIΔ to take into account the competitive effects of common ownership. This is unwarranted for a number of reasons beyond the preliminary nature of the AST results. First, in mergers of commonly owned firms, while incorporating MHHIΔ may affect the threshold at which enforcement officials look closely at mergers on the grounds that HHI understates the pre-merger competitive condition, it will likewise reduce the significance of any increases in HHI resulting from the merger (on the same grounds). Second, while focusing on MHHIΔ points in the right direction in the review of mergers between a large incumbent and a non-commonly owned maverick firm, merger policy already subjects such mergers to enhanced scrutiny. In such cases, focusing on MHHIΔ adds little.

But suppose that the relation between increased industry concentration, increased oligopoly profits, and increased common ownership since 2000 is all the result of a common cause. What if it turns out that the rise of superstar firms, driven by changes in information technology, network effects, regulation, and/or globalisation, is responsible for the simultaneous increase in concentration and common ownership? What are the implications?

This will be an important debate going forward. Some will argue that the rise of superstar firms should justify stricter merger control.Footnote ⁴⁷ Others, however, will argue that the rise of the superstar firms – firms that become and remain superstars because they reduce costs and increase output at the same time as they increase profits – calls into question the fundamental assumptions of current merger regulation. If the superstar firm hypothesis is confirmed, these will be among the most important debates of the next era of antitrust enforcement.

14.1 Introduction

The recent literature on the anticompetitive effects of parallel holdingsFootnote ¹ can be seen as one of the many expressions of both a scientific and societal concern for the increasing economic influence acquired by big investment management corporations throughout the world.Footnote ² The findings of the authors of the seminal papers in this areaFootnote ³ – although highly contested –Footnote ⁴ may highlight the existence of a (present or potential) antitrust issue, which still needs to be understood in all its complexity.

If we widen our perspective from the mere price effects correlated to the presence of parallel holdings to a larger range of variables, we may soon understand that the parallel holdings antitrust issue is nested within a complex set of systems and subsystems, which cannot be ignored while studying the possible policy reactions to the antitrust issues arising from common ownership. For instance, the raise of common ownership has occurred during a period of increasing concentration in product marketsFootnote ⁵ and of significant changes in corporate governance worldwide.Footnote ⁶ The latter also aimed at pursuing ESG objectives that were previously considered as mere externalities in corporate governance.Footnote ⁷

Even if common ownership was unequivocally proven to be correlated with price increases, a well-designed policy reaction would not necessarily need to target parallel holdings or parallel holdings alone. On one hand, dismantling parallel holdings may jeopardise the efficiencies of such an investment system as well as their role in promoting ESG objectives.Footnote ⁸ On the other hand, it would need to be proven that a shift in corporate ownership structure can effectively improve competition ceteris paribus (i.e. despite highly concentrated product markets and despite a rather unpromising promotion of stewardship).Footnote ⁹ On top of that, given the massive presence of the biggest investment corporations worldwide, dismantling of parallel holdings and the entry of new foreign investors may trigger a series of geopolitical consequences among the three main areas of economic influence in the world (i.e. the US, Europe, and China) whose long-term effects may be hard to predict.Footnote ¹⁰

14.2 Parallel Holdings: A New Antitrust Frontier Also for Europe?

The academic debate on the anticompetitive effects of common ownershipFootnote ¹¹ started as a pure US antitrust one.Footnote ¹² Nevertheless, the growing weight of Indexed Funds’ equity holdings in the ownership structure of USFootnote ¹³ and non-US corporationsFootnote ¹⁴ seems to have a substantial impact on the overall structure and functioning of the economic, financial, and industrial systems worldwide. Hence, such a debate appears to be inherently global in nature – not only from the point of view of pure financial economics or industrial organisation theory but also for its wider policy and socio-political implications. Recent papers have argued that common ownership may produce significant wealth transfers and therefore affect the welfare of shareholders and consumers alike.Footnote ¹⁵ It may also create a sort of monopsony on the labour markets, while pushing managers to pursue cost reduction strategies, which in turn often entail dismissals and/or salary cuts.Footnote ¹⁶ The other flip of the coin is that large investment corporations – apart from their renowned capability to contain investment management costs and to promote investment efficiency and investment democracy through indexing –Footnote ¹⁷ have also become increasingly able to advance effectively ESG objectives.Footnote ¹⁸ Hence, common ownership by institutional investors seems to interact with a very wide range of stakeholders – maybe society as a whole – which makes any potential issue surrounding parallel holdings a matter of complex balance of conflicting interests.

The importance of the welfare effects (and competitive harm) attributed to common ownership seems to contrast with the limited decisional power historically enjoyed by common owners in relation to the strategic choices of most of their target corporations.Footnote ¹⁹ In most cases, none of the common owners seems to be able to exercise any control right on their target corporations, at least not on a standalone and stable basis.Footnote ²⁰ This is a classic situation where owners could, at least in principle, acquire a degree of influence on corporations through coalitions. But, even if that was to become the case, the outcome of coalitions is often hard to predict and their stability can be challenged also by very subtle changes in the equity holdings.Footnote ²¹ This sort of influence, or ‘potential influence’ without proper control, is notably a conundrum for antirust and especially for EU competition law.Footnote ²²

The unprecedented situation triggered by common ownership does not puzzle only antitrust and competition law experts. It is also a corporate law and governance primary issue.Footnote ²³ In contrast with what had been foreseen by Hansmann and Kraakmann in their famous essay on the ‘end of the history of corporate law’, the world financial and industrial systems, including the European ones, have not converged to a full extent and univocally towards what the authors call ‘the standard model’.Footnote ²⁴ Today, the US corporation looks increasingly dissimilar from that ‘standard model’ and has not become the end point of any other corporate law model in the world.Footnote ²⁵ Hence, a degree of legal diversification is actually the standard.

In the presence of parallel holdings, corporate directors may not be as strong as they used to be (or at least as they used to be depicted) when there was more significant capital dispersion.Footnote ²⁶ This may be the case especially when common owners adopt active strategies in relation to some of their holdings.Footnote ²⁷ But even if they stay passive the dimension of their blockholding – the now famous ‘800-Pounds gorilla’Footnote ²⁸ – may pose a potential threat to corporate directors’ inefficient behaviours.Footnote ²⁹

Despite the primary importance of Indexed Funds in the US, the rest of the world still hosts a completely different corporate reality in terms of ownership structure. For instance, continental European corporations still are characterised by the prevalence of one or a few controlling shareholders – very often an entrepreneurial family.Footnote ³⁰ But think also of the People’s Republic of China’s (PRC), whose economic system may be described as ‘socialism with many capitalists’.Footnote ³¹ In PRC, since the end of the 1970s many private corporations have emerged as significant alternatives to the traditional State-owned industriesFootnote ³² – including today several cross-border active high-tech giants, such as LenovoFootnote ³³ – while the Chinese Communist Party has acquired a substantial influence on the country’s private business reality.Footnote ³⁴ An influence which seems to have increased after the mixed-ownership reform of State-Owned Enterprises (SOEs).Footnote ³⁵ The PRC example shows that the combinations between institutional and political variables can reach levels of creativeness that perhaps we could not have expected in the 1980s, when the world was largely polarised between the capitalist and the socialist/communist models.

Even in a globalised world, the diversity of the local business and organisational models is mindboggling. Therefore, even if Indexed Funds invest worldwide, their investments in such diversified corporate environments cannot be understood in the light of a monolithic theoretical framework. The real challenge ahead is not only understanding the potential competitive consequences deriving from parallel holdings but also adapting any potential policy response in the most adequate way possible to different business and organisational realities. This may require avoiding early and easy fixes and considering the wider social and political variables triggered by the intertwining between finance, industry, and corporate governance.Footnote ³⁶

14.3 Non-Controlling Equity Holdings and Minority Shareholdings: Pieces of a Scattered Puzzle

Indexed Funds, a type of Exchanged Traded Fund – had hardly ever attracted negative public comments until 2016.Footnote ³⁷ Indexed Funds are based on an ingenious intuition by John ‘Jack’ Bogle who published copious essays and books in defence and for the promotion of his investment strategies.Footnote ³⁸ The merits of indexing as an investment technique is proven by its success: Indexed Funds have become prevalent over alternative international financial investments in 2003.Footnote ³⁹ Bogle’s investment strategy consists of mimicking financial indexes: Indexed Funds have been renowned for being rarely outperformed by alternative investment strategies.Footnote ⁴⁰ Their expanded diversification system offers ideal risk containment; the prevalence of passive investment strategies for such holdingsFootnote ⁴¹ entails very low management costs, with consequent lower fees for investors.Footnote ⁴²

Despite parallel equity holdings being mostly qualified and treated as passive investments,Footnote ⁴³ they are significantly different from other forms of non-controlling minority holdings and/or passive investment, such as minority non-controlling holdings by competitors or cross-shareholdings in the same product market or across the supply chain – on which EU competition law literature abounds.Footnote ⁴⁴ They also differ from the so-called circular ownership, a situation in which each firm on a given relevant market owns equity in at least one of its competitors.Footnote ⁴⁵

Even though they may at times have anticompetitive effects, holdings in a competing firm (i.e. nothing to do with holdings deriving from investment based on indexing, which are normally in non-competing firms) can be supported by a financial and/or an industrial rationale.Footnote ⁴⁶ On one hand, a company may be willing to invest in one’s competitor simply because that competitor is well positioned for meeting the market demand and such information is easily retrievable by other actors within the same industry.Footnote ⁴⁷ On the other hand, a company may want to have a stake in one of its competitors in order to increase collaboration, for instance for the purpose of innovation – i.e. pursuing dynamic efficiency.Footnote ⁴⁸ There may also be cases where two competitors set up a jointly owned corporation expressly as an R&D joint venture and their interests are (limitedly) aligned as equity owners. Finally, the acquisition of non-controlling equity holdings may anticipate an intention to merge, such as in the case of certain types of earnouts M&A operations.Footnote ⁴⁹ None of such strategy is present in parallel holdings – where the inherent rationale is purely financial and is based on maximum diversification through indexing.

Secondly, the influence of the investment performance of one single minority holding in a competitor’s balance sheet may be rather significant, as it will be diversified only against the industrial activity of the acquirer.Footnote ⁵⁰ By contrast, for an indexed fund, the relevance of the investment performance of one single non-controlling equity holding will fade against the myriad of additional holdings represented in the reference index that the investment fund aims at mimicking. Therefore, the kind of action in which a competitor is ready to engage is likely to be different from the actions undertaken by an indexed fund. As a matter of fact, when Indexed Funds have been active, they have normally focused on wide common objectives and not on one single holding.Footnote ⁵¹ For the reason outlined above, one cannot consistently employ the theoretical findings on minority holdings in a competitor for tackling the problems which might arise in the case of parallel holdings.

14.4 The Peculiarities of Indexed Funds’ Parallel Holdings in the EU Corporate Context with Prevalence of Traditional Controlling Shareholders

In the US, the phenomenon known as common ownership arose from the ashes of a corporate world that was previously characterised by dispersed corporate ownership structure.Footnote ⁵² But dispersed ownership structure is not a common feature of European corporations,Footnote ⁵³ with the exceptions of the UK and Ireland.Footnote ⁵⁴ In certain industries dispersed corporate ownership structure is present also in continental Europe – although far from being the dominant model.Footnote ⁵⁵ Therefore, if we purport to understand the impact of Indexed Funds’ investments on competition in European product markets, the case for a potential competitive harm deriving from common ownership needs to be analysed considering two alternative hypotheses: dispersed and concentrated corporate ownership structure – concentrated ownership being the most commonly found in practice. The literature on competitive harm in case of dispersed corporate ownership structure is already abundant; yet an overall agreement over the existence of such harm has not been reached yet.Footnote ⁵⁶ Had an agreement on such harm to be reached, as we will see, US literature would not necessarily be of great use in the EU institutional context. Neither from a legal,Footnote ⁵⁷ nor from a policy perspective.Footnote ⁵⁸

When it comes to concentrated corporate ownership structure, the chances that the same mechanisms described in the US literature on common ownership apply within the EU corporate context are extremely remote. In a system characterised by concentrated corporate ownership structure directors will chiefly respond to the controlling shareholder(s).Footnote ⁵⁹ As recently shown, this will occur despite the emphasis put by stewardship codes on the role of institutional investors in corporate governance in those systems characterised by concentrated ownership structure.Footnote ⁶⁰

Yet, one may still hypothesise a residual anticompetitive role of common owners in a context of concentrated ownership. For instance, one might want to investigate whether controlling shareholders’ (industrial families or other institutional owners) interests may occasionally or stably be aligned with those of Indexed Funds towards a collusive objective.Footnote ⁶¹ But the role of Indexed Funds in these cases could only be marginal. For instance, one might think of a potential role of investment funds in relation to information exchanges within the same industry, so to ease up the collective action problems that are normally seen in collusive contexts with many players.Footnote ⁶² But such a hypothesis, unless corroborated with sound empirical research, should be considered as little less than fantasy for the time being.

Another difference between the US and the EU context is that Indexed Funds may not employ the same indexes in the two economic environments. In fact, the equivalent of the S&P index in Europe is the FTS Eurofirst 300. Yet, it does not seem to enjoy much success when it comes to indexing. The EU is rather diversified both from an institutional and industrial perspective – which makes regional areas extremely differentiated from an investment perspective.Footnote ⁶³ Therefore, for investors, such as BlackRock, it makes sense to create Indexed Funds based on country-based indexes (e.g. iShares MSCI Germany Index Fund (NYSE: EWG)).Footnote ⁶⁴ Indexed Funds’ overall investment techniques may be characterised by more significant fragmentation in the European context, as compared to the US one. Such a fragmentation may reduce the degree of coordination vis-à-vis potential interventions within the governance of the target companies. But, as already mentioned, what represents one of the core obstacles in tackling potential issues deriving from common ownership is the legal variable.

14.5 EU Competition Law and the Absence of an Adequate Legal Framework for Parallel Holdings

The hypothetical competitive harm scenario depicted by the common ownership industrial organisation literature encompasses two alternative explanations.Footnote ⁶⁵

According to one of them, the modification of the investees’ corporate ownership structure, subsequent to the increase in common ownership, may induce a behavioural change in the members of the target companies’ boards of directors. Corporate directors would tend to compete less vigorously in order to please common owners – without actually being actively encouraged or forced to do so, i.e. consistently with a passive investment scenario. It has been hypothesised that such behaviour may be reinforced by the fact that directors are remunerated on the basis of the performance of the market, instead of on the basis of the performance of the corporation they serve.Footnote ⁶⁶

A completely different explanation considers an active involvement of Indexed Funds, directed to curb target companies’ directors’ behaviours towards a collusive approach. As highlighted by Hemphill and Kahan, this would require the generation of the appropriate strategies, a transmission of the information to the targets’ directors and a behaviour inducement.Footnote ⁶⁷ Indexed Funds could also act ‘behind the scene’ and try to influence corporate directors in the course of private meetings – instead of risking to leave evidence of their intentions in their voting preferences.Footnote ⁶⁸

Hemphill and Kahan have claimed that most of the abovementioned strategies are unlikely to occur – basing their analysis on the relationships between Common Owners (CCOs) and Non-Common Owners (NCOs). According to the authors, the only effect of common ownership on competition could consist of some cases of selective omission vis-à-vis Indexed Funds’ stewardship engagement, because ‘[c]ompared to NCOs, CCOs would tend to push less for aggressive competition where more aggressive competition would increase firm value (because of its effect on the value of competitors in which the CCO has a stake)’.Footnote ⁶⁹

But let’s hypothesise that Hemphill and Kahan got it wrong and Indexed Funds’ managers want to engage also in active behaviours in European companies: let’s see how this could be tackled by EU competition law. Let’s first consider both the active and passive hypothesis within the EU corporate environment characterised by the same features of the US one – i.e. absence of a controlling shareholder – a type of corporate ownership structure which, as already said, is rather uncommon in continental Europe.Footnote ⁷⁰ A legal framing of such behaviour under EU competition law may not be without difficulties. Unlike presumed by US literature,Footnote ⁷¹ it has been demonstrated that at least two potentially relevant EU competition law rules may not apply to such cases.

Firstly, rules on collective abuse of a dominant position will not apply because of the lack of the requirements outlined in the AirtoursFootnote ⁷² and ImpalaFootnote ⁷³ cases.Footnote ⁷⁴ So far, there is no evidence of any coordination – and therefore of any shared understanding regarding such (non-existent) coordination. As a consequence, there is not even question of monitoring and sustainability of such coordination – nor of a threat of competitive constraints jeopardising such coordination.Footnote ⁷⁵

Secondly, rules on concerted practices may also prove to be of extremely difficult application, both at the level of the market for financial investments and at the level of product markets of the target companies.Footnote ⁷⁶ As a matter of fact, none of the requirements identified in the Anic caseFootnote ⁷⁷ seem to occur in the case of parallel holdings. Not a concertation between undertakings (no evidence of concertation among investment management corporations nor among the investees), nor a specific behaviour of the undertakings pursuant such (non-existent) concertation.

Not even the Philip MorrisFootnote ⁷⁸ doctrine – had it to be considered living and applicable –Footnote ⁷⁹ would probably be a good point of reference for such cases. In the Philip Morris line of cases the CJEU and later the EU Commission assessed the potential competitive harm deriving from an investment of a firm in one of its competitors.Footnote ⁸⁰ But in the EU law pre-dating the first EUMR, and in the context of merger reviews under TEU article 86 (now TFEU article 102) the concept of influence emerged as a way to assess concentrations.Footnote ⁸¹ Such a concept – had it been properly developed and expanded as a standalone concept – might have helped framing the intermediate, nuanced, stages which may emerge in practice as corporate ownership arrangements that lay in between an anticompetitive agreement and a proper merger, both in their active and passive versions.Footnote ⁸² Nonetheless, neither the EU Commission nor the CJEU ever seemed to give specific relevance to the concept of influence as an independent legal category to be applied beyond the boundaries of EU merger law pre-dating the first EUMR.Footnote ⁸³

The concept of influence could probably be interpreted as a phylogenetic trace of a former theoretical framework (the pre-EUMR merger one), which no longer plays any role in the present EU competition law system. Even if we considered Philip Morris’ concept of influence as a living and fully applicable one, its adaptation to common ownership cases would require excessive logical stretching and, worse, a possible misreading of the rationale followed by the CJEU. In common ownership cases there is no issue of minority equity investment in competitors involved, because Indexed Funds operate in industries that have little or nothing to do with their target companies: hence the underlying facts are different from the Philip Morris scenario, which assessed investment in competitors.

As a matter of fact, the industrial distance between Indexed Funds and their target companies renders most of the EU competition law targeting influence down the supply chain inapplicable – as such law postulates a degree of industrially functional relationship between upstream and downstream firms.Footnote ⁸⁴ Such functional relationship is found both in horizontal (competitors) and vertical (supplier/producer/distributor) cases. But in the case of common ownership – as the word suggests – we are in presence of mere ‘owners’ that are rarely operative in industries bordering with their targets.Footnote ⁸⁵

Perhaps, a possible way to catch hypothetically relevant Indexed Funds’ managers behaviours could be framing them within a hub and spoke scheme, where Indexed Funds, the hubs, retrieve and process information, which they distribute down the chain of their investees. But it goes without saying that there is no evidence of a similar behaviour so far – and the likeability that anticompetitive activism is in the interest of Indexed Funds is rather low.Footnote ⁸⁶ Apart for the limited economic incentives of such conducts, given the degree of public attention that Indexed Funds have received so far, it would be very unlikely that they would engage in such activity out in the open – which in turn may render the hub and spoke case law extremely difficult to apply.Footnote ⁸⁷

But what is even more difficult to catch under present EU law are passive investments by Indexed Funds. The point of reference would be ‘quasi-mergers’ and that set of extremely vague situations surrounding the acquisition of a degree of influence over other corporations. Such nuanced situations normally become relevant for EU law only under merger reviews by the DG-Comp – hence posing the existence of a notified or notifiable merger as a pre-condition for the scrutiny of such situations.Footnote ⁸⁸ In such a context the DG-Comp has already assessed the anticompetitive potential of common ownershipFootnote ⁸⁹ and it may do so in future on a casuistic base.Footnote ⁹⁰

Beyond the present state of the art, one may wonder whether new rules might be introduced within the EU competition law system in order to address the potential competitive harm deriving from corporate ownership structure modifications brought by common ownership in the rare cases of dispersed ownership found in the continental European context. One might think this will occur soon, because of the recent studies that the EU CommissionFootnote ⁹¹ and the EU ParliamentFootnote ⁹² have dedicated to the common ownership debate. Despite the accuracy of the cited studies, the likelihood that such new rules are introduced light-heartedly may be rather low. Or better, it is precisely the accuracy of such studies which may induce us to predict that such likeability is low.

In fact, since the first decade of the twenty-first century, the EU Commission has investigated a connected, although significantly different case – i.e. that of the potential anticompetitive effects of (non-controlling) minority shareholdings.Footnote ⁹³ Such an investigation had culminated in a White PaperFootnote ⁹⁴ which has not resulted in any modification to the EU competition law framework. The attempts to innovate the EU merger legislation in order to accommodate an assessment of minority shareholdings were dismissed on grounds of insufficiency of evidence with regard to the competitive harm deriving from minority equity holdings.Footnote ⁹⁵ Such a dismissal confirmed the cautious attitude of the EU Commission when it comes to introduce completely novel competition rules. A similar cautious approach accompanied the last EU Commission’s decision on parallel holdingsFootnote ⁹⁶ and the subsequent declarations by DG-Comp Commissioner Vestager.Footnote ⁹⁷

14.6 EU Policies on Corporate Governance, Competition Goals and Common Ownership: in Search for Consistency

Imagine that, as some of the authors of the seminal papers on common ownership have claimed, none of the criticisms brought to their work is solidly founded;Footnote ⁹⁸ and/or that, in future, more convincing evidence emerges as to the competitive harm brought by common ownership on a wider sets of product markets than those object of the abovementioned papers.Footnote ⁹⁹ One may still wonder whether the most advisable policy reaction would be to dismantle or limit common ownership, as several US academics suggested.Footnote ¹⁰⁰ A core guideline in policymaking should be that not necessarily a solution to a problem is found at the same level of that problem. This is because what may be perceived as an issue or as a negative externality may actually produce also positive externalities at the same level: therefore, in such a case policy action could be more fruitfully carried out at a different level, in order to preserve the positive externalities.

To exemplify this principle with reference to common ownership, it is undoubtable that – besides the efficiency of indexing as an investment technique per se –Footnote ¹⁰¹ evidence is emerging on the biggest funds’ managers capability to provide replies to the ‘macro legal risks’ (i.e. those corresponding especially to ESG) of our times in a far more efficient way than traditional owners.Footnote ¹⁰² Even Azar, one of the authors of the seminal common ownership papers, has co-authored a paper analysing the role of the ‘Big Three’ (BlackRock, Vanguard and State Street) in containing carbon emissions.Footnote ¹⁰³

ESG objectives, such as anything which has to do with the protection of the environment, the fight against climate change, and ultimately an urgently needed adjustment towards a more sustainable way of doing business, are very close to contemporary consumers’ preferences (as well as to emerging investors’ preferences)Footnote ¹⁰⁴ – and in recent times strongly endorsed by policymakers.Footnote ¹⁰⁵ One may argue that positive role played by IF’s managers may not concern most European corporations, because they may be unable to act as stewards for such objectives in corporate environments where concentrated ownership structure prevails.Footnote ¹⁰⁶ But it is equally true that for such European corporations the potential anticompetitive harm brought by IF funds is still unclear and unproven – and it is unlikely to be particularly relevant.Footnote ¹⁰⁷

Besides that, possible policy interventions on common ownership should always consider EU competition law general principles. A core concept around which EU competition law has evolved is consumer welfare, which in the present interpretation refers to price effects. Nonetheless, it has been suggested that in future we may become more and more uncomfortable identifying consumer welfare exclusively with low prices – as we tend to do today.Footnote ¹⁰⁸ This again may bring ESG positive externalities into the domain of consumer welfare, therefore rendering a dismantlement of common ownership controversial.

Regardless of such potential (and probable) interpretative evolution, consumer welfare does not even represent the only goal of EU competition policies. As the recent empirical paper by Iacovides and Stylianou has shown, in practice consumer welfare is one of the many objectives that are at the core of EU competition policies.Footnote ¹⁰⁹ At that core there seem to be more a kaleidoscope of rather contradictory objectives than one univocal standard. Some of the objectives at the core of EU law may be efficiency, welfare, fairness, entrepreneurial freedom from competitors, market structure, European integration, and the protection of the competitive process.Footnote ¹¹⁰

Hence, when it comes to introduce new EU competition rules capable to tackle potential issues arising from common ownership, it is inevitable that such rules need to be tested in the light of the different goals advanced by the EU in terms of competition policy. On one hand, given the multiplicity of objectives pursued by EU competition policies, it may be easy to find justifications to any potential way forward, i.e. emphasising one of such policy objectives over the other. But it may equally become easy to find reasons for not regulating common ownership – hence leaving any intervention open to an almost infinite set of potential adjustments – also in the light of the evolution of such standards for adapting to the pursuance of ESG objective.Footnote ¹¹¹

But given the ever-expanding reach of EU policies, an intervention on common ownership would not be exclusively a matter of internal consistency (i.e. with the goals pursued by competition law). A direct intervention against indexing would entail a redesigning of corporate ownership structure of many companies and such a change would reverberate on their corporate governance. Such shift would in turn raise the question of consistency with EU policies that lay beyond the competition law ones – for instance, corporate governance policies – stewardship, at least for those European companies that are characterised by dispersed ownership structure.Footnote ¹¹² One may wonder whether a dismantlement of the biggest Indexed Funds would be consistent with requests for enhanced stewardship, when institutional investors, such as Blackrock or Vanguard have been among the most actively engaged investors in pursuance of sustainability.Footnote ¹¹³ A direct intervention on Indexed Funds’ holdings may be similar to one of those unfortunate Mikado picks that make the whole stick-tower crumble. Hence, before dismantling parallel holdings, it would be better to know what financial reality will come next. And a more cautious way forward would entail examining the overall situation from a better viewpoint and putting all the different pieces of the puzzle in their right place – so to create the lesser frictions possible among different EU policies.

14.7 Zooming-out and Zooming-in: Looking for Alternative Ways Forward – Economic and Political Trajectories

If the issue of parallel holdings is so deeply intertwined with several coexisting layers of policy choices, one may wonder what could be the alternative ways forwards that may produce the most positive externalities while reducing negative externalities.

A first point concerns the effects of the academic and institutional debate on Indexed Funds’ managers. If Indexed Funds’ managers had ever thought about actively engaging in restricting competition, the present trend of literature on common ownership may have persuaded them to refrain from doing so – by recalling the public attention on their activity and therefore increasing their chances they get caught.Footnote ¹¹⁴ Hence, today the main threat – if there is any – may derive from common owners’ passive investment.Footnote ¹¹⁵ A possible way to analyse this case from a broader perspective is to consider one by one the different variables surrounding common ownership, as identified in the papers that have fuelled the debate on this topic. Occasionally, one can apply a counterfactual analysis to the potential modifications brought to each variable – hence comparing different policy alternatives.Footnote ¹¹⁶

To exemplify, policies directed to avoid or limit the potential anticompetitive harm deriving from common ownership may attempt to operate on: the structure of the target companies’ product markets;Footnote ¹¹⁷ the structure of the financial services product market;Footnote ¹¹⁸ the investment techniques adopted by companies active in the financial instruments’ product market;Footnote ¹¹⁹ and the governance activity carried out by the financial firms within their target companies.Footnote ¹²⁰

Some of the policy ways forward proposed by US academics aimed at banningFootnote ¹²¹ or limiting indexing.Footnote ¹²² If we apply a counterfactual reasoning to the proposals aimed at banning or limiting indexing, the chances that removing common ownership may improve competition are at least questionable. Big investment management companies own extremely large amounts of stock and that such financial instruments would need to find new owners. One may wonder whom would purchase that stock. First, although the study of strategies for State direct investments in the economy are not unknown within the EU Member States’ political arenas, especially in the midst of the COVID-19 crisis,Footnote ¹²³ it is unlikely that this becomes a generalised trend.Footnote ¹²⁴ Hence, in line with the liberal economic model that the EU pursues, one may imagine that the equity holdings dismissed by Indexed Funds would be purchased by other private economic actor on the stock market, accompanied by some mechanisms due to temper the effects of a temporary oversupply of stock. Among potential purchasers, one might think of entrepreneurial families – the typical continental European blockholder – or for instance of other (non-indexed) investment funds. As to entrepreneurial families, this would contradict the present trend of families looking for external equity investors, more than co-investing in third-parties’ companies.Footnote ¹²⁵

Other (non-indexed) investment funds would probably be the best candidates to substitute Indexed Funds in the ownership of such equity holdings. But literature has shown that the majority of institutional investors not employing indexing have not been champions in activism so farFootnote ¹²⁶ – with some notable exceptions, such as hedge funds.Footnote ¹²⁷ But hedge funds have traditionally been backed by other institutional investors for their active strategies.Footnote ¹²⁸ Moreover, hedge funds are renowned for not been particularly well aligned with ESG concerns.Footnote ¹²⁹ And this may again reverberate against the quest for internal EU policy consistency.

An alternative way to obtain a better competitive outcome might be to impose corporate governance rules that prompt investors to actively intervene in governance. One way could be promoting regular meetings for discussing pro-competitive issues, such as for example how to make the company more price efficient, how to make its products more attractive and ultimately how to reap one’s competitors’ market shares. An even simpler way could be to impose a duty to actively promote competition. But it is unlikely that an omission could be sanctioned in a reasonable way: as a matter of fact, this would entail intruding in the discretionary choices of investors and especially directors which in modern corporate law have to abide by the business judgement rule.Footnote ¹³⁰ Moreover, such an intrusion would introduce a level of public ordering that would go well beyond its level even in the most State intervention-prone jurisdictions.Footnote ¹³¹ A possible further intervention in governance could consist of limiting the potential contacts among different companies by way of dismantling interlocking directorships – which still seem to be very common both in Europe and in the US – and which may be the cause of anticompetitive harm beyond what might be causes by common ownership.Footnote ¹³²

Another way forward may consist of intervening on the structure of the target companies’ product markets.Footnote ¹³³ It goes without saying that such an intervention – at least in the US – would need to be backed by stronger evidence than presently existing.Footnote ¹³⁴ But in Europe, where most product markets are less concentrated, a generalised policy against concentration may at times produce inefficiencies and it may also contrast with the political objective of the EU. For example, further market integration by facilitating cross-border transaction or even the creation of European national champions – an objective that is already pursued less vigorously than needed by present EU policies.Footnote ¹³⁵ Moreover, in certain cases, such interventions would not even be needed, as some industries in Europe are national by nature or as a consequence of the legislation.Footnote ¹³⁶

Possibly, the most consistent way forward would entail intervening on the structure of financial services’ product markets – i.e. trying to promote new entrants by bringing down barriers to entry and/or intervening with structural remedies, i.e. forcing the largest incumbent investment management corporations to split in a higher number of fund management companies. This idea has been proposed by Goshen and Levit, in relation to a parallel problem which seems connected to the emersion of the largest Indexed Funds, i.e. a monopsony in the labour markets. This has progressively caused a transfer of wealth from employees to equity holders,Footnote ¹³⁷ and has also created labour instability, which seems at odds with the corporate governance-based sustainability objectives highlighted above.Footnote ¹³⁸ Nonetheless, there is question as to whether such a modification in the ownership structure of big investment management corporations would grant the pursuance of the sustainability objectives that have been promoted so efficiently also thanks to the present financial services market concentration. And finally, this would be more a matter for US than EU competition authorities, given the limited reach of common ownership in Europe.

Concluding, there seem to be no perfect fix to the potential competitive harm deriving from parallel holdings by Indexed Funds, while, in a world characterised by urgent environmental and social problems, there seem to be a need for consistency among policies in bordering areas.

14.8 EU Policies for Common Ownership within the Global Context

As I started this brief excursus by considering the EU policies on common ownership with a glance to the global perspective, it is worth concluding by considering such a wider viewpoint. And so far, at least one crucial player has not been included yet in this analysis: the PRC. To my knowledge, the MOFCOM has not released any statement concerning common ownership yet. Nonetheless, the present silence of Chinese authorities on this subject matter does not mean that the US and EU policy choices in relation to common ownership are not without consequences for the Chinese economic actors.

As a matter of fact, in the Chinese context, common ownership was already a hot topic far before than the whole Indexed Funds ‘scandal’ broke out. One may recall the fact that the EU – probably misunderstanding the way the PRC and its Communist Party manage their investment in SOEs – has qualified Chinese holdings in EU companies as common ownership.Footnote ¹³⁹ Advancement on the analysis of the concept of common ownership in the context of Indexed Funds may in turn inform also the policies on SOEs’ parallel holdings and vice versa. But beyond a potentially interesting comparison between Indexed Funds and SOEs common ownership law rules, a far more important question concerns the global financial implications triggered by a potential request addressed to Indexed Funds to dismiss part of their holdings –Footnote ¹⁴⁰ or to allow for new entrants in the market for financial services.Footnote ¹⁴¹ In fact, earlier we identified other investment funds not employing indexing as potential acquirers of such holdings, or alternatively the entry of new investment management corporations in the financial market. But we have not discussed potential acquisitions of dismissed holdings by PRC SOEs, which are renowned for their interest in cross-border acquisitions.Footnote ¹⁴²

The inability to replace entirely the present common owners with alternative European or US owners might provide far-East investors with unprecedented financial investments opportunities. And especially Chinese FDI may focus more intensively on Europe, given the limitations brought by the FIRRMA legislation approved under the Trump administration to foreign investment in US companies.Footnote ¹⁴³ But if the EU Commission is still persuaded that most PRC holdings tend to be under the control of the Chinese Communist Party, this may well sound to the DG-Comp like throwing such holdings from the frying pan into the fire.Footnote ¹⁴⁴ Faced with a potential increased demand from Chinese investors, EU policy-makers might be prompted to build even higher walls against PRC purchases in EU Member States’ companies. This is turn may generate a spiral of tit-for-tat reactions – as the one we have already seen between PRC and US – which may be hard to contain and certainly not beneficial for the process of economic integration.Footnote ¹⁴⁵

As a matter of fact, the problem of common ownership – although seemingly only technical in nature – may have geopolitical implications hard to imagine for those who are mostly concerned with its potential price effects. And this might be another reason why the EU competition authorities are walking on eggshells – on one hand trying to gather as much data as possible on this subject matter and on the other trying to avoid untimely interventions.Footnote ¹⁴⁶

Ultimately, what needs to be understood before a final word is spent on common ownership is how the sustainability/growth conundrum will be solved at a corporate governance level and how each of the largest world economic blocks will carve their role in the pursuance of even far wider objectives – among which we must certainly include the protection of Human Rights.Footnote ¹⁴⁷ If trust among global players and convergence in their progressive policies increases,Footnote ¹⁴⁸ we may see a far larger set of cross-equity holdings of various nature across the globe and in every possible and imaginable direction. But suspicion and localism may put the word end to such cross-border interactions.Footnote ¹⁴⁹ The hope is that the democratisation brought by investment funds in the financial arena will evolve into something even more positive – helping spreading progressive value beyond what the world is seeing today.

15.1 Introduction

Common ownership is the talk of the town in antitrust land. Surrounded by mystery and noise, the competitive implications of rival firms being partially owned and controlled by a small set of overlapping owners are both fascinating and hotly contested. The fascination comes from the fact that the source of potential competition harm may be minority shareholder control in a setting of widely held companies.Footnote ¹ In fact, the common ownership phenomenon is so pervasive, in particular in the US,Footnote ² that if this new theory of harm is true, most markets could be beset by serious antitrust concerns. At the same time, scepticism among academics and policy-makers abounds. Most notably, critics wonder about the likely prospect, quantum and mechanics of common owners’ influence driving any pro- or anticompetitive effects.Footnote ³ It is often stressed that the antitrust analysis of common ownership is clearly distinguishable from that of cross-shareholding links between competitors.Footnote ⁴ Indeed, the novel concern caused by common shareholdings derives from indirect, and possibly partial, shareholder overlaps across rival firms rather than directly from competitive overlaps in product markets. A comprehensive account of partial ownership, capturing the competition dynamics of both cross- and common shareholding and the incentives of both individual and institutional investors, is notoriously missing.Footnote ⁵ Yet, so far, the spirited debate between antitrust and corporate law and economics scholars centres on whether this ‘knowledge gap’ is material, set to be filled by better understanding and experience as a matter of course or whether it is a fictional problem and an empty inquiry that is theoretically implausible and empirically unrealistic to unfold.Footnote ⁶

Against this backdrop, this chapter aims to illuminate some of the latent connecting points in this debate, by looking back into the past and then fast forward to the future. There are two key takeaways from this analysis. The historical split of corporate and antitrust laws and their gradual specialisation in targeting different issues with distinct objectives in mind has unwittingly created regulatory gaps. This is the source of the present-day problem posed by minority and common shareholdings for competition law. With this understanding clear, the analysis moves on to offer a new taxonomy of (partial) shareholdings in light of their partial control characteristics, with distinguishable classifications based on competition and corporate law, as compared to broader economics-focused notions of control. The industrial organisation perspective reveals that commonly thought passive and highly diversified minority holdings are not necessarily innocuous in terms of their competitive implications. Rather, minority common shareholdings may give rise to actual or potential ‘competitive influence’ under certain circumstances (‘influential’ shareholdings).

The corollary is that antitrust cannot afford to neglect such corporate ownership structures and for this reason, it is called to look into the actual corporate governance dynamics in the substantive assessment of cases and in designing appropriate remedies. Yet, the historical and economic analyses further suggest that merger control needs to recalibrate its jurisdictional scope and embrace a ‘structural’ approach, combined with ‘case-by-case’, fact-specific analysis, also for shareholdings falling below legal thresholds of control. This reorientation would not only fill enforcement gaps and capture new theories of harm relating to common shareholding but also reconnect merger control to its corporate law origins in a way that holistically addresses agency costs and market power concerns linked to such shareholding. At the same time, corporate law and governance should be cognizant of these parallel developments and tread softly when shaping their own regulations so that they do not augment any antitrust concerns.

The structure of the chapter is as follows. Section 15.2 provides relevant background on the two-sided history of regulating shareholding acquisitions under corporate and competition laws. Section 15.3 illustrates antitrust’s embeddedness in pre-existing corporate laws and forms, documenting the early unity and progressive quiet disconnect of the two fields in regulating ownership structures and intercorporate links. Section 15.4 presents the contemporary common ownership (hypo)thesis and the distinct challenges and opportunities that it poses for both antitrust and corporate law. Section 15.5 develops a working taxonomy of (minority) shareholding types and their (partial) control characteristics from different perspectives with a particular focus on competition economics. Section 15.6 focuses on the economic attributes and competitive effects of common shareholding seen and analysed through the lens of corporate property rights theory. Section 15.7 concludes with an urge to competition and corporate governance and finance policy-makers for harmonic progression in seeking regulatory solutions to address common ownership and with further implications for competition law following the preceding analysis.

15.2 Mergers and Minority Shareholding: A Two-Sided History

Minority shareholding is an old story in the realm of competition or antitrust laws. It takes us back to the origins of antitrust, or even to preceding developments in corporate law that led to its birth.Footnote ⁷

Ever since its inception, antitrust was designed to tackle a rampant wave of mergers and acquisitions fuelled by technological and industry developments as well as holding structures (‘trusts’)Footnote ⁸ between competing companies that essentially led to concentrated economic power and monopolistic market outcomes. The antitrust movement was an immediate reaction to evolving corporate laws. The emergence of the trusts and multistate corporate mergers was the result of a ‘race to the bottom’Footnote ⁹ with US states ‘competing’ for corporate charters and New Jersey being the first to enable ‘interstate’ holding structures intended to monopolise or cartelise national industries.Footnote ¹⁰ US federal antitrust law was born in 1890 in an attempt to rein in this ‘accommodating’ interstate competition in corporate laws among different states. Up to that point, state law treated jointly ‘issues of antitrust and corporate authority’.Footnote ¹¹ Yet, not all anticompetitive mergers or restraints of trade were captured by those initial antitrust laws. In fact, the Sherman Act originally targeted ‘loose’ combinations (cartels) rather than ‘tight’ ones (mergers)Footnote ¹² but later case law (1904) also applied it to the holding company (‘single business firm’).Footnote ¹³ Stock acquisitions only became a specific antitrust target of US merger control with the coming into force of the Clayton Act in 1914.Footnote ¹⁴

Similarly, the founders of the EU avoided incorporating rules controlling corporate ownership structures in the Treaty of Rome, which included solely behavioural rules on cartels and abuse of dominance. Only in 1990, EU Members States agreed to have a pan-European Merger Regulation (‘EUMR’) in place to address cross-border mergers and acquisitions. Till then, the available EU antitrust rules were used as a de facto merger control regime.Footnote ¹⁵ Indeed, over time, EU authorities decided to make use of Article 102 TFEU to address mergers and ‘majority’ acquisitions.Footnote ¹⁶ While later on, EU case law further applied Article 101 (and 102) TFEU to minority shareholdings linking competitors and giving rise to ‘some influence’.Footnote ¹⁷ In fact, part of the reason for the adoption of the EUMR was the foregoing ‘expansive’ use of Articles 101 and 102 TFEU by the European Commission to go after ‘minority’ share acquisitions in competitors.Footnote ¹⁸ Under pressure, Member States decided to compromise by yielding part of their regulatory powers checking anticompetitive mergers, acquisitions, and joint ventures under national law rather than be de facto completely swept away by creeping EU antitrust competence. The end result of this pragmatic political settlement was that the EUMR was designed to jurisdictionally cover only cases of ‘concentrations’ that give rise to a ‘lasting change of control’, i.e. multistate corporate combinations that confer upon the acquirer positive or negative ‘decisive influence’.Footnote ¹⁹

Perhaps counterintuitively and for different reasons, US and EU merger control laws shared a little noticed, common trajectory in their origin. Their absence was conspicuous in the inaugural design of cross-Atlantic antitrust rules. EU antitrust law had a ‘top-down’ inception inspired by a high-level political commitment towards an internal market integration objective. Coming only later and independently, EU merger control was the political product of Member States agreeing to a ‘lesser evil’ against pressing supranational antitrust expansion.Footnote ²⁰ Initially, however, cross-border mergers were seen as a positive force furthering European integration and the competitiveness of its industry rather than in need of any legal constraint.Footnote ²¹ In contrast, US antitrust law had clear ‘bottom-up’ origins reflecting populistic sentiments of the time against cartelising business trusts and monopolistic merger combinations. US merger control also came into force later but predominantly to fill the gaps left by state corporate laws. Accordingly, the initial omission of merger law both in the EU and the US from the traditional body of antitrust law was not a random policy choice. US antitrust legislators consciously chose to break free from the ‘formalities’ of state corporate law and its ‘structural model’ of dealing with the ‘trust problem’ while opting for a ‘strategic model’ merely governing agreements and combinations in restraint of trade ‘based on economic theory that purported to distinguish between competitive [and] anticompetitive’ ones.Footnote ²² The distinct concern of the antitrust approach when examining a merger or combination was ‘its object or effect’ on market competition, not ‘its form’.Footnote ²³ In addition, it was initially believed that issues of firm formation, ownership transfer agreements, and the sale and purchase of property or stock acquisitions, either by a corporation as a legal business entity or by its business owners as physical persons, were beyond the regulatory ambit of antitrust rules.Footnote ²⁴

Seen in this light, the noted ‘foundational deficit’Footnote ²⁵ of EU competition law in reaching to partial ownership structures and share acquisitions has not been a singular EU story. Traces may also be found in US legal history, albeit in subtler forms. Yet, what is noteworthy about US antitrust law is its ability for fast(er) adaptation to emerging business and economic realities, such as the disaggregation of state corporate and antitrust policy and the proliferation of multistate business firms, as well as to the rapidly changing content and scope of state corporate laws towards more liberal and enabling rules favourable to private ordering. Indeed, jurisdiction and substantive review under US merger control now resolutely rely on ‘effects-based’ tests, whereas EU merger control jurisdiction retains its ‘formalistic’ reliance on a narrow legal conception of ‘control’.Footnote ²⁶

15.3 Antitrust Embeddedness in Corporate Forms: The Quiet Disconnect

But the interplay between competition and corporate laws goes one level deeper. Antitrust choices regarding rules or analytical frames had been implicitly premised on pre-existing corporate law and practice in the formative era. Early corporate law in the US till the end of the nineteenth century was much more restrictive and unitary in nature in regulating business entities, their structure and operation, as a legal and social phenomenon. Corporate law not only included far more outright prohibitions (mandatory rules) rather than balancing or enabling rules regarding ownership structure and governance practices within any individual firm but also strictly regulated intercorporate relations. Indeed, in the early days of US corporate law agency problems within the firm were not a major concern as the law and surrounding circumstances at the time ensured there were:Footnote ²⁷

1. i) no ‘separation of ownership and control’,Footnote ²⁸

2. ii) no ‘separation of ownership and consumption’,Footnote ²⁹

3. iii) no ‘separation of control (voting rights) and investment (financial interests)’.Footnote ³⁰ Or even further, from the perspective of shareholders:

4. iv) no ‘separation of ownership from ownership’,Footnote ³¹ and

5. v) no ‘separation of ownership and awareness’.Footnote ³²

The first legally organised companies had been novel combinations of ‘private investment and state-granted monopoly privileges’ to undertake important community projects under special charters.Footnote ³³ Once chartered companies obtained ‘perpetual existence’ and ‘strong entity shielding’, shareholders acquired a legal ‘right to sell their shares without the consent of other owners’ in exchange for their lost ability to withdraw from the joint venture at will.Footnote ³⁴ Such monopoly grants served two purposes, one balancing against the other. Monopoly rents were the bait to attract self-interested investors against the risk of ‘control-person opportunism’.Footnote ³⁵ At the same time, the public interest was also served by enabling large-scale ventures that would not have been possible otherwise. With the corporate form becoming widely available under general incorporation statutes,Footnote ³⁶ exclusive privileges were no longer a priori guaranteed. Democratisation of the corporate form paved the ground for free market competition. Success in the marketplace was now the driver of private profit-seeking venturers and also the only assurance for firm survival. Competition among independently operating companies was the new modus operandi for serving the public and consumer welfare.

Thus, although anachronistic today, it is no surprise that in its early days, US corporate law was primarily seen as a response to problems of monopoly and market power rather than concerned over agency problems inside the firm.Footnote ³⁷ Prominent examples are rules regarding voting (caps) or purpose restrictions (ultra vires doctrine),Footnote ³⁸ prohibition of separately allocating cash flow and control rights or splitting shareholders’ property interests in the firm (bar on the separation of ownership/ investment and control),Footnote ³⁹ prohibition of acquisitions of foreign (out-of-state) companies,Footnote ⁴⁰ prohibition of ‘intercorporate stock ownership’ (interlocking shareholding or ‘common stockholders’)Footnote ⁴¹ and ‘intercorporate influence’ (interlocking directorates or ‘shared directors’).Footnote ⁴² During this era, it was not obvious that separate corporate entities could acquire or own property or equity interests (shareholding) in others or engage in combinations (mergers) of assets (productive facilities) or stock (capital).Footnote ⁴³ The legal and economic environment of the time was in other words quite different to what is observed today.

Progressively, however, most of these restrictions were abandoned as corporate law relaxed and redirected its focus on the ‘internal affairs’ of firms,Footnote ⁴⁴ aiming to minimise agency costs and conflicts of interest. Companies were gloriously emancipated from early state (corporate) control; at the age of adolescence, the only credible constraint on their market and transactional activities was (federal) antitrust law.Footnote ⁴⁵ Given its liberalisation trends, corporate law was now oriented on developing alternative means of protecting investors, mostly notably minority shareholders,Footnote ⁴⁶ rather than providing any form of consumer protection against them. While antitrust grew to fill in those gaps, it only comprehensively did so with regard to mergers. Minority share transactions have been loosely regulated especially in the EU whose merger control rules imported corporate law norms, conceptions, and formalities to single out controlling acquisitions from presumably harmless ‘non-controlling’ ones.Footnote ⁴⁷ Thus, the staggering specialisation of corporate and competition laws on firms and markets, respectively, had its own unintended consequences as the regulation of minority shareholding came to ‘fall between the cracks’.

15.4 The Common Ownership (Hypo)thesis: Corporate Sensibility or Antitrust Overkill?

Nowadays, concern over potentially anticompetitive minority shareholding has taken novel forms. The buzzword is ‘common ownership’Footnote ⁴⁸ or ‘horizontal shareholding’.Footnote ⁴⁹ The dramatic growth of large institutional investorsFootnote ⁵⁰ and the indirect concentration of (partial) ownership of publicly listed firms it brought with it, not only signalled the promise of improved corporate governanceFootnote ⁵¹ but also created a major ‘challenge to market competition’Footnote ⁵² or indeed the ‘greatest anticompetitive threat of our times’.Footnote ⁵³ More fundamentally, however, common institutional shareholding has both deep and mixed implications for corporate as well as for competition laws.Footnote ⁵⁴ Arguably, parallel horizontal shareholdings by institutional investors may be perceived as the ‘new trusts’: a modern version of horizontal shareholder structures interconnecting competing firms.Footnote ⁵⁵ The interest and curiosity in increasing common ownership by institutional investors arouse not only due to its a priori ambiguous welfare effectsFootnote ⁵⁶ but also even more because it challenges the fundamentals of antitrust (and organisational) conventional wisdom.

Traditionally, minority cross-shareholdings have been a natural object of competition law concern and attention considering the direct competitive overlaps between firms operating in a (horizontal or vertical) competitive relationship.Footnote ⁵⁷ Now, a new ‘economic blockbuster’Footnote ⁵⁸ has become the epicentre of ground-breaking competition law and economics scholarship: the same group of large, concentrated, and diversified financial intermediaries partially own and control significant parallel shareholdings in the major competing firms within a given industry across the economy. On the one hand, the stakes held by each institutional investor in individual firms are small in absolute terms, thus considered ‘non-controlling’ on a stand-alone basis from a governance perspective, and often ‘passive’ given the indexation and portfolio diversification investment strategies employed by institutional investors from a finance perspective. However, empirical and theoretical economic research reveals that they may (and do) nonetheless affect competition outcomes in product markets.Footnote ⁵⁹ Modern finance theory and the evolution of capital markets have transformed the investment landscape towards increasing diversification and institutional investment, with indirect (and unintended) consequences for corporate ownership, governance, and industry structure.Footnote ⁶⁰ The intriguing possibility raised by the ‘common ownership hypothesis’ is that the combination of institutional re-concentration of ownership and portfolio diversification has systemic corporate governance and market competition effects.Footnote ⁶¹ In the case of minority common shareholdings, the (partial) shareholder overlaps in the ownership structure of the major competitors in concentrated industries that are said to (indirectly) increase the ‘effective’ market concentration and also produce competition harm and possibly productive efficiencies.Footnote ⁶² In effect, the common (financial) owners of rival (industrial) firms may have the incentives and ability to affect the operation of firms and markets away from individual profit maximisation leading to increased prices and reduced industry output.Footnote ⁶³

The fundamental antitrust question is: is this ‘new wine’ that needs to be distilled and fit into ‘old (legal) bottles’ or would such a fit simply be unnatural – an ‘antitrust overkill’? The myriad of new concerns and possibilities common ownership raises in a variety of legal and economic fields may easily let the debate go astray. Yet, the intimate relation between competition and corporate governance and finance lay at its heart, both in terms of theory and practice. What policy-makers decide on either side shall have profound implications on the way firms are organised and governed as well as on how financial and product markets operate.Footnote ⁶⁴ Therefore, the aforementioned functional regulatory schism offers no excuse for overlooking the systemic consequences of the issue in point.Footnote ⁶⁵

15.5 Shareholding Types and Antitrust: The Controlling, the Passive, and the Influential

Let us then take a step back and refocus the analysis in order to better appreciate where we stand. From a competition perspective, the interesting cases of minority shareholding have been those involving ‘non-controlling’ or ‘passive’ financial stakes acquired in rival firms, which may escape antitrust scrutiny.Footnote ⁶⁶ Despite the gradual updating of antitrust rules to capture new forms of potentially anticompetitive practices, it remains a matter of debate whether their scope or interpretation extends to ‘partial’ ownership of a competitorFootnote ⁶⁷ when participation in the share capital is limited to a minority position (nominal equity holding) and not accompanied by majority voting control (corporate legal control),Footnote ⁶⁸ or any other form of active influence (by means of governance actions or activist intent) over the commercial activity of the competing company.Footnote ⁶⁹ Complex theoretical and factual issues at the intersection of competition and corporate laws naturally arise in the analysis of those cases. Changes in the ownership structure (shareholder base) of firms may impact corporate governance (managerial and firm behaviour) which in turn affects competition (market concentration and industry performance).

Usually, control is seen as a key determinant of an antitrust theory harm and also part of the mechanism that translates (partial) common ownership into suboptimal corporate and market outcomes.Footnote ⁷⁰ Yet, ‘control’ is a complex and multifaceted concept, and ‘partial’ control arising from minority shareholding is not clear or well established in legal or economic theory.Footnote ⁷¹ Indeed, it is often a form of ‘factual’ control situation that may heavily depend on the surrounding context and specifics of the particular case.Footnote ⁷² At this point, it becomes both interesting and instructive that minority shareholding alludes to the ‘many faces’ of ownershipFootnote ⁷³ and ‘shades’ of control, with each combination leading to different kinds and degrees of competition effects. The variety in effect contrasts sharply with our limited word stock that is often misleading or inaccurate given the overlapping use of common terms such as (ownership or) control for different purposes and bodies of law. In order to dissolve some of the unnecessary confusion and elucidate the competitive harm potential of distinct shareholding types, I elaborate on the different layers of control attending a given minority position. Accordingly, minority shareholding can be classified as follows:

1. i) controlling or non-controlling – from the perspective of (EU) competition law – depending on whether the acquirer is able to exercise formal (legal) control over the target or not;Footnote ⁷⁴

2. ii) solely or jointly (partially) controlling – from the perspective of (EU) competition law – depending on whether there is a single dominant shareholder with clear (de jure) sole control over the target or control is (de facto) shared among many individual minority shareholders, in ex ante unascertainable ways (e.g. if joint control exists on the basis of ‘changing coalitions’ and no ‘stable’ majority can be established even in the presence of equal equity positions and identical rights among the shareholdersFootnote ⁷⁵);

3. iii) active or passive – from the perspective of corporate law – depending on the acquirer’s ability to exercise some active (economic) influence over the target or not, usually given its shareholder rights or corporate governance actions;Footnote ⁷⁶

4. iv) totally or partially controllingFootnote ⁷⁷ (actively influential)Footnote ⁷⁸ – from the perspective of competition economics and corporate governance – depending on whether there is a dominant shareholder with clear total (legal) control, due to either a majority or a minority equity holding, or a formally non-controlling minority shareholder with some (economic) influence over the target, due to a de facto ‘blocking minority’ (veto power) or some other situation of ‘informal influence’ arising out of statutory corporate law or contractual rights (e.g. voting rights, information rights, disproportionate board representation, board observer seats);Footnote ⁷⁹ and

5. v) actively or passively (strategically)Footnote ⁸⁰ influential – from the perspective of competition economics and corporate governance – depending on whether an ‘active’ shareholding directly affects the behaviour of the acquired firm given the acquirer’s ability to exercise active influence over the target, by operating within its corporate governance, or a formally ‘non-controlling’ or ‘passive’ shareholding affects the acquirer’s own incentives to compete due to the strategic interaction between rival firms in oligopoly even if the competitors are linked by purely financial interests without any apparent influence or control in the target’s governance.Footnote ⁸¹

The above exposition reveals that the legal and economic views of the different shareholding types are not fully overlapping. That is, some shareholdings that are: a) only ‘partially’ or not standalone controlling or b) completely ‘non-controlling’ and ‘passive’ as a matter of competition or corporate law, and possibly outside the reach of antitrust or merger laws, may still turn out to be ‘competitively influential’ as a matter of industrial organisational theory. The economics perspective also vividly illustrates that there is a continuum of effects on competition due to the multiple shades of control and non-control that accompany minority shareholdings. Effectively, this economics-informed analysis adds another shareholding type in the traditional legal dichotomy – controlling, influential, passive – that may be of competition concern. That said, this continuity in effects does not necessarily suggest that there is a linear progression in terms of the magnitude of potential harm: occasionally a totally controlling shareholding (active sole control) may be quantitatively more detrimental to competition than a full merger,Footnote ⁸² or similarly, partially controlling or mutually influential shareholdings (de facto joint control) may be equally harmful to a full merger.Footnote ⁸³

A visual representation of the taxonomy of shareholding types based on their control qualities from the three distinct analytical perspectives employed above is shown in the following table. The cells highlighted in grey illustrate the potentially problematic (competitively influential) minority shareholdings that may escape competition scrutiny in certain jurisdictions such as the EU or others that follow its example and the underlying reasons that trigger this situation, i.e. legal gaps due to formalistic definitions of shareholdings in competition or corporate law. In light of the above, it is also important to realise that both active and passive minority shareholdings may give rise to ‘competitive influence’, which may flow from either governance influence or strategic influence, respectively.Footnote ⁸⁴ The table thus visually reflects how precisely and what specific types of minority shareholding came to ‘fall between the cracks’.

Seen from the broader economic point of view, competitively ‘influential’ minority shareholding may be further subcategorised, considering the time horizon, intensity of economic control, and degree or reciprocity of profit internalisation, as follows:

1. i) statically or dynamically influential – depending on whether the acquirer is able to exercise current influence over the target, and thus have an observable impact on competition (present effects) or possibly exercise future influence (potential effects), in light of its theoretical shareholder rights that could put into use (e.g. voting or other special contractual rights such as rights of first refusal);Footnote ⁸⁵

2. ii) proportionately or disproportionately influential – depending on whether the acquirer’s degree of internalisation of rival profits (the ‘profit weight’ as a function of financial interest and controlFootnote ⁸⁶) is proportionate to its partial shareholding investment in that rival (the ‘control weight’ is equal to the nominal percentage of the equity positionFootnote ⁸⁷) or disproportionate (the actual degree of control vis-à-vis the target’s management discretion (‘agency costs’) is more or less than the nominal equity position);Footnote ⁸⁸

3. iii) one-directionally or bi-directionally influential – depending on whether only the acquirer is induced to internalise its partially acquired rival’s profits, due to its financial interest linked to its shareholding investment, or also the target is induced to take into account the acquirer’s shareholding,Footnote ⁸⁹ due to the latter’s corporate influence exercised in the target firm’s governance,Footnote ⁹⁰ or due to target’s own parallel shareholding and financial interest in the acquirer.Footnote ⁹¹ Importantly, a ‘bi-directionally influential’ minority shareholding (mutual internalisation) need not be harmful to competition (softening of competition or collusion) due to the internalisation of competitive externalities but may also be welfare enhancing (pro-competitive or efficiency creating) due to internalisation of productive or innovation spillovers.Footnote ⁹²

15.6 Common Shareholding in Antitrust and Corporate Governance: From Competition Effects to Property Rights

It follows from the above analysis that minority shareholding that is considered ‘non-controlling’ or ‘passive’ in the legal sense may well be de facto competitively ‘influential’ in the economic sense given its present or potential impact on the corporate control and competition dynamics. Further, from a corporate perspective, so long as the shareholding is ‘voting’ stock it is not really a passive one but rather dynamically influential in that the power of the vote may always be exercised later in the future,Footnote ⁹³ or in fact, the implicit threat stemming from its mere existence may produce current results and change equilibrium outcomes given its deterrent disciplining effect.Footnote ⁹⁴ In addition, so long as stock is voting and ‘freely tradable’ (either as a package of share plus vote or simply the vote), there is no real ‘separation of ownership and control’Footnote ⁹⁵ in the sense that corporate control is contestable.Footnote ⁹⁶ Management is disciplined by both the stock market and the market for corporate control, while shareholders remain ‘residual claimants’Footnote ⁹⁷ with the power to hold/exercise and buy/sell their votes. The latter bear both the residual risk and the corresponding residual rights to control of the corporate property (due to their equity share in the corporation’s profits). Thus, the ‘atom of property’,Footnote ⁹⁸ albeit diffused among many shareholders and partially split between principals-passive owners and agents-actual managers in large public corporations,Footnote ⁹⁹ is remarkably solid: out of all corporate constituents, only shareholders as a class generally bear the marginal gains and losses of corporate actions or omissions and thus have the right incentives at the margin to check and redirect management towards improved corporate performance by the (actual) use or (deterrent effect of) possession of voting rights.Footnote ¹⁰⁰ Consequently, shareholders’ private profit motive remains the (valid) driver for financial investment in corporations and management discipline as well as for more efficient use of an industrial property. Private property, free markets, and competition reassuringly remain the ‘holy triad’ upon which modern corporate and industrial organisation solidly relies.Footnote ¹⁰¹

The above analysis also makes clear that investor ‘passivity’ does not necessarily translate into permanent corporate ‘silence’ or would justify an a priori antitrust immunity.Footnote ¹⁰² If the minority shareholding is accompanied by voting rights and the shares or votes can be exchanged, potential control persists. Yet it is difficult to tell ex ante if and how such shareholder power may be exercised.Footnote ¹⁰³ Considering this ex ante uncertainty and the importance of the surrounding factual circumstances (shareholder control dynamics, market and legal constraints) to assess the competitive significance and effects of any minority shareholding, it is unlikely that a purely ex ante merger control regime will be effective in distinguishing and addressing potentially harmful cases of minority cross- or common shareholding. This is so even if such prophylactic regimes were not structural (corporate law model) or formalistic (as in EU merger control) but effects-based (as in US merger control).

Nonetheless, the US merger regime has a resolutely sound economic structure: its open-ended scope for liability (no safe harbour for any actual ‘lessening of competition’) combined with its ‘passive investment’ exemption from filing a notification provide both flexibility and reduced regulatory burden yet leave the door open for ‘residual ex post enforcement’ in case a ‘passive’ investor’s initial intent changes later and becomes ‘active’. Furthermore, the ‘solely for investment’ exemption explicitly does not apply in case of cross-shareholding (the acquirer is a competitor) or interlocking directorates (a controlling shareholder, director, officer, or employee simultaneously serves as an officer or director of the issuer), and more generally if the holder of voting securities (1) nominates a board candidate or is represented in the corporate board; (2) submits a shareholder proposal for approval; or (3) solicits proxies.Footnote ¹⁰⁴ Therefore, US merger law is fit to capture a range of potentially problematic minority acquisitions by applying: i) an ex ante licensing regime (transaction filing and regulatory approval) when the potential of harm is most likely to materialise (active financial investments) and foreseeable (present corporate influence) and ii) an ex post safety valve (potential future liability and enforcement) when passive intent is negated by the acquirer’s subsequent corporate governance actions (ex post opportunism) or actual competition harm is evidenced (passive financial investments, individually or collectively). Seen in this light, ‘passive’ minority shareholding merely indicates cases where a full fact-specific antitrust analysis is required. If needed, such analysis will become relevant after the actual acquisition.

While investor passivity and the vertical agency problem are an inherent part of the modern corporation model and no obstacle to antitrust enforcement, the real challenge for both corporate and antitrust law is investor diversification. Common shareholding that is not only ‘passive’ but also ‘diversified’ changes the analysis completely: it surely can be ‘influential’ in the antitrust sense (in terms of its competition impact) but in counterintuitive ways (in terms of its mechanics). The ‘separation of ownership from ownership’Footnote ¹⁰⁵ and the ensuing horizontal agency problem (potential conflicts between diversified versus undiversified shareholders) challenge long-standing foundations and analytical frames in corporate finance and governance (shareholder unanimityFootnote ¹⁰⁶ or homogeneity as a classFootnote ¹⁰⁷) and competition law (control-based and entity-centric antitrust analysisFootnote ¹⁰⁸). The ‘democratisation of investment’Footnote ¹⁰⁹ brought about by the ‘index investing revolution’Footnote ¹¹⁰ has triumphantly enabled access to low-cost, widely diversified portfolios via a single investment product for a wider part of the population. However, this paradigmatic shift towards passive portfolio investment strategies has transformed not only the investment landscape but also it has created far beyond ripple effects yet to be fully appreciated. For one, the resultant widely diversified ownership structures (shareholder overlaps in many competing firms across industries) may well make firm-specific or market structure irrelevant, signalling a fundamental change not from firm ‘independence’ to inter-firm ‘control’ (the focal point of traditional antitrust analysis), but from shareholder ‘focus’ to investor ‘indifference’ (the new corporate reality brought about by financial innovation).Footnote ¹¹¹ Diversified investors are rationally interested in the aggregate return gained from their portfolio investments,Footnote ¹¹² following a portfolio-wide investment and governance strategy,Footnote ¹¹³ and less so in the individual performance of portfolio companiesFootnote ¹¹⁴ or spurring atomistic competition in product markets.Footnote ¹¹⁵ Common diversified shareholding takes the ‘depersonalisation of ownership’ of the public corporationFootnote ¹¹⁶ one step further: private property interests have become not only ‘split’ as in the age of managerial capitalism (specialisation of ownership and management functions) but also ‘parallel’ and ‘concurrent’ in the current capitalism era of professional portfolio managers and savings plannersFootnote ¹¹⁷ (many small shareholders are partial ‘co-owners’ not only of a single but also several competing corporate enterprises at the same time). Corporate ownership is rendered both diffuse and collectivised at the same time: the concern now is not the ‘objectification of the corporate enterprise’Footnote ¹¹⁸ (with business management separate and independent of its own shareholders – ‘owners’) but rather the ‘institutionalisation’Footnote ¹¹⁹ of investment and savings (with institutional intermediaries separate and independent from any individual business corporation, and its undiversified shareholders). Consequently, the cherished image of a (homogeneous) shareholder as a corporate ‘property owner’ is shattered, and we may no longer speak of individual shareholders (even if passive investors) as ‘residual risk bearers’ identifiable with a distinct corporate organisationFootnote ¹²⁰ as they may not fully bear any firm-specific risk or be concerned about targeted governance actions.Footnote ¹²¹

The horizontal ‘separation of capital from capital’Footnote ¹²² creates a ‘double split’ in the ‘atom’ of corporate property, both with regard to the unity and identity of its key actors (shareholders) and components (shares). On the one hand, institutionalisation has produced two sets of ‘owners’: ‘ultimate’ owners, who retain investment authority (but not necessarily voting control authority) and the associated risk for their investment choices, and ‘beneficial’ owners, who are only entitled to a future stream of profits from the invested funds (a claim that is more of a fixed contractual than residual nature).Footnote ¹²³ It will depend on the circumstances to decide whether and who of any institutional or individual investors have in fact residual or beneficial ownership status. On the other hand, diversification introduces another form of ‘decoupling’Footnote ¹²⁴ financial interests (risk) from corporate control (influence).Footnote ¹²⁵ That is, diversification of investment may render (legal) ownership ‘empty’Footnote ¹²⁶ and (voting) control ‘hidden’:Footnote ¹²⁷ the legal title is not congruent with economic interest and voting power need not follow the residual claim. As the economic link between cash flow and control rights is broken, economic and voting ownership is no longer proportionate to the nominal equity holding of a shareholder.Footnote ¹²⁸ Thus, the nominal level of the shareholding does not automatically reflect the level of economic risk and governance power a ‘shareholder’ may have.

This ‘double split of ownership’ in equity shareholding – i.e. ‘two faces of ownership’ and the ‘risk-bearing dilution’ brought about by passive, diversified investmentFootnote ¹²⁹ – has important implications for corporate governance. First, it significantly complicates the analysis as to who and to what extent is a residual claimant and thus raises questions as to the actual allocation of property rights in firms (real versus nominal owners). Second, it may lead to ex ante unforeseeable or de facto ‘morphable’ control situations. Given the fragmentation of shareholding in large public corporations and in the absence of a large dominant block holder or special asymmetric governance structures, control is likely shared among several shareholders and not ex ante ascertainable or fixed.Footnote ¹³⁰ Control is ‘morphable’ in that although no single shareholder has standalone control, some of them have the de facto ability to form control coalitions as a group.Footnote ¹³¹ Third, it raises the possibility that the individual self-interest may become ‘destructive’ both for the corporation and society as the pursuit of private profit and the exercise of voting rights may produce externalities on third parties (e.g. undiversified shareholders, other stakeholders, consumers) under certain circumstances.Footnote ¹³² Put differently, the concern is not only that individual investors may not be the ‘real’ owners and institutional investors are but also that no one is a truly responsible and concerned ‘owner’ in the traditional sense (sole proprietor).Footnote ¹³³ With the (partial) separation between financial risk and control due to diversification, it is not that self-interest is extinguished but rather that the very diversified investors’ self-interest is oriented towards the maximisation of their collective interests flowing from the pool of their portfolio invested firms’ profits (portfolio value maximisation). In addition, there are actors (institutional investors and business managers) that could potentially implement such altered preferences – to the extent they come to benefit themselves from any strategic shift away from individual firm profit maximisation and atomistic market competition. Indirectly, this complex and opaque constellation of agency relationships and fragmentation of property rights may further raise concern as to the robustness of market forces (capital market, market for corporate control, product markets) to efficiently allocate financial or investment capital, to drive corporate management towards improved performance and to move economic resources to their most productive uses and users across society.

The implications for competition law are equally significant as diversification and common ownership by large institutional investors induce corporate shareholders to be both rationally ‘apathetic’ and ‘indifferent’Footnote ¹³⁴ – a mutation of the archetype.Footnote ¹³⁵ At this point, it is worth revisiting the potential competitive effects of common shareholding in light of the above taxonomy on the various ‘shades’ of partial control related to partial ownership. Accordingly, common diversified shareholding, although individually a minority one with no standalone control, may be dynamically influential (potential effects on competition)Footnote ¹³⁶ as it may affect both the commonly held firms’ incentives to compete and their corporate governance.Footnote ¹³⁷ Such common shareholding may further qualify as disproportionately influential (given the potential concentrated influence and financial interest of the common owners linked to the shareholding)Footnote ¹³⁸ and also as bi-directionally influential (due to the potential mutual influence such shareholding may induce among the commonly held rival firms).Footnote ¹³⁹ Thus, depending on the circumstances, minority common shareholding that is considered passive and diversified may lead to situations of either ‘hidden control’ (shareholder concentration) or ‘hidden reciprocity’ (internalisation of externalities) in terms of effects despite its form (direction or symmetry of equity holding). Potential aggregation of individually small passive shareholdings and their parallelism in interest may give rise to cumulative de facto control and interactive, compound or network-like, competitive effects.Footnote ¹⁴⁰

The transformed quality of risk-diluted shareholding as closer to a debt holding in nature may add to such possibility. Interestingly, in this sense, ‘negative’ financial decoupling (disproportionate influence compared to the risk and size of a shareholding)Footnote ¹⁴¹ may indirectly lead to ‘positive’ linking of profits between competing firms (de facto profit correlation due to the rivals’ competitive relationship and their common ownership links).Footnote ¹⁴² Mutual internalisation of rivals’ profits may thus arise either due to common owners’ asymmetric corporate influence or their (symmetric) parallel financial interests, given their de facto control in portfolio firms’ governance and their aligned incentives to compete, or the induced financial dependence linked to their potential de facto position as largest shareholders and creditors of portfolio companies.Footnote ¹⁴³ Besides, when ‘exit’ is excluded for index funds,Footnote ¹⁴⁴ the ‘voice’Footnote ¹⁴⁵ of diversified institutional investors is amplified, thus potentially having a greater weight in managerial and firm governance decisions and indirectly in competition outcomes in product markets.Footnote ¹⁴⁶ Indeed, the voice of large ‘long-term’ institutional investors is actively encouraged by regulators and has become quasi-permanent (although ad hoc in its manifestation) and systemic (reaching their whole portfolio of invested companies).Footnote ¹⁴⁷ It follows that there might be several ways through which passive diversified investors may be competitively influential, either at present or in the future, and not only in one direction but potentially bidirectionally.

15.7 Implications and Conclusions

Coming full circle and looking back to press forward, history teaches us that it repeats itself. New forms of minority shareholding pop up as new ways of aggregating capital and savings and linking businesses are devised. Common shareholding by institutional investors brings to the fore the early unity and much-needed congruence between competition and corporate laws. The lesson for their continued interaction is for each discipline to assume a measure of modesty rather than ‘going all the way’ – alone – in solving the ‘common ownership trilemma’.Footnote ¹⁴⁸ ‘New finance’ with all the good that it brings for firms, markets, and people and the ‘new trusts’ with all their potential implications and distortions for the competition are one and the same problem: despite their modern functional split and specialisation, antitrust and corporate governance cannot bypass their deep interdependence both in terms of theoretical foundations and balanced regulatory solutions.

Yet, in this novel and unwieldy setting, a new role encounters antitrust: competition law enforcement may help rebalance and restore the initial allocation of property rights within firms (shareholders’ residual claim) and thus indirectly protect undiversified shareholders, to the extent corporate law control mechanisms (fiduciary duties) are ineffectual. Antitrust could therefore be used to protect noncommon shareholders who may be harmed – along with consumers – by suboptimal outcomes in the performance of individual corporate entities and industries. Developing an antitrust policy to tackle common ownership by large diversified institutional investors (shareholder concentration and diversification) and to supplement existing merger policy (market concentration) may unexpectedly return antitrust to its corporate law origins. Increasing diversification in financial investment and concentration and parallelism in corporate ownership calls antitrust to shift its operation from a pure conduct-oriented (‘strategic model’) to a structure-oriented approach: acting as a de facto early corporate law-like ‘structural model’ of checking the corporate structure and shareholder property rights exchange, internalising portfolio investment and governance externalities, and ensuring ongoing private versus public interest balancing.Footnote ¹⁴⁹

While antitrust is to retain its focus on economic effects, the source of potential competitive effects from minority shareholding and common ownership originates in changed corporate ownership structures well below any formal competition law threshold of ‘control’ (as in EU merger control). The foundations of competition and merger control when designing their scope and aims may be challenged, however, considering: i) the liberalisation of corporate law lifting intercorporate ownership and influence restrictions and ii) recent organisational and financial market developments undermining early corporate law assumptions (no separations of ownership or investment from control or consumption) that antitrust inherited. By turning its look inside corporate governance and shareholder structures, antitrust could provide an effective alternative tool to holistically capture problematic minority shareholdings that create both agency problems or conflicts of interest and market power concerns (just as early corporate law regulation of mergers and shareholding acquisitions did). Such a solution would not only rebalance disperse shareholders’ private interests (property rights) inside the firm but in view of the historical split between corporate and competition law, it would also aim to safeguard and prioritise the public interest (consumer welfare). Against this backdrop, antitrust can simply not afford to not look closer at minority shareholding structures that ‘came to fall between the cracks’. Essentially, and unlike the main antitrust rules that target behaviour (Sherman Act Sections 1 and 2; Articles 101 and 102 TFEU), merger control rules within modern competition law regimes have inherited and preserved to a certain extent corporate law’s ‘structural’ regulatory approach. The question now is that merger control adjusts its jurisdictional ambit and tools below and beyond obsolete legal thresholds and economic assumptions. The concentration of corporate ownership through novel shareholding structures and intermediaries may produce competitive influence, in significant although unfamiliar ways, which impels antitrust to be alert in shifting its attention and finetuning its enforcement tools. Curiously, as the new reality of the such evolved capitalist environment and organisational structure sets in, ‘anti-trust’ may be found anew to be well worth its name.Footnote ¹⁵⁰

At the same time, on a substantive level, common institutional shareholding calls antitrust to be not merely future proof but ‘future perfect’: new intriguing possibilities, both with regard to novel theories of harm and efficiencies, will have to be acknowledged and incorporated into competition law enforcement if it is to remain relevant. In light of its mixed effects, competition law should go past imposing hard limits or any per se prohibition against common ownership, even if a ‘structural’ approach is systematically employed to assess ad hoc their competitive significance and effects in the specific case. In such case, antitrust will have to enrich, refine, or complement its merger law-based measurement tools (HHI and MHHI), depending on the circumstances. In other words, the ‘structural’ approach will need to be combined with case-by-case analysis based on the specific factual context. In this connection and as illustrated above, it is important to realise that the combination of concentration and diversification related to common institutional ownership lends corporate property rights ‘dual’ or ‘quantum’ qualities: 1) minority shareholding may be both ‘passive’ and ‘dynamically influential’ at the same time, so long as there is some competitive relationship between the interlinked firms that may be undermined;Footnote ¹⁵¹ 2) diversified shareholders may be both ‘owners’ of the firm but ‘empty’ of any economic interest in its performance, challenging their residual claim status; 3) control may be ‘hidden’, concentrated and disproportionate in its actual ex post manifestations compared to its ex ante hypothetical properties considering the size and risk attached to a stand-alone shareholding; 4) in the presence of common ownership, the quantum ‘atom’ of corporate property may be composed of not only ‘solid particles’ (control rights) but also ‘invisible waves’ (parallel interests), the latter fundamentally changing the traditional identity of an equity share and a shareholder.Footnote ¹⁵² Yet, the existence and impact of these parameters are hard to pin down in the abstract; better observation and empirical evidence are needed and also bold theoretical leaps forward to fill gaps in our understanding of the emerging reality. While by no means easy, ‘the only way out is through’.Footnote ¹⁵³

Generations of law and economic scholars will no doubt devote lifetimes to unravel the ‘gordian knot’ presented by common ownership. As Minority ReportFootnote ¹⁵⁴ reminds us one may not be able to predict the future from the past or credibly form a single-minded assessment of the likelihood of harm materialising or not. Still, it is possible to change the future once one is aware of its prospect. For the sake of a sustainable capitalist future and society, may both public officials and economic agents be mindful of the power they possess and use it with wise restraint and prudent foresight.

16.1 Introduction

In recent years, the competition law community has become absorbed in discussions around the role of competition law in the digital economy. Debate typically centres on the largest digital platform companies: Google, Apple, Facebook, Amazon, and Microsoft (‘GAFAM’).Footnote ¹ Such studies have helped to shift competition law beyond the neoclassical price theory framework that underpins the consumer welfare standard, raising significant issues including what is the most appropriate unit of analysis for understanding competitive interactions in the digital economy and how to define power in the digital age.Footnote ² These are pertinent questions that will rightly continue to attract the attention of scholars and regulators.Footnote ³ Still, in this brief chapter, we wish to highlight important shortcomings of the existing literature and therefore promote a different perspective on the digital economy than the ‘monopoly power’ narrative that has dominated competition law discourse regarding digital platforms in recent years.

The current dominant narrative on the power of digital platforms stays silent on the cui bono question: Who benefits from this digital transformation? To answer this, one needs to address the inner logic of modern financial capitalism, the prevalence of the shareholder value principle,Footnote ⁴ the transformation of the ‘shareholder’ concept with the emergence of institutional investors as primary holders of global capital as an important episode in the trend towards financialisation,Footnote ⁵ and the way this impacts on the competitive strategies of Big Tech.

Financialisation is a broad, and somewhat contested, term. Economist Gerald Epstein offers a frequently cited definition of financialisation as ‘the increasing role of financial motives, financial markets, financial actors, and financial institutions in the operation of the domestic and international economies’.Footnote ⁶ This study focuses on the corporate aspects of financialisation, namely the growth of the financial dimension of value generation in the digital economy and the reorientation of corporate governance around the principle of shareholder value maximisation. Following Hyman Minsky, we also highlight the increasingly speculative nature of economic behaviour under financialisation.Footnote ⁷

Subject to a few exceptions, the extant literature fails to acknowledge the importance of financialisation in driving the growth of GAFAM.Footnote ⁸ While the majority of scholarship on the digital platforms focuses on network effects and market tipping in promoting their growth, financialisation is also key to understanding value capture in the digital economy and in particular the competitive strategies employed by the same platforms. As noted, here we understand financialisation to mean the growth of the financial dimension of value generation and the reorientation of corporate governance around the principle of shareholder value maximisation.Footnote ⁹ Significantly, it is also associated with an increasingly speculative approach to investment.Footnote ¹⁰ Nowhere is this more apparent than in the digital economy. Expectations around the future profits of technology firms are extremely buoyant, and often quite distinct from the reality of markets as they presently exist.Footnote ¹¹ By achieving central positions within digital ecosystems, GAFAM are beneficiaries of such financial speculation. They enjoy enormous financial clout that then enables them to further entrench their power, primarily through intensive merger and acquisitions (‘M&A’) activity. This has allowed them to expand outside of their core business activity into adjacent and/or overlapping fields of activity. That is, financialisation plays a vital role in intensifying the accumulation of power, and therefore rents, in the digital economy. Too much focus has been placed on digitisation without comprehending that the shift may not have had such dramatic economic and social consequences if it was not paired with the emergence of financial capitalism.Footnote ¹²

This financial capitalism perspective invites us to consider who really benefits from GAFAM’s dominance. Competition law typically considers participants in the economy in the binary terms of consumer welfare and producer welfare, with the terms ‘consumer’ and ‘producer’ stylised concepts that mask the variety of positions an economic actor may hold within an economy. Under financialisation, people are not simply consumers but may also be investors. It is therefore worth considering the extent and distribution of equity ownership in digital platforms. Given that equity ownership is heavily skewed towards the wealthiest and most privileged in society, we consider that the role of competition law may be to represent those who do not share in the power of GAFAM through their involvement in capital markets.

The further limitation we highlight is a failure to consider GAFAM’s dominant position in the global economy from an organisational perspective. As in the field of competition law more broadly, scholarship on competition in the digital economy pays little attention to corporate governance. While many commentators observing GAFAM’s M&A strategy draw a comparison with the conglomerates that dominated the American economy following World War II, there has been insufficient work to uncover the accuracy of this analogy. The post-war conglomerates were run by and for a managerial class, largely free from interference by shareholders.Footnote ¹³ This may be contrasted with the corporate governance of firms following financialisation, in which the interests of shareholders are paramount. We undertake empirical work to uncover whether GAFAM truly do resemble historical conglomerates rather than the financialised firms typical of the post-1980 period. We find that GAFAM’s corporate governance regimes, to varying degrees, exhibit a hybrid blend of characteristics and resist easy categorisation.

The chapter is split into three substantive sections. Section 16.2 uncovers the interaction between financialisation and growth of the digital platforms. Section 16.3 addresses the question of who benefits from their dominance and the implications of this for competition law. Section 16.4 presents empirical evidence on the corporate governance regimes of the GAFAM firms, comparing and contrasting them with the managerial primacy of the post-war conglomerates and the shareholder-led approach that typifies financial capitalism. Section 16.5 concludes.

16.2 The Changing Competitive Game: Financialisation, Futurity, and the Dominance of Digital Platforms

Work on competition in the digital economy has helped to move competition law beyond its traditional focus on output restriction by product market monopolists.Footnote ¹⁴ This is apparent, for example, in the literature on digital ecosystems, which have emerged as a new structure of economic relationships.Footnote ¹⁵ Such work helps us to understand the centrality of GAFAM in the digital space; they have attained ‘architectural power’, positioning themselves so that they influence the way that ecosystems are structured and shaping the allocation of value between ecosystem participants.Footnote ¹⁶ Undoubtedly, this governance architecture may generate value and promote innovation. However, competition law scholarship still does not fully grasp the dynamics driving GAFAM’s dominance of the digital economy and its long-term impact on social and consumer welfare.

Arguably, the most important factor is the consolidating effect of financialisation. Modern competition law holds that if a firm engages in some type of conduct, or wishes to merge with another, this is more than likely because it would enhance efficiency.Footnote ¹⁷ Yet, under financial capitalism, competitive behaviour is not only about improving efficiency to succeed in product markets in the present, as assumed by the competition orthodoxy. Rather, they engage in ‘futurity-led’ competition, in which firms seek capital asset appreciation by convincing investors of their expected dominance in the future – futurity denoting the reorientation of economic activity towards the future and firm valuation coming to rest on expected future profits.Footnote ¹⁸

In a 1925 manuscript, the institutional economist John Commons explained futurity in the following manner:

Thus, futurity is ‘a factor that indicates anticipation’.Footnote ²⁰ Writing a century ago, Commons identified the emergence of futurity by analysing a series of cases heard by the US Supreme Court at the end of the nineteenth century. Prior to this time, firm valuation was based solely on the estimated market price of tangible assets. That is, firms were treated as if they had ceased to trade and were simply a collection of illiquid assets awaiting liquidation. Commons’ analysis of futurity centres on the replacement of this practice by the treatment of firms as living entities, as ‘going concerns’, that are expected to earn profits in the future.Footnote ²¹ By the turn of the twentieth century, the US Supreme Court had come to recognise that firms can own something that they do not actually possess, expected future profits, the value of which began to be incorporated into firms’ value.Footnote ²²

Futurity is inherently tied to the financial system because it is finance that provides ‘the necessary link between the present and the future’.Footnote ²³ Under financial capitalism, unfettered financial markets are driving an intensification in futurity. With their ‘animal spirits’ unleashed, investors in the modern economy care less and less about the profits and cashflow firms achieve in the present day and instead aspire speculatively towards realising tremendous profits many years in the future.Footnote ²⁴ In this configuration, where futurity is key, value is not determined and generated by the present market exchange but by an expected succession of events. Consequently, we see a ‘subtle shift of mindset from profit (and isolating mechanisms) to wealth creation (and the potential for asset appreciation)’.Footnote ²⁵

Nowhere is futurity-led competition more apparent than in the digital economy, in which economic decision-making is driven by speculative visions of extraordinary profits in the future that is seemingly detached from present-day reality. Consider, for example, the extraordinary valuations of companies like Uber and Tesla that are yet to record meaningful, sustained profits.Footnote ²⁶ As noted by economist Ronen Palan:

The major digital platforms have benefitted immensely from this trend, with their market capitalisations skyrocketing in the last decade. At the time of writing, GAFAM are five of the six most valuable companies by market capitalisation globally (their hegemony broken only by Saudi Armco): Apple is valued at $2.6 trillion, Microsoft is valued at $2.5 trillion, Amazon is valued at $1.8 trillion, Alphabet is valued at $1.9 trillion, and Facebook (Meta) is valued at $0.9 trillion.Footnote ²⁸ Together, they constitute 24.7% of the Standard and Poor’s (S&P) 500’s total market capitalisation.Footnote ²⁹ Such valuations are principally motivated by high expectations for phenomenal profits in the not-so-immediate future because of their position as gatekeepers controlling important bottlenecks in digital ecosystems (for example, operating systems, search engines, app stores, and the cloud). To the extent that, in view of its essential characteristic of futurity, the main source of value in the digital economy emanates from financial markets valuing expected returns, ecosystem orchestrators attract important investments, in particular, if they command control over an essential bottleneck for complementors (gatekeepers).

In turn, the futurity-driven market capitalisations enjoyed by GAFAM provide them with immense financial firepower, which enables them to further shape the digital economy in their own favour.Footnote ³⁰ There appears to be a ‘growth-funding’ feedback loop, ‘by which the future supply of effective funding increases as a result of the conditions created by a speculative expansion, and ends up lowering the cost of capital’.Footnote ³¹ Moreover, digital platforms directly benefit from financialisation as they own large stocks of financial assets (bonds, cash, or other financial instruments).Footnote ³² Rodrigo Fernandez and his co-authors note that the seven largest Big Tech companies (Apple, Microsoft, Alphabet, Facebook, Amazon, Alibaba, and Tencent) own more than $700 billion financial assets, including cash, government bonds, corporate debt securities, mortgage-backed securities, investments in money market funds, and equity securities, among others,Footnote ³³ which represent a much higher percentage of their total assets than traditional S&P 500 companies.Footnote ³⁴

Significantly, GAFAM are able to use their enormous financial clout to further entrench their power through engaging in intensive M&A activity. In the period from their founding to 2018, Alphabet has acquired 214 companies, Amazon 77 companies, and Facebook 65 companies.Footnote ³⁵ Similarly, between 1991 and 2018, Microsoft has acquired 189 companies and Apple 89 companies.Footnote ³⁶ Through such transactions, GAFAM have been able to further develop digital ecosystems and cement their critical positions within them. Amazon, for instance, started off as an online retailer of books before being vertically and horizontally integrated with other entities which enabled it to become a vendor of various products and a media and entertainment company that competes with other media and entertainment companies the products of which it also sells on its platform. Amazon has also expanded its activities into Internet cloud business and storage and the transmission of content to consumers.

Financialisation and futurity, therefore, create a virtuous feedback loop for these firms in which power in the digital economy precipitates financial power, which in turn enables them to deepen their power in the digital economy. In this manner, the speculation that characterises the wider digital economy prompts the development of a more ‘rational bubble’ in GAFAM stock valuations.Footnote ³⁷ The perception that they (could) control a valuable bottleneck that may provide them with a sustainable competitive advantage and abnormal profits in the long term is a crucial driver for strategic action by the management of the relevant firm. This relatively new phenomenon is reinforced by the learning-by-doing effects engendered by the use of data accumulation, advanced artificial intelligence capabilities, and the central positioning of digital platforms as gatekeepers in various economic sectors in the digital economy, which enable them to draw on their significant predictive power and to potentially leverage that to a central positioning in the financial anchorages of the global economy.Footnote ³⁸

This financial capitalism account sheds light on the dynamics driving the expansion of digital platforms, moving the discussion beyond efficiency-based explanations and inviting us to view their growth with a greater degree of scepticism.

16.3 Who Benefits from the Dominance of the Digital Platforms? Distributional Implications and Agent-Based Modelling

Beyond highlighting the role of financialisation and futurity-led competition in driving the growth of the digital platforms, the financial capitalism perspective invoked here also invites us to consider who actually benefits from GAFAM’s dominance of the digital economy. The current market-based competition law framework does not attempt to provide a holistic answer to this vital question, focusing narrowly on consumer welfare. Yet, following financialisation, people are not only consumers but also may be investors who share in the benefits of GAFAM’s power through stock ownership.

Notably, despite a widespread myth about the ‘democratisation of finance’, shareholding remains highly concentrated among the wealthiest and most privileged in society.Footnote ³⁹ In the US, for example, the vast majority of people own very little or no corporate equity: at the end of the first quarter of 2021, the wealthiest top 1% of Americans owned 20% of corporate equities and mutual fund shares, with the top 10% holding 13% and the entire bottom 50% just 0.2%.Footnote ⁴⁰ As highlighted in the following section, institutional investors, who invest on behalf of this wealthy minority, own significant stakes in each of the GAFAM firms. Clearly, then, it is the most well-off who benefit from GAFAM’s dominance via stock ownership.

This distributional impact of competition law enforcement has not yet been adequately examined, although there have been some efforts, still uncomplete, to move towards this direction.Footnote ⁴¹ To the extent that competition law only focuses on consumer welfare, defined as the behaviour of a representative agent, the consumer, it is agnostic as to distributional effects to other sociological categories of agents, such as workers, investors, and other categories of individuals that may be delineated according to their average income or wealth.Footnote ⁴² Such analysis, if it is to be complete, needs to be performed at the level of each jurisdiction, taking into account all the affected stakeholders.Footnote ⁴³ One can imagine that such analysis is performed at least intuitively when deciding about the stance of a specific jurisdiction with regard to the social costs (and benefits) of the economic power of Big Tech in order to design their specific regulatory response, in view of their institutional capabilities and resources.Footnote ⁴⁴ A more systematic analysis would require the use of different tools than those traditionally employed by competition authorities, in order to map the power of these digital platforms and assess their incentive and capacity to produce effects in each jurisdiction and determine the social ‘geography’ of such direct or spill-over effects.

An important tool that has not been adequately explored so far by competition law literatureFootnote ⁴⁵ is agent-based modelling.Footnote ⁴⁶ This tool provides a bottom-up approach to simulate a system of heterogeneous autonomous agents, thus accounting for different attributes, such as the size, the business model, as well as the specific ownership structure and corporate governance of undertakings, and could also integrate a dynamic perspective by designing these agents to be adaptive through learning. A similar modelling can be done for various sociological categories of individuals, such as ‘investors’, ‘labour’, and ‘consumers’, accounting for their income, education or wealth level, varying degrees of rationality, thus not relying on the average behaviour of individuals defined in abstracto but on the basis of their real attributes and those the theory/hypothesis to be tested considered important. The model may not only focus on price-system intermediated interactions but also centre on or combine non-price ones. It may be possible to also develop a typology of realistic rule sets to be applied to all or categories of agents, as well as different agent environments (taking into account the different spheres of competition – markets, ecosystems, sectors) that more comprehensively account for the complexity of these interactions and relationships (for example, competition, cooperation, co-opetition, ownership, control, influence) and open up to various behavioural frameworks that fit the research question asked (this will be different, for instance, if the research focuses on the impact on privacy, prices and output, quality, innovation, democracy, among other dimensions). The interactions to take into account may be financial flows, unique visitors metrics and time spent on a website, information exchange/data flows, and the expression of emotions (‘likes’, ‘dislikes’, ‘friends’, ‘followers’) in order to determine the ‘ties’ between the various agents and the topography of the network.

Calibrating such models may take significant resources, and naturally, their degree of validity may depend on the way the model matches with the available data and on the initial conditions chosen to design the model. Although such tools also require significant sources of data, it is easier than it has ever before to gather in view of digitisation and the expansion of the digital economy. The agent-based model will run on various simulations and other computations and will eventually provide important insights through the visualisation of the interactions between agents, and the predicted evolution and outcomes of such interactions in different virtual worlds. The economic process would thus be modelled as a dynamic system of interacting agents. The topology of such interactions between agents is complex as the scale of the system/environment the agent-based model aims to explain is driven by the specific social phenomenon of interest. The tool may thus enable competition authorities to better capture emerging phenomena and to improve their understanding of the broader social impact of the examined behaviour in the context of a specific jurisdiction, not only at a purely abstract level but also taking into account a more realistic depiction of the status and motives of the agents. However, one should note the limitations of such tools, in view of the important complexity of adaptive systems, and the evidential value of simulation methods in legal processes. Notwithstanding, the tool may be employed more safely for case selection and prioritisation.

16.4 Are Digital Platforms Typical Conglomerates?

The intensive M&A activity undertaken by GAFAM is a well-documented feature of the digital economy. In light of their merger-induced diversification, commentators frequently invoke a conglomerate analogy to describe the GAFAM firms.Footnote ⁴⁷ However, analyses have so far stopped short of testing this analogy by evaluating their corporate governance regimes. Most work has so far focused on the governance of the Big Tech ecosystems, which is of course important and crucial if one is to superpose some form of public governance to the private governance structures that have been put in place by Big Tech platforms, through technology, the use of contract law or even what some have named the ‘uncontract’,Footnote ⁴⁸ but little has been done to explore from a competition law perspective their corporate governance. As a discipline, competition law treats firms’ internal dimensions as a ‘black box’ that left to corporate law.Footnote ⁴⁹ Here, we open this black box and interrogate the validity of the conglomerate hypothesis, assessing whether GAFAM resemble the conglomerates that were a prominent feature of the post-war economy or if they are closer to the financialised firm typical of the post-1980 period.

In the decades following World War II, a combination of strict financial regulation and atomised shareholding left managers as the predominant force in US corporate governance.Footnote ⁵⁰ Unlike under financial capitalism, shareholders were paid little mind.Footnote ⁵¹ Managers sought growth through conglomeration. These large conglomerates dominated large chunks of commerce and emerged as multinational corporations organised as an M-form business organisation that expanded their reach in different parts of the globe with the view to develop ‘synergies’. Conglomeration and the concept of synergies were however increasingly subject to extensive criticism in business literature in the 1980s: Michael Porter criticised the relevance of portfolio management, at least for advanced economies, and explained in his work ‘how diversified companies do not compete; only their business units do’ and that ‘diversification inevitably adds costs and constraints to business units’.Footnote ⁵² These conglomerates also relied on the central role of important ‘industry captains’, such as Harold Geneen of the ITT Corporation or James Ling, of the now defunct Ling-Temco-Vought conglomerate, that built the acquisitive conglomerate model through a succession of M&As in the 1960s and early 1970s.Footnote ⁵³

Notably, conglomerates pioneered the strategy of corporate growth through capitalising on financial markets – a strategy now being taken up by the digital platforms.Footnote ⁵⁴ Many conglomerates acquired more than 50 firms during the decade up to 1968,Footnote ⁵⁵ with the very largest, including ITT, making more than 20 acquisitions in 1968 alone.Footnote ⁵⁶

One of the reasons and possible advantages of conglomerate mergers during this period was the need to develop internal capital markets at the level of the conglomerate, as the managers of the firm were thought to have information advantages over the external capital markets that were not that developed in the 1960s in order to allocate capital for projects with higher rates of return. Conglomeration enabled cross-subsidisation between different divisions within the same diversified company.Footnote ⁵⁷ Some firms were perceived to have information advantages over the external capital markets, for instance in the allocation process of capital and the operational aspects of each business division. However, as external capital markets develop, ‘many firms can provide company-specific information to the capital markets directly’ and thus ‘more easily bypass firm internal capital markets for investment funds’.Footnote ⁵⁸

Crucially, conglomerates were a key driver of technological progress. Profits were retained and reinvested into the corporation. Economist John Kenneth Galbraith noted how managerialism entailed a ‘shift of power in the industrial enterprise […] from capital to organised intelligence’, allowing for the significant commitment of time and resources necessary to produce technical innovation.Footnote ⁵⁹ Yet, the concept of conglomeration came under attack in the context of the structural crisis of the 1970s.Footnote ⁶⁰

Furthermore, the development of external capital markets in the 1970s, in particular with the development of technology enabling external (to the firm) capital markets to work around the clock and provide immense amounts of information and data for analysis, reduced the importance of the information advantages of internal capital markets.Footnote ⁶¹ This has led to the emergence of different organisational structures.

Following financialisation and the onset of shareholder primacy, the corporation was mostly seen as a portfolio of activities, managed according to their financial performance (in terms of rate of return on investment), rather than defined in terms of synergetic productive capabilities at a conglomerate level. To raise share prices, managers divested large swathes of their firms. By 1995, the average large firm that had started the 1980s operating in dozens of industries operated in only one.Footnote ⁶² In parallel, the adoption of open-system standards by major players in the computer industry led to the weakening or abandonment of internal research and development within major corporations in favour of patenting, cross-licensing, outsourcing, and the takeover of start-ups. Technically, this was accompanied by the design and development of modular components that were manufactured by offshore companies and vertically integrated into niche markets. Financially, the shift was made possible through the rise of organised venture capital, cushioned by large investments from large retirement and pension funds.

This technological and economic transformation was built around a new mantra: shareholder value maximisation, which forms a key tenet of financial capitalism. The shareholder primacy principle changed managerial priorities from that of maximising growth by re-investing corporate savings in the long-term productive potential of the corporation (the ‘principle of retain and re-invest’) to that of maximising stock value through extensive buybacks of corporate stocks (share repurchase) in order to inflate stock prices as the resulting artificial scarcity of shares boosts their value.Footnote ⁶³ Disciplined by a corporate market for control dominated by financial interests, in particular institutional investors, corporate managers became increasingly aligned with the interests of shareholders, and adopted strategies aiming to increase the price of their corporate stocks. They downsized their corporations (in particular cutting labour costs) in order to create short-term shareholder value and distributed the freed-up corporate revenues to financial interests, particularly shareholders, instead of re-investing them in the corporation (the principle of ‘downsize and distribute’).Footnote ⁶⁴

Literature on financialisation tells us that there are various characteristics of a ‘financialised’ firm: 1) pronounced institutional investor shareholding; 2) extensive influence of such institutional investors, including through voting rights; stock-based corporate remuneration; and 3) the distribution of cash to shareholders through dividend issues and share buybacks, typically at the expense of productive investment (for example, in research and development).Footnote ⁶⁵ Put together, we may imagine a ‘financialised firm’ index, composed of these three elements, that indicates the extent to which a firm’s corporate governance is guided by shareholder primacy. By examining GAFAM through the lens of these factors, we may determine whether their corporate governance resembles that of the post-war conglomerate or that of the post-1980 firm.

Current accounts of the development of digital conglomerates do not take into consideration these internal (ownership structure and corporate governance-related) drives for the behaviour of Big Tech platforms. Nicolas Petit builds an argument for a competition law immunity of Big Tech platforms (which he calls ‘moligopolies’) distinguishing them from monopolies, as moligopolies ‘channel sizeable amounts of resources into R&D’ and invest in human resources, particularly entrepreneurship, thus indirectly referring to the fact that moligopolies retain their earnings and do not distribute.Footnote ⁶⁶ However, this assertion is not factually supported. Marc Bourreau and Alexandre de Streel make the analogy with conglomerates, but they arrive at different conclusions than Petit, noting the gatekeeping role of Big Tech platforms and their ability to pre-empt competition by, from the outset, killing any opportunity for a potential competitor to emerge.Footnote ⁶⁷ Similarly, Jean Tirole raises concerns as to the adoption of possible bundling practices that may exclude new entrants from the markets in which are active these conglomerates.Footnote ⁶⁸ What is crucially missing from these analyses, however, is the consideration of the internal dynamics of each of these firms, as this is determined by their ownership structure and corporate governance, this time not in order to assess the broader social costs of digital platforms, but in order to gather elements that would help policy-makers to predict their behaviour.

We examine the ownership structure of Microsoft, Apple, Amazon, Alphabet, and Facebook. We observe that each firm is predominantly owned by institutional investors. Table 16.1 illustrates the proportion of GAFAM shares held by institutional investors versus non-institutional investors. Evidentially, institutional investors are major shareholders in GAFAM, cumulatively holding between three-fifths and four-fifths of their shares. However, these figures overstate the importance of institutional investors within the digital economy, especially within Google, Facebook, and Amazon. Google and Facebook offer dual-class shareholdings. Share class differentiation reduces the strength of the link between observed shareholding and voting rights, skewing voting rights heavily towards firm insiders. Although Amazon’s share structure is more traditional, with one vote per share, Jeff Bezos owns a clear majority of shares.Footnote ⁶⁹ Microsoft and Apple also adhere to a traditional share structure, with institutional investors dominating share ownership and therefore voting rights. This is illustrated in Table 16.2, which shows the investors with notable voting rights in Google, Facebook, Amazon, Apple, and Microsoft, respectively. Although institutional investors hold a majority of stock in GAFAM, they are not necessarily endowed with the greatest voting rights. Rather, individuals, typically founders, retain the most significant voting power in Alphabet, Facebook, and Amazon.

After institutional investor share ownership and voting power, the next aspect of firm financialisation to consider is how corporate executives are remunerated. In this regard, GAFAM companies may be viewed as more conventionally financialised. Although the chief executive officers of Alphabet, Amazon, Apple, and Facebook did not receive stock-based compensation in 2020, the other C-suite level executives at these firms did. All C-suite executives at Microsoft, including its chief executive officer, Satya Nadella, received stock-based compensation in 2020.

The final component of the hypothesised financialised firm index is to consider how much cash each of the GAFAM firms distributes to shareholders through dividend issuances and share buybacks. Ordinarily, there is a view that financialised firms distribute their cash to shareholders at the expense of making longer-term productive investments. Again, the situation with GAFAM is complex and does not fit neatly into this narrative. While both Microsoft and Apple buy back shares and issue dividends, Amazon does not buy back its own shares or issue dividends, and Alphabet and Facebook, although they buy their own stock, do not issue dividends. Furthermore, as is frequently pointed out, these companies also invest heavily in research and development. Digital platforms seem to constitute a hybrid of the old managerial ‘retain and reinvest’ and the financialisation ‘downsize and distribute’ models.

Based on the above observations, we find an interesting paradox – despite the tremendous goodwill that GAFAM enjoy in capital markets, they are not uniformly financialised in accordance with the three parameters noted above. Rather, we find a mixed picture, with the corporate governance regimes of Amazon, Facebook, and Google closer to the managerialism of the post-war conglomerates and Apple and Microsoft closer to the shareholder-centric firm that characterises financial capitalism. We hope that the exercise undertaken here, opening the black box of corporate governance, may promote a more nuanced understanding of GAFAM going forward.Footnote ⁷⁰

16.5 Conclusions

In this chapter, we explore a missing dimension of the discussion over Big Tech platforms in competition law: financialisation and its impact on competitive strategies and the social impact of Big Tech’s economic power. We draw on insights from economics and corporate governance to offer a broader analysis of the five major digital platforms. We offer three contributions. First, we argue that financialisation and futurity have played a prominent role in GAFAM’s expansion. Second, taking into account the social dimension of financialisation and the heavily skewed distribution of stock ownership in society, we argue that the distributional implications of Big Tech Power and of competition law enforcement, respectively, need to be assessed more systematically. We examine that the tool of agent-based modelling may contribute to this analysis and help us move beyond the traditional focus on consumer welfare. Third, we explore as an additional dimension of financialisation the role of ownership structure and corporate governance in influencing the broader economic model of behaviour followed by Big Tech platforms and we offer an empirical examination of GAFAM’s corporate governance regimes, interrogating the conglomerate analogy frequently invoked in the digital economy,