A PYMNTS Company

A Paradigm Shift in the Antitrust Analysis of Common Ownership In India

 |  March 1, 2021

CPI COLUMNS South Asia cover

By Saatvic1

 

Introduction

While antitrust authorities have developed time tested methods to analyze the competitive effects of mergers, assessing acquisitions of minority stakes that create common ownership links has been a sticking point.  Traditional quantitative tools like concentration indices are only useful when control completely switches hands.  Minority investments lead to multiple owners, and the degree to which each owner influences the actions of the company are not usually discernable.

The Competition Commission of India (“CCI”) has traditionally sidestepped the issue and followed a “worst case scenario” approach – identifying any firms in the investor’s current portfolio whose activities overlap with that of the firm whose shares are being acquired and analyzing the transaction as a full-blown merger between such portfolio firms and the target.  The pitfalls of this approach are not hard to see – at some point the CCI would be confronted with a situation where an investor owning less than 10 percent of one firm wants to acquire less than 10 percent of another firm in the same industry, and together these firms supply a “high” proportion of some product market.  Naturally, such a situation would fail the “worst case scenario” test, and the CCI would have to confront issues surrounding common ownership directly and rigorously. Given the continued mushrooming of institutional investing in India, the results would provide valuable clarity to future investors eager to do business in India.

The inevitable happened when ChrysCapital, which already owned small stakes in various pharmaceutical companies, wanted to up its stake in Intas Pharmaceuticals from 3 to 6 percent.2  The CCI approved the transaction3 subject to behavioural remedies, including removing a nominee committing to not exercise veto rights related to certain subjects at a competitor.  However, while the approval came on April 30, 2020, the CCI has not yet uploaded a final order at the time of writing (though a draft order4 was briefly uploaded and then rescinded in 2020 (“Draft Order)).  When the final order is issued, it will likely have profound effects on the behaviour of institutional investment in India. The direction of that effect is still an open question – any perceived heavy-handedness could deter further investments, while the establishment of a replicable and analytically sound procedure to deal with common ownership would significantly reduce regulatory uncertainty and boost further investments.

   

Assessing Unilateral Effects

Theory of Harm

While the unilateral theory of harm in such transactions is qualitatively the same as in the full-merger case – increased concentration might relax competitive constraints between firms – the mechanisms are not straightforward.  Say an investor with a prior stake in firm A now also acquired a stake in firm B.  B’s manager might realize that the common owner would benefit from some increase in price, since part of the revenue lost to A would flow back to the common owner as profits through their share in A. However, there are two complicating factors. First, the common owner owns only a part of A, so not all the profits B corners flow to the common owner. The incentives are lower than what they would be if the common owner fully owned A. Thus, the anticompetitive effects change with the level of the common owner’s financial interest (the percentage of profits they are entitled to) in each firm. Second, the common owner is just one of several owners, and the non-common owners would not want to increase price, since they do not receive any part of lost profits captured by other firms. Thus, the common owner’s corporate interest, or the degree to which the common owner can control the firm’s actions, is also pivotal.

A Tailored Concentration Index

Real-world transactions require antitrust authorities to not just acknowledge these insights, but also form an opinion on their effect on the magnitude of likely anticompetitive effects.  To this end, economists have developed a Modified Herfindahl-Hirschman Index (“MHHI”), which allows firms to have multiple owners, each with potentially different financial and corporate interests in multiple firms. 5 The resulting index is analogous to the traditional Herfindahl-Hirschman Index (“HHI”).  In fact, the MHHI is simply the HHI plus a measure of additional concentration arising from common ownership. Its calculation requires data on market shares, as well as financial and corporate interest structures of firms with common owners. Like the HHI, it is positive, but it is not bound at 10,000 because of the additional effects of common ownership. The higher degree of control by common owners, the higher the MHHI; and for a given degree of control, the higher the common owners’ financial interest, the higher the MHHI.

The MHHI is thus a powerful screening tool like the HHI and is accordingly recognized the European Commission Horizontal Merger Guidelines6 and working documents7. The European Commission considered or referenced MHHI estimates in its analyses of the Exxon/Mobil,8 Alcan/Pechiney,9 Schneider/Legrand,10 Munksjo/Ahlstrom,11 Glencore/Xstrata,12 and Dow/Dupont13 transactions.

A key sticking point has been the need to take a stand on the degree of corporate interest of various owners. One might think that the percentage of voting stock is a good indicator, but there are two confounding issues.  First, different stocks come with different rights, e.g. veto rights over certain issues and the ability to appoint board members, which can confer significant power.  Moreover, even if we only considered voting stock, someone controlling 51 percent of the votes would get their way 100 percent of the time, so the relationship is not straightforward.

Since the European Commission last passed an order that referenced the MHHI, new research14 has suggested that the relationship between shareholding and control/corporate interest is best described by the Banzhaf Power Index (“BPI”),15 which was originally developed to examine block voting in political elections where voting power is unequally distributed (e.g. the U.S. Electoral College). Under the BPI, the probability of large shareholders achieving their preferred outcomes is higher than their proportional shareholding, while for small shareholders it is lower than their proportional shareholding.  As the shareholding approaches 50 percent, the level of control approaches 100 percent.  Thus, the BPI gives small shareholders less, and large shareholders more control than their shareholding level.  Given that common ownership issues usually arise with respect to minority stakeholders, adopting a level of control proportional to shareholding would therefore be the conservative option, as it would confer a higher degree of control than the BPI would imply.

Thus, there is now a good case for the MHHI to become an integral part of antitrust reviews of transactions that create common ownership links.

Unilateral Assessment in ChrysCapital/Intas

The Draft Order, combined with the nature of the imposed remedy, strongly indicates that the CCI has not followed a crude “worst-case scenario” approach, but has conducted its unilateral effects assessment taking the degree of financial and corporate ownership into account. Despite noting that ChrysCapital’s portfolio companies and Intas together serve over 30 percent of several markets, the Draft Order does not discuss unilateral effects in individual markets.  Instead, it expresses concern that “(t)he common interest of the Acquirers in Mankind and Intas would give them the ability to pursue anti-competitive goals such as allocation of product or geographic market, or customers; streamlining innovation efforts; price arrangements; and/or bid-rigging, in the concentrated markets.”  This is clearly a coordinated effects theory of harm. Moreover, the remedy involves foregoing the appointment of a board director in Mankind, Intas’ competitor, which would do little in and of itself to assuage any unilateral actions.

The CCI’s apparently sophisticated unilateral effects analysis, possibly using tools such as the MHHI, would be extremely positive news if replicated in the final order – it signals that the CCI recognizes the limitations of the “worst-case scenario” approach, and will conduct more sophisticated analyses should the need arise.

   

Assessing Coordinated Effects

Theories of Harm

There are two potential sources of concern related to coordinated effects when it comes to common ownership.  First, given a certain level of transparency, the transaction could change the market structure and underlying incentive structure to such an extent that that collusion becomes sustainable. Second, common owners can act as conduits for information exchange between portfolio companies.  While the first concern can be analyzed as in full mergers (while taking care to correctly estimate the increase in concentration), it is the second concern that has been the focus of discourse in the common ownership context, especially since even minority shareholders have access to information that outsiders are not privy to. This is potentially exacerbated if minority owners obtain access to the boardroom, where extremely sensitive information is discussed.  However, “the overall impact of common ownership on the likelihood of tacit collusion will depend significantly on whether an industry is already susceptible to collusion (based on factors such as homogenous products and multimarket interactions), and on the ability of an investor to encourage and facilitate a collusive agreement.16

Empirical Evidence on Coordination through Common Owners

The empirical evidence strongly suggests that increased transparency due to common ownership has almost no effect on the incidence of collusion. A recent study17 examining the association between interlocking directorates and collusion in Europe between 1976 and 2003 found that of 3,318 interlocking directors identified during this period, only 12 were involved in a known cartel, even though a total of 2,348 collusive ties were formed. This rate was even lower for the most recent period – only 1 in 570 during 1998-2003. Lastly, and most importantly, an analysis of case documents in these cases found that the interlocks were completely extraneous to the occurrence of a cartel, which were facilitated entirely through other means.

Coordinated Effects Assessment in ChrysCapital/Intas

While the CCI’s main theory of harm was centered on coordinated effects, and the remedy imposed was geared at breaking an interlocking directorate, the Draft Order contains no backing analysis. If this is also the position in the final order, then not only will the CCI have missed out on providing vital guidance on how it would assess coordinated effects in common ownership cases, it will also create the fear among investors that any creation of interlocking directorates will be dealt with severely. Given the empirical evidence cited above, this is unfortunate. While institutional investors contribute funds to fledgling businesses, they also transfer world class management and strategic expertise by inviting seasoned experts to help guide the firm.  Depriving Indian businesses of this expertise can only serve to make them less productive, and India a less attractive investment destination. Therefore, one hopes that the CCI addresses the coordinated effects theory of harm more fully in its final order.

 

Conclusion

Regardless of the form the final order takes, this case represents a paradigm shift in the CCI’s approach to transactions involving common ownership. Observers will take heart from the apparently sophisticated unilateral effects analysis that goes well beyond the “worst-case scenario” approach and possibly incorporates purpose-built tools such as the MHHI. However, given the coordinated effects-oriented theory of harm and remedy, more details in the final order on its coordinated effects analysis would further remove regulatory uncertainty and provide valuable guidance to future investors when negotiating deals with Indian companies.

Click here for a PDF version of the article


1 Saattvic is Founder and Expert at Sapient Econ (saattvic@sapient-econ.com). Sapient Econ advised ChrysCapital during the transaction analyzed in this article.

2 Combination Registration No. C-2020/04/741.

3 https://www.cci.gov.in/sites/default/files/press_release/PR09202021.pdf.

4 Available at https://www.mergerfilers.com/decisions/wSMCTwKzM97FZOOrder741%20Intas%20Order.pdf at the time of writing.

5 O’Brien, D. P. & Salop, S. C. (2000). Competitive Effects of Partial Ownership: Financial Interest and Corporate Control. Antitrust Law Journal, 67(3), 559-614.

6 European Commission. (2004). Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings. 2004/C 31/03, 5-18, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52004XC0205(02)&from=EN.

7 European Commission. (2013). ANNEX to the COMMISSION STAFF WORKING DOCUMENT Towards more effective EU merger control. SWD(2013) 239 final Part 2/3. https://ec.europa.eu/transparency/regdoc/rep/10102/2013/EN/SWD-2013-239-F1-EN-MAIN-PART-2.PDF.

8 Case M.1383.

9 Case M.1715.

10 Case M.2283.

11 Case M.6576.

12 Case M.6541.

13 Case M.7932.

14 Azar, J. (2017).  Portfolio diversification, market power and the theory of the firm; and Brito, D., Osorio, A., Ribeiro, R. & Vasconcelos, H. (2018). Unilateral Effects Screens for Partial Horizontal Acquisitions: The Generalized HHI and GUPPI. International Journal of Industrial Organization, 59, 127-189.

15 Banzhaf, J. F. (1965).  Weighted voting doesn’t work: a mathematical analysis.  Rutgers Law Review, 19, 317-343.

16 OECD (2017).  Common ownership by institutional investors and its impact on competition: Background Note by the Secretariat.  DAF/COMP(2017)10, https://one.oecd.org/document/DAF/COMP(2017)10/en/pdf.

17 Buch-Hansen, H. (2014).  Interlocking directorates and collusion: An empirical analysis.  International Sociology, 29(3), 249-267.