Ryan Hawthorne and Genna Robb, March 2019
A recent merger decision by the Competition Tribunal provides a useful clarification of the importance of economic evidence in an assessment of information exchange. Following a series of Competition Tribunal merger decisions which imposed limitations on cross-directorships post-merger due to coordination concerns , the Competition Commission recommended the imposition of a similar condition in a merger in the property sector. After consideration of the economic evidence led, the Competition Tribunal determined that cross-directorships were unlikely to lead to collusive effects in the market in question and approved the transaction without conditions. This reversed a trend of almost blanket application of restrictive conditions relating to cross-directorships in mergers, which often had no basis in the economics of information exchange and coordination.
The transaction in question was part of a series of transactions concerning the establishment of DiverCity, an urban property investment fund focused on investing and developing inner-city precincts and renewing dense urban precincts. The merger resulted in the acquisition of an 18% stake in DiverCity by Nedbank and by RMH Property Holdco 5 which is controlled by entities ultimately controlled by RMB Holdings Limited (collectively referred to as the RMH Group).
The RMH Group had various property investment holdings through stakes in Propertuity, Atterbury and Genesis Capital Three. Nedbank, together with its shareholder, Old Mutual group Holdings (SA), held various commercial, residential and industrial properties, including through a stake in Vestfund. Pre-merger, Propertuity, Atterbury, Genesis Capital Three and Vestfund were the shareholders in DiverCity, with the result that pre-merger, the RMH Group already had an indirect interest in DiverCity. The transaction would result in both Nedbank and RMH Property Holdco 5 acquiring a direct stake in DiverCity and gaining the ability to appoint directors to the board. DiverCity’s property portfolio included two small shopping centres and office and residential property in Johannesburg, Durban and Pretoria.
The Competition Commission found that the merger would not give rise to any unilateral effects in any of the markets in which the merging parties’ activities overlapped. However, it raised a concern with respect to the exchange of competitively sensitive information between the parties to the transaction. It therefore approved the transaction subject to conditions on cross directorships to prevent information exchange between DiverCity and Nedbank and RMH’s other property investments. The Commission did not, however, provide a theory of harm explaining how information exchange would be likely to lead to anti-competitive harm in this case, particularly given its conclusion that the merging parties’ market shares in the various markets were low and in the light of the large number of competitors in the South African property market.
From an economic perspective, the impact of information exchange on competition is ambiguous and depends on market structure and dynamics. In some circumstances, information exchange is pro-competitive, as recognised by the Competition Commission in its draft guidelines on information exchange, including where it leads to: “improvement of investment decisions; improvement of product positioning; provision of organisational learning; facilitation of entering an industry; lower search costs; benchmarking best practices; and more precise knowledge of market demand”.  On the other hand, information exchange can also facilitate coordination between competitors, where the market characteristics are conducive to such conduct.
Economic analysis of coordination shows that for a cartel to be sustained it must be in the best interests of each cartel member because each member of a cartel always has some incentive to cheat on the cartel. Each member of a cartel thus weighs (1) the profits from maintaining the cartel arrangement over a period of time against (2) the profits from cheating (deviating from the agreement by undercutting its competitors in order to win a greater share of the market) now and reduced profits in later periods when they are punished. In economics, these are known as the incentive constraints to cartel participation. If these incentive constraints are not met, cartel membership is not in the best interests of each firm concerned. The factors typically considered in an assessment of coordinated effects are linked to these incentive constraints and include the likelihood of reaching agreement, monitoring that agreement and enforcing the agreement. At the most basic level, where there are many outside, non-participants to a potential cartel as is the case in the property sector (and there are many property developers, listed and unlisted property funds, financiers, and other participants in the property sector outside of DiverCity and its shareholders) then coordination will be difficult to reach and sustain. Where products are homogenous and prices are transparent, it is easier for market participants to reach an agreement and monitor it. There is significant product differentiation in property markets, even within particular segments such as retail or commercial property. There are also a range of different types of market participants with different roles and objectives. These factors suggest that the incentives of market participants may not be easy to align.
In a merger transaction, a key question is whether the merger changes any of the market characteristics such that coordination is more likely and therefore a substantial lessening of competition. All of this requires an analysis of economic evidence on a case by case basis.
The Competition Tribunal and Competition Appeal Court debated the economic effects of joint ventures and coordinated effects at some length in the Primedia / NAIL transaction. The transaction resulted in Primedia, the owner of the Highveld radio station, having the right to appoint a director onto the board of NAIL, which partially owned a rival radio station, Kaya FM. After the Competition Appeal Court referred the matter back to the Competition Tribunal, the Competition Tribunal found that a range of factors mitigated against the possibility of coordination in this case. In particular, the Tribunal recognized that in a differentiated market, the different entities involved in a joint venture are likely to face divergent incentives in terms of setting prices which would make a coordinated agreement difficult to achieve and sustain. The Tribunal further explained that “In order to make the case for co-ordinated effects the evidence needed to be stronger than the mere holding of an interest in a rival and the right to appoint a director to its board.”
In a similar vein, according to the ICN merger guidelines, the correct manner in which to conduct an analysis of coordinated effects is first to consider whether the market has conditions which are conducive to coordination, and then to assess whether “the merger will make coordination easier or more likely, considering the specific features of the market that affect the merged firm’s ability and incentives to exercise market power in coordination with rivals." They explain that "Agencies should focus on whether the merger will materially alter firms’ ability or incentives to achieve and sustain coordination." Both the ICN and the EC merger guidelines correctly treat information exchange as one factor to be considered as part of a wider coordinated effects assessment, where the characteristics of the market and the ability of firms to reach, monitor and sustain a collusive agreement are to be interrogated in detail as well as the likely impact of the merger on each of these factors.
In the case of the DiverCity merger, the available evidence did not suggest that the market was one in which coordination was likely or that the merger made this substantially more likely. The Tribunal assessed “whether the Commission showed that the acquiring firms each have other property investments that are rivals of DiverCity to the extent that an exchange of confidential information might lead to a collusive outcome.” It concluded that the Commission had not provided evidence that this was the case and approved the merger without conditions. In particular, it found that the exchange of information was unlikely to have a collusive effect as the merging parties’ property assets were differentiated and catered for different market segments.
This case therefore provides a useful reminder that, contrary to some of the recent practice by the competition authorities, information exchange resulting from merger transactions needs to be assessed with regard to economic evidence and cannot be assumed to lead to a substantial lessening of competition absent consideration of the competitive dynamics of the relevant market.
Acacia Economists Ryan Hawthorne and Genna Robb provided expert economic advice to the merging parties in relation to the transaction.
 See for example Investec/Assupol, case number LM042May18; ARC/Ooba, case number LM257Mar16; Arrowhead/Synergy, case number LM076Aug16; and, Pembani/Shanduka, case number LM041Jun15.
 Case number LM083Jun18.
 See ‘Draft Guidelines on the Exchange of Information between Competitors under the Competition Act’, available here.
 See, for example: Motta, M. (2004). Competition policy: theory and practice. Cambridge University Press.
 See the Competition Appeal Court decision in case number 68/CAC/MAR/07 and Competition Tribunal decisions in case number 39/AM/MAY06.
 ‘ICN Recommended Practices for Merger Analysis’ p.23.
 ‘ICN Recommended Practices for Merger Analysis’ p.24; EC (2004) ‘Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings’ paragraphs 47 and 51.