Lauralyn Kaziboni and Reena Das Nair
Competition investigations on abuse of dominance by near-monopoly beer producers have typically been limited to national boundaries and within the jurisdiction of single national competition authorities. However, it is increasingly recognised that viewing transgressions as neatly falling within political borders is restrictive, and often misses the ‘bigger picture’ of the firm’s overall strategy and conduct. For instance, conduct characterised as abuse of dominance in one country may be part of a broader picture of that firm being ‘allocated’ that country or region as part of a collusive agreement, while fellow cartel members are allocated other countries or regions. The implicit (or explicit) understanding may be that the firms will not target each other’s allocated countries or regions. Therefore, addressing the abuse of dominance issues in a single country when such arrangements are in place is unlikely to solve the problem. This article highlights the importance of understanding behaviour and strategies of firms holistically, and across regions.1
In general, cartel conduct is more likely to occur in markets where there is a high concentration of firms, relatively homogenous products, high barriers to entry, stable demand conditions, firm symmetry, multi-market contact between firms, and cross ownership, among other facilitating factors, including sharing of disaggregated information.2 Furthermore, cartel developments across countries have in the past been facilitated by the lack of an active regional regulator to oversee cross border arrangements.
It is well accepted that cartel behavior is detrimental to consumer welfare. Cartel conduct stifles competition as new entrants face substantial barriers to entry due to the actions of entrenched incumbents.3 Lessened competition implies that dynamic rivalry is reduced, and consumers cannot benefit from variety, improved quality and lower prices.4 In light of these detrimental effects and the increasing appreciation of regional cartel conduct, the COMESA Competition Commission has commissioned a study on cross border cartels in Africa that seeks to determine the effects of such arrangements.5
Findings from previous studies, such as in the cement industry, are illustrative of the coordinated strategies large firms could have in industries where scale economies and large investments are important. Cement is often produced by multinational companies that develop strategies on a wider regional basis as opposed to a country-by-country basis. Companies have operated in different countries by agreement with one another, either through the export of cement to those markets or by establishing plants locally. During the cement cartel period in the Southern African Customs Union (SACU) for example, PPC supplied the Botswana cement market, while Namibia was supplied by Afrisam in accordance with the cartel agreement. The understanding was that each would not target the other’s territory. The companies were able to monitor the collusive agreement, inter alia, by sharing monthly sales information. The conduct resulted in higher prices and higher (economic) profit margins for the firms involved.6
Developments in the beer industry
Similar regional market allocation arrangements are present in the beer industry in Africa. Market allocation through cartel arrangements serves to maintain the dominant positions of incumbent firms in individual country markets, and often gives rise to incentives to abuse this position of market power. As put by the communications director of SABMiller (2009), the second largest beer producer globally, in reference to the relationship between Castel and SAB (discussed later):
“There may be antitrust laws at the national level, but none covering the continent. I don’t see what the problem is.”7
Such market allocation arrangements have led to competition concerns against beer producers in a number of countries in Africa. A recent example of this is the regional cartel that was uncovered and prosecuted by the Competition Commission of Mauritius (CCM). The arrangement between Stag Beverages Ltd, a subsidiary of Castel, and Phoenix Beverages, was the first cartel investigation in Mauritius. A financial penalty of MUR27 million8 (approximately USD900 000) was imposed wherein Phoenix received lower penalties as they took advantage of the leniency programme offered by the CCM, although behavioural remedies were also imposed. Stag and Phoenix had agreed to share the Mauritian and Malagasy beer markets in contravention of Section 41 of the Mauritian Competition Act. Under the agreement, Stag was to exit the Mauritian beer market, and dismantle their operations in Mauritius while continuing its operation in Madagascar, with the reciprocal exit of Phoenix from Madagascar.9
In other countries on the continent, SABMiller has maintained its position as the dominant beer producer, with brewing and beverage operations in Africa across 15 countries and new capacity investments in Zambia, Nigeria and Uganda. Furthermore the firm is represented in an additional 21 countries through a strategic alliance with the Castel Group (excluding South Africa and Namibia). Since 2001, Castel holds a 38% stake in SAB’s African subsidiary while SAB has a 20% stake in Castel’s African operations. According to SAB, the strategic alliance was undertaken to hedge against risk in the African political environment and to increase investment opportunities. Moreover the strategic alliance in SAB’s view aims to “capitalise on bulk purchasing opportunities; enhance SAB’s strategic position in African markets and benefit from future opportunities to increase cooperation between the SAB and Castel.” This strategic alliance further allows SAB a pre-emptive right to acquire Castel’s Africa operations should it wish to sell, and effectively means that the beer producers do not compete against one another in the allocated territories.10
Similar arrangements have been observed in other African countries. In 2009, East African Breweries, a subsidiary of Diageo,11 traded 20% of their shares of its local subsidiary, Kenya Breweries, for a similar proportion in Tanzania Breweries, a subsidiary of SAB International. This ultimately resulted in Kenya Breweries and Tanzania Breweries allocating the Kenya and Tanzania markets, respectively, as they both exited the other market akin to the SAB-Castel and the Phoenix/Stag arrangements.12 The agreement not only resulted in job losses, but led to the elimination of competition in both markets, and the end of a price war that had kept prices stagnant for five years.13 Dismantling these arrangements would most likely stimulate competition in the region.
Because competition authorities tend to assess the conduct of the dominant beer producer within geographic markets often defined along political borders, competition investigations historically have been focused on abuse of dominance conduct in each country and not broader regional arrangements that may have given rise to those monopolies. In 2009, Tanzania Breweries Limited (TBL), a subsidiary of SABMiller was accused by Serengeti Breweries Limited (SBL), a subsidiary of East African Breweries, of abusing its dominance by entering into branding agreements with outlet owners which excluded the participation of SBL and other smaller breweries.14 More recently in 2014, the Competition Tribunal of South Africa dismissed the distribution case between the Competition Commission and SABMiller and 14 SAB appointed distributors.15
Despite the ostensible rationale for cross shareholding agreements, these also present a risk to competition in that they eliminate any competition between large beer companies. These strategic alliances and extensive cross ownership agreements have not been interrogated by national or regional competition bodies, and the cases discussed suggest a need to do so. These cases highlight that the analysis in certain competition matters should gravitate from analysis of abuse of dominance in a single country, towards analysis of possible regional cartel activity as the COMESA Competition Commission hopes to do in future.
Please click here for a pdf version of the article.
- Recent studies show that this applies across several different industries, including in cement and sugar. See, for example, Paelo, A. ‘Regional dimensions of competition and trade in sugar and cement’. CCRED Quarterly Competition Review (August 2014).
- Bernheim, B. D. & Whinston, M. D. (1990). Multimarket contact and collusive behaviour. The Rand Journal of Economics, 1-26.
- Roberts, S. (2012). ‘National Dominant Firms, Competition Law and Implications for Economic Development in Southern-Africa: case study of energy, beer and food’. Instituto de Estudos Sociais e Económicos Conference, Conference Paper 17.
- Levenstein, M., & Suslow, V. Y. (2003). Contemporary international cartels and developing countries: economic effects and implications for competition policy. Antitrust LJ, 71, 801.
- Nkambule, N. ‘Puma Energy mergers with Excel Swaziland’ (7 October 2014). Swazi Observer.
- See note 1.
- See note 3.
- A financial penalty of MUR 20, 299, 355 was imposed on Phoenix, while MUR 6, 575, 377 was imposed on Stag.
- See Competition Commission of Mauritius. ‘Commissioners Endorse the Recommendations of the Executive Director and Direct Phoenix Beverages Ltd and Stag Beverages Ltd to Pay Financial Penalties of MUR27 Million for Collusive Behaviour’ (22 August 2014).
- South African Breweries. Annual Report 2001, p. 18.
- East African Breweries is Eastern Africa’s largest brewer and is in fact a subsidiary of Diageo, one of the largest brewers in the world.
- Wahome, M. ‘East Africa: Brewers Move to End Turf Wars’ (15 May 2002). All Africa.
- Jenny, F. ‘Competition Enforcement in Testing Times: beyond the national level’. Presented at Competition Principles Under Threat, IDRC Pre-ICN Forum on Competition and Development, Zurich, Switzerland, 2 June 2009.
- Judgement by the Fair Competition Commission of Tanzania. May 2010. Serengeti Breweries Limited vs Tanzania Breweries Limited. Complaint No. 2 of 2009.
- Kaziboni, L. ‘SA Tribunal finds no case against SAB distribution’. CCRED Quarterly Competition Review (August 2014).