---

IN THE COMPETITION TRIBUNAL

REPUBLIC OF SOUTH AFRICA

Case No: 19/LM/Feb00

In the large merger between:

BROMOR FOODS (PTY) LTD

AND

NATIONAL BRANDS LTD

Tribunal’s Reasons for Decision

Conditional Approval

1. On 31 August 1999 Bromor Foods acquired the Game powder and Sports Drink business from National Brands as a going concern. We have decided to approve the transaction subject to certain conditions, which we have set out below.

The Merger Transaction

2. National Brands sold its sports drink business known as Game to Bromor Foods in August 1999. The business entails the manufacture, bottling and canning, marketing, distribution and sale of a product known as Game. The product known in the beverage industry as a sports drink is produced in both a powder and ready-to-drink (RTD) form. The ready-to-drink product is distributed in 500 ml plastic bottles.

3. The business was purchased as follows:

The intellectual property rights to the Game brand - R4 million

The recipes and formulas - - R 1

The canning, bottling equipment, plant and fridges - R 2 020 139

Stock in trade - (figures not available to us)

4. Since the merger was implemented on the day immediately prior to the date on

which the Competition Act came into effect 1 September 1999 it is considered in terms of the transitional measures found in Schedule 3 of the Act. This has various implications for what we say later.

Background

5. National Brands, a subsidiary of Anglovaal Industries, is in the business of manufacturing and marketing consumer brands. In February 1999 the NB board took a decision to dispose of the Game business, which was considered to be non-core. A willing buyer was found in Bromor Foods a wholly owned subsidiary of Cadbury Schweppes. Bromor was already a player in the Sports Drink market through its range of ready to drink and concentrated beverages marketed under the Energade brand name. The merger was notified to the Competition Commission in November 1999. The Commission recommended that the merger be approved unconditionally.

6. Presently there are three major brands marketed specifically as RTD sports drinks. In descending order of market share they are: Bromor’s Energade, Coca Cola’s PowerAde and Game, the subject of this merger.

7. Game is the oldest brand in the South African market having been introduced in the 1970’s. Game unlike its rival brands is also available in a powdered form. Certain beverages are marketed as energy as opposed to sports drinks. The major energy drink brand is Red Bull.

Evaluation of the Merger

The Relevant Market

8. Defining a relevant market for consumer products is notoriously difficult. Delineating a relevant market for beverage products is especially difficult because one is faced with not only the subjective proclivities of consumers but also the marketing stratagems of firms as they attempt to differentiate their products in response to competitive threats.

9. Beverage antitrust cases have long been the subject of bitter contestation over relevant market definition. On the one hand merging parties contend they are merely minor players fighting for their “share of the throat”, in a market where the fizzy drink competes with anything that can be imbibed from fruit juices to tea. On the other hand competition regulators argue that the fizzy drink is the relevant product market.¹

10. Ultimately each case must be determined on its own facts and foreign judgments can do no more than give us guidelines to method for they cannot serve as a way for us to come to a conclusion on facts. The behavior of a teenage consumer of carbonated beverages in Texas is no more use to us as evidence than the behavior of the French consumer of carbonated mineral water.

11. The Commission in this case is in the highly unusual position of a competition regulator arguing for a wider market definition than the merging parties. The Commission says the relevant market is the market for non-alcoholic beverages. On this basis the market share of the sports and energy drinks is a mere 2,6 % and of which the merged entity would comprise a modest 1%. Not surprisingly, on this definition of the market they conclude that the merger raises no concerns.²

12. The merging firms did not share the Commission’s expansive view of the market. Whilst we cannot say they contradicted the Commission on this point, in argument before us and in their documentation they did not pursue the point with any conviction choosing instead to focus on barriers to entry.

13. However internal documents supplied by both parties to the Tribunal subsequent to the Commission’s report point to a much narrower definition of the market - that for RTD sports drinks only. As we argue below the behavior of both merging firms suggests that this is how they understood the relevant market.

14. We say this for the following reasons-

14.1 In its business plan Bromor identifies sports drinks as a category. This is not merely labeling, as the company is also involved in other non-alcoholic beverages and significantly does not deal with them in the same category. The only competing products named are other sports drinks and the only competing brand specifically named is PowerAde.

14.2 National Brands in its Business Plan for the Game Sports Drink dated July 1998 and six months prior to the merger having been mooted, considered an ambitious plan to re-launch the Game RTD and powder products. The document is premised on the assumption that the market is a sports drinks market and specifically identifies Game’s competitors as Energade, PowerAde and Lucozade. In a market survey conducted for them by Markinor in January 1998 only the three leading brands, i.e. Game, PowerAde and Energade are compared.

14.3 The Sale Agreement between the parties contains a restraint of trade prohibiting National Brands from carrying on a competing activity for a period of five years. “Competing activity” is defined as an activity in the “ the sports energy drinks category”.³

14.4 A market survey performed by Nielson compares total annualized volume yields on “Sports Drinks”.⁴

14.5 The prices of the sports drinks differ considerably from that of the nearest possible beverage substitute namely energy drinks. Although all the prices of these products fluctuate depending on the outlet at which they are sold Red Bull is consistently and significantly more expensive than the RTD products.

14.6 Post merger Bromor was able to sustain an 8 % price increase on its Energade product and an 8.7% increase of its newly acquired Game brand including bringing price parity between the two brands.

14.7 The rationale for the merger from Bromor's point of view was anti-competitive and the intention was to prevent the Game Brand from becoming available to either a new entrant or to an existing competitor viz. PowerAde. Mr. Cowie the marketing director of Bromor informed the Tribunal that the Cadbury –Schweppes board was concerned that a “ formidable competitor” like Coca Cola could easily have bought the brand as well.

14.8 There is a lack of price competition in the sports drink market, which facilitates the ability of Bromor as the leading firm on its own or Bromor and Coca Cola jointly to raise prices to a supra-competitive level. This is evident from the Game business plan referred to above where in paragraph 4.4 of the pricing plan the authors note that the RTD market is “not a price sensitive market” This observation is made prior to the merger being contemplated. Mr. Cowie in his submissions refers to the fact that what distinguishes brands is whether they are actively marketed or not “otherwise you land up competing on price.” The parties have also indicated that prices are not uniform and depend on the nature of the outlet where the product is sold. The consumer purchasing from a refrigerator at a “point of sweat” will pay considerably more than the consumer purchasing off the shelf at the super market.

15. None of these factors on their own is decisive as to the relevant market but the accumulation of these factors suggests that the relevant market is for RTD sports drinks. On this basis and adjusting the Commission’s figures to exclude Red Bull the market pre and post merger is highly concentrated.

Product	Pre-merger Market Share (%)	Pre-merger Concentration (HHI)	Post-merger Market Share (%)	Post-merger Concentration (HHI)
Energade	55.1	3036	60 (incl. Game)	3600
Powerade	30.5	930	30.5	930
Game (RTD)	4.4	19
Lucozade	1.0	1	1.0	1
Other	8.9	9 (a)	8.9	9
Total		3995		4540

Source: Based on figure in Competition Commission’s Report

We have assumed that 50% of the Game product is sold in RTD form and that the figures quoted for the other products are for RTD sales.

Market share percentages and Herfindah- Hirschman Index (HHI) values have been rounded-off.

(a) The HHI calculation assumes that ‘other’ comprises 9 firms of equal share ( 1%).

15. As far as the geographic market is concerned there is no dispute that this is the whole of South Africa.

Barriers to Entry

15. As we stated earlier the merging parties have focused their defence of the merger on the basis that barriers to entry in the industry are low because there are no significant regulatory hurdles and capital expenditure on plant is not significant. This they say is evidenced by the proliferation of small brands in the market.

17. If we exclude the four largest brands the remaining brands account for only 9 % of the sports drinks market. Given that the fourth largest Lucozade only has a share of 1% we assume that none of the present remaining brands has a market share significant enough to discipline the behavior of PowerAde and Energade / Game in the market.

18. This leaves us to consider the role of potential competition as a deterrent effect. Whilst the parties are correct in contending that capital is not a serious deterrent to entry in this market, establishing a brand is. Since brands are essential in the market for fast moving consumer goods, no firm will enter unless they are willing to sink significant sunk costs in marketing a brand that can compete successfully with the two market leaders. By sunk costs are meant costs that cannot be recovered if the entry is a failure. Advertising to create a brand image is a classic example of a sunk cost.⁵

19. In one of the best-known merger cases in the history of the beverage industry the FTC successfully challenged Coca Cola’s attempts to acquire rival carbonated soft drink producer Dr Pepper. In 1986 a Federal District Court granted a preliminary injunction that thwarted the merger. In delivering the judgment Judge Gessel made the following observations about sunk costs that we find instructive,

20. “ to establish a major new brand requires large expenditures for advertising to fix the brand name and image in the mind of the consumer – expenditures that cannot be recovered if the introduction fails. … Effective entrants must also match the considerable promotional budgets of the dominant companies in targeting their brands for effective distribution through retailers … Finally, it has been the experience of the industry that effective entry against dominant companies is likely to require years of sustained effort for any continuing success.”⁶

21. Shepherd argues that advertising costs for a new entrant are higher than those for existing firms with established brands. This is because to enter a firm must meet penetration costs, which escalate sharply with as a firm tries to increase output.⁷

22. The marketing information submitted by both parties is consistent with the view outlined above. Even Energade the leading brand was concerned that PowerAde, with Coca Cola’s resources behind it, might win the battle to obtain celebrity sports endorsements. Indeed the very demise of Game is illustrative of what can happen to an established brand, which fails to sustain its marketing edge over its rivals.

Rationale for the merger

23. Even if the merger may lead to a substantial lessening of competition we are obliged to consider whether it has any pro-competitive effects. At the hearing we explored this issue with the parties and they adduced no satisfactory evidence on this point. On the contrary, the representative of Bromor indicated that when National Brands put Game on the market there was concern that the brand might be purchased by a competitor, more specifically Coca Cola. The decline in the Game brand since February 1999 suggests that the purchasers were less concerned with reinvigorating a past champion brand and more with keeping it away from competitors.

24. A document placed before the Board of Cadbury Schweppes indicates their intention to utilise Game as a “ fighting brand against regional competitors in the sports drinks market.”⁸

25. We are further concerned that the real intention of Bromor was to remove the Game RTD brand from the market. In their letter to the Tribunal dated 9th March 2000 Bromor observe that the market for the Game liquid has all but collapsed. The thrust of Mr Cowie’s oral submissions to the Tribunal was to the same effect. Consumer indifference to the brand is cited as the reason for this. Yet in March 1999 total Game sales were at a peak for the 12 month period from July 1998 to June 1999. The decline in the brand was conceded by Bromor, who said they had no interest in maintaining it once it was going to be sold. Although the agreement was only signed in August the agreement to sell appears to have been finalized in February or March. Had Bromor seriously intended to retain a vital brand it would have taken the normal steps to ensure the brand remained viable during the period between the conclusion of negotiations and the effective date. The extraordinary decline of the brand in this period suggests Game RTD’s demise rather than its continued vigour may have been their real purpose.

Sale of the plant and powdered IP rights

26. The sale of the plant raises no competitive concerns as market power in this market derives from control over brands in the RTD sports drinks market not ownership of plant. Secondly we also conclude that the powdered drinks constitute a separate market from the RTD. The powder product is not ready to drink and hence attracts a different consumer. It pricing and marketing are different as well and it also is less constrained in methods of distribution i.e. not requiring fridge space. Since Bromor was not in this market before the merger it raises no concerns.

1. Conclusion

27. We find that the merger substantially prevents and lessens competition in the market for sports drinks because:

1. The merger will lead to a higher levels of concentration in an already concentrated market;

2. There is an increased likelihood of collusion between the two remaining brands;

3. The Game brand has been removed as the most likely effective competitor to the two dominant brands;

4. There are no pro-competitive efficiencies or public interest considerations which otherwise justify the merger.

2. Remedy

28. Since this is a Schedule 3 merger it means that the parties were lawfully entitled to implement the merger without prior approval from the Tribunal. At the time of this decision the merger will have been in effect for over seven months.

29. Given our conclusion that the merger is anticompetitive we have three possible remedies-

1. to prohibit the merger

2. to approve the merger subject to an appropriate structural remedy

3. to approve the merger subject to a behavioral remedy.

30. Prohibiting the merger is too drastic a remedy given that-

1. the merger has already been implemented and the Game brand has since weakened;

2. not all aspects of the merger are anti-competitive. There are for instance no concerns about the sale of the plant or the sale of the powdered Game product;

3. Separating the powdered brand from the RTD is difficult since they are both marketed under the same brand name.

31. The next option would be to approve the merger but make it subject to an appropriate structural remedy. The most obvious structural remedy on the facts of this case would be to order Bromor to divest itself of the Game RTD brand’s intellectual property⁹ to a third party acceptable to the Tribunal. This remedy as well is too drastic in our view.

32. If we had to order divestiture an appropriate time period would have to be given for Bromor to sell the Game intellectual property to a third party. A period of at least six months would be appropriate. This means that at least thirteen months would have elapsed since the sale of Game to Bromor and nearly twenty months since the brand was last viable. There is no guarantee that at that stage the Game brand would be significantly more useful to a new incumbent than establishing a new brand given that there would be no incentive for Bromor to retain the brand for the benefit of a future vigorous competitor even at the expense of a short-term loss on the price achieved through a sale.

33. Normally the practice in the circumstances would be for the competition authority to appoint a trustee to administer the assets to be divested so as to retain their value prior to a sale. Given that in this case we are concerned with divesting a brand and not a separate business this remedy is completely impractical let alone not worth the expense.

34. Secondly the Game brand also attaches to the powder form product, which, as we have noted, is in a separate market and does not raise competition concerns. Practically separating the brand is impossible and means a divestment remedy would have to include the powdered product. Thirdly the rights to Game include rights to the brand in other non- South African markets. We have no jurisdiction to assess competitive effects beyond our borders in terms of section 3 of the Act.

35. Divestiture is too extreme a remedy in the circumstances nor is there any expectation that it will be an effective one either. It is not surprising that other jurisdictions are loath to impose divestiture as a remedy where a merger has already been implemented, hence the rationale for pre-merger notification. The Federal Trade Commission noted in their 1999 study on their divestiture process that divestiture after consummation is frequently inadequate. One example they give is of this is that the goodwill of the acquired firm may be dissipated making it a weaker competitive force after divestiture.¹⁰

36. One of the primary changes in the merger regime between the present act and its predecessor, the Maintenance and Promotion of Competition Act (Act 96 of 1979), was the introduction of compulsory pre-merger notification. If a structural remedy is considered appropriate it is less offensive to the merging parties settled rights if imposed prior to consummation because one does not have to unscramble a merged entity.

37. We conclude that the only appropriate remedy is a behavioral one. Although the remedy we are proposing is mild and may be entirely academic it is based on the recognition of the disciplining effect of potential competition. We find the restraint of trade imposed on National Brands unnecessarily restrictive and not related to the goodwill of the Game brand as they could only enter with a new brand without violating the intellectual property rights to the Game brand owned by Bromor. Since National Brands with its experience of the consumer market is always a viable potential entrant its potential as an entrant may discipline the remaining players more than any other potential competitor.

38. Secondly given the fact that we find the merger anti-competitive the potential for future restrictive practices to take place in this market is by no means remote. These could take the form of either an abuse of a dominant position or a horizontal restrictive practice between the major players. A possible remedy if this does occur and no adequate behavioral remedies are available is for the Commission or complainant to seek an order of divestiture against Bromor. This potential remedy which may have a disciplining effect on the firms in the market will only be feasible if the Game brand retains some value so that divestiture to a third party is more attractive than establishing a new brand. For this reason we have made the additional order that Bromor must maintain the Game RTD brand for a period of at least two years. If a successful restrictive practice case is brought against Bromor in this period and divestiture is a competent remedy the retention of the Game brand will also ensure that it is a practical remedy. We have been at pains not to be prescriptive in this regard as we are sensitive about interfering with Bromor’ s commercial freedom unduly.

Order

39. We approve the merger subject to the following conditions-

39.1 that clause 12 of the sale agreement dated 31 August 1999 is declared void with effect from the date of this order;

39.2 that Bromor continue to maintain the Game ready to drink brand in the national market at levels not substantially less than it does currently including expenditure on advertising or otherwise promoting the brand for a period of not less than two years from the date of this order;

39.3 that the obligation in sub-paragraph 2 does not prevent Bromor from selling the Game intellectual property or parts of it to a third party during this period provided that-

1. third party is-

   1. not Cadbury-Schweppes , Coca Cola or a firm controlled by either of them or Bromor ; and

   2. genuinely at arms length form any firm contemplated in sub-paragraph (i) ; and

2. the transaction is notified to the Competition Commission prior to implementation.

________________ Date: 14 April 2000

N.M. Manoim

D. H Lewis and S. Zilwa concurred.

---