Pick 'n Pay Retailers (Pty) Ltd and Boxer Holdings (Pty) Ltd (52/LM/Jul02) [2002] ZACT 30 (26 April 2002)

COMPETITION TRIBUNAL

REPUBLIC OF SOUTH

AFRICA

Case No: 52/LM/Jul02

In the large merger between:

Pick ‘n Pay Retailers (Pty) Ltd

and

Boxer Holdings (Pty) Ltd

_______________

Reasons

Approval

The Competition Tribunal issued a Merger Clearance certificate on 7 August 2002 approving the merger between Pick ‘n Pay Retailers (Pty) Ltd and Boxer Holdings (Pty) Ltd. The reasons for our decision are set out below.

The Transaction

Structure

Pick n Pay Retailers will acquire all the issued share capital in Boxer Holdings and Boxer Superstores. Both Boxer Holdings and Boxer Superstores will after the merger become wholly owned subsidiaries of Pick ‘n Pay Retailers.

The parties

The primary acquiring firm is Pick ‘n Pay Retailers (Pty) Ltd which is controlled by Pick ‘n Pay Stores Limited. The Pick n Pay group owns 14 Pick ‘n Pay Hypermarkets, 113 Pick ‘n Pay Supermarkets, 1000 franchised Pick ‘n Pay Family Stores, 39 franchised Pick ‘n Pay Minimarkets and 116 Score Supermarkets.

The primary target firm is Boxer Holdings (Pty) Ltd and Boxer Superstores (Pty) Ltd. Dumakude Investments (Pty) Ltd, I O E Holdings (Pty) Ltd, Ndumu Investments (Pty) Ltd and Smithhold (Pty) Ltd each hold 25% of the shares in Boxer Holdings. Boxer Holdings directly controls Boxer Superstores through its majority shareholding of 78.09% in the issued share capital of Boxer Superstores.

Both, Pick ‘n Pay Retailers and Boxer Superstores, are active in the retailing sector, selling groceries and a range of other household products.

Rationale for the transaction

According to the merging parties the merger will increase Boxer’s competitiveness in the retail market since Shoprite/Checkers, Boxer’s main competitor, is the only group that has national coverage in the LSM1 group in which Boxer operates. At the same time this merger will assist Pick ‘n Pay to enter a market sector in which it is not currently represented.

Pick ‘n Pay has advised that it intends to continue with the Boxer brand and will not remove it from the market.

Evaluating the merger

The relevant market

Both Pick ‘n Pay and Boxer serve clients with a full range of supermarket products, which includes bakeries, butcheries and delis. Since Boxer does not sell products such as hardware, clothing and household appliances these products are not included in the relevant market.

We are convinced by the Commission’s view that the profile and target market of the different format stores are important factors that indicate in which market the merged company will compete. We therefore agree that the LSM classification should be used in this case to define the product market.

Boxer concentrates on the LSM 1-4 categories focusing on “no-frills”, low cost fixtures and fittings with stores situated in rural areas. The only grocery outlets within the Pick ’n Pay group that do not trade in the LSM 6-8 market are the Score Supermarkets. They too focus on the LSM 1-5 categories although the concept, in terms of how each store looks and trades, differs. Boxer is set up much like a warehouse type operation selling in bulk to the lower LSM groups selling a lot of 25 – 50kg bags of mealie meal, rice and other basic commodities. Score supermarkets, on the other hand, tend to have a lower cost or spend per consumer with customers shopping more regularly. It has a slightly more up-market vision that tends to attract customers from the middle LSM group.2

We also agree with the Commission that the geographic market is local. The relevant market accordingly is the retail grocery market, serving consumers in the LSM 1-5 categories, within a local geographic market, being the area immediately surrounding the stores of the parties.

The towns in which both parties compete are Vryheid in KwaZulu Nata, Rustenburg in North West, and Umtata, Lusikisiki, Mount Frere, Bizana, Engcobo, Idutywa, King Williams Town and Butterworth in Eastern Cape.

Effect on competition

In this relevant market Spar and Shoprite/Checkers are regarded as the main competitors since they cater for all the LSM categories. The Commission also presented evidence that wholesalers/ cash & carry outlets also compete with the outlets of the parties due to families or villages in and around town that buy in bulk from the wholesalers for own consumption and not for resale.

Barriers to entry are not high with large groups such as Shoprite and Spar continually opening new stores, such as Shoprite, which opened in Lusikisiki in May 2002.

The market shares of the merged entity and its main competitors in each town are:

TOWN

OUTLET

ESTIMATED MARKET

SHARE Vryheid Merged entity Shoprite Metro Cash & Carry Checkers Spar 13.47 16.33 16.33 16.33 8.98 Burgersfort Merged entity Spar Lebowa Wholesalers Metro Cash & Carry 14.36 11.88 14.85 14.85 Rustenburg Merged entity Shoprite Checkers Trans Cash & Carry 21.30 12.03 15.04 10.03 Lusikisiki Merged entity Shoprite Metro Cash & Carry Browns Cash & Carry Lusiki Cash & Carry 23.05 9.38 14.06 12.50 15.63 Mt Frere Merged entity Soli’s Spar TG Wholesalers Weirs Cash & Carry 22.73 12.63 18.18 15.15 Bizana Merged entity Ingele Supermarket Ingele Wholesalers Browns Bargain Wholesalers 14.20 14.79 14.79 17.75 28.40 Idutywa Merged Entity Empumalanga Metro Cash & Carry Spar Weirs Cash & Carry 13.80 25.10 12.55 5.02 7.53 Butterworth Merged Entity Spargs Emphumalanga Weirs Cash & Carry Spar 17.95 23.08 25.64 16.03 9.62

Market shares fluctuate between 13% and 23% in the markets where there are overlaps between Boxer and Score stores. According to the parties calculations only the post merger HHI’s in Bizana (1820 points) and Butterworth (1920 points) will exceed 1800 points, which the USA antitrust agencies regard as highly concentrated. However, Boxer and Score operate in markets that are highly price sensitive and evidence was put before us, which shows that if prices are increased sales drop.3 We are therefore convinced that the merged entity will not be able to gain market power in these areas where there are other competitors.

Taking into account the number of competitors and their market shares, as set out in the above table, as well as the slight difference in market focus between Boxer and Score the Tribunal is satisfied that post the merger competition will not be substantially prevented or lessened in any of the relevant markets.

Public Interest

The transaction does not raise any adverse public interest issues.

_____ 26 August 2002

D.H. Lewis Date

Concurring: N. Manoim, U. Bhoola

For the merging parties: Mr R Wilson of Sonnenberg Hoffmann Galombik

1 Living Standard Measurement (LSM) segmentation divides the population into 8 LSM groups, from 1 (the lowest) to 8 (the highest), using criteria such as degree of urbanisation and ownership of cars and major appliances.

2 The average amount spent per shopping basket in Boxer stores fluctuates between R25 – R30 and the average Score basket R12.

3 See page 123 of the record where examples of price behaviour is discussed and the example of Boxer in Umtata that sells a lot of cigarettes to hawkers increased its price of cigarettes and sales dropped..