EC, November 15, 2004, No M.3464

Dear Sirs,

Subject:  Case No COMP/M.3464 - Kesko/ICA/JV

Notification of 8.10.2004 pursuant to Article 4 of Council Regulation (EC) No 139/2004

1. On 08.10.2004, the Commission received a notification of a proposed concentration pursuant to Article 4 of Council Regulation (EC) No 139/20042 (“ECMR) by which the undertakings Kesko Food Ltd. (“Kesko”, Finland) and ICA Baltic AB (“ICA”, Sweden) combine their retailing activities in the Baltic States into a newly created company constituting a joint venture (“the JV”) that they will jointly control within the meaning of Article 3(l)(b) of the ECMR.

1. THE PARTIES

2. Kesko is mainly active in groceries trade. It is a wholly-owned subsidiary of the Kesko Group, which is also active in other areas of trading (such as agricultural and machinery trade), but food trade is its largest business division. In Estonia and Latvia Kesko is active in the retailing of daily consumer goods, whereas it does not operate as a wholesaler. It is not active on any market in Lithuania.

3. ICA is a subsidiary of ICA AB, jointly controlled by ICA Forbundet Invest AB (Sweden),

Canica (Norway) and Royal Ahold (NL). The ICA Group of companies is one of the Nordic region's leading food retail companies, with 3.000 of its own and associated stores in Scandinavia and the Baltic countries. ICA is active in wholesaling and retailing of daily consumer goods mainly on the Swedish and Norwegian markets. Through a number of subsidiaries, ICA is also active in retail of daily consumer goods in Estonia, Latvia and Lithuania. However, ICA does not operate as a wholesaler in these countries.

II. THE OPERATION

4. Pursuant to the proposed transaction Kesko and ICA will combine their respective businesses in Estonia, Latvia and Lithuania within procurement and retail of daily consumer goods by setting up a joint venture. According to the parties the JV will operate through its own store chains in the following shop formats: hypermarkets, food- supermarkets and hard discount chains. Kesko and ICA will transfer to the JV the shares of their respective companies operating in Estonia, Latvia and Lithuania in these businesses. The JV will be established as a limited liability company to be incorporated under the laws of Sweden with is registered office also in Sweden.

III. THE CONCENTRATION

5. Kesko and ICA will exercise joint control over the JV, since they will each acquire 50% interest in the JV, be entitled to the same number of representatives in the Board and have veto rights over all major decision.

6. The JV will operate on a lasting basis and not be dependent on its parents for supply, demand or resources. It will also act in its own name and for its own account and have a dedicated management. Thus it will be a full-function joint venture within the meaning of Art. 3 (4) of the ECMR.

IV. COMMUNITY DIMENSION

7. The undertakings concerned have a combined aggregate world-wide turnover of more than EUR 5 billion (ICA EUR 7893 million; Kesko EUR 7070 million). Each of ICA and Kesko have a Community-wide turnover in excess of EUR 250 million (ICA EUR [...]; Kesko EUR [...]), but they do not achieve more than two-thirds of their aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension. It does not constitute a co-operation case under the LEA Agreement.

V. RELEVANT PRODUCT AND GEOGRAPHIC MARKETS

8. Both Kesko and ICA are active in the retail of daily consumer goods in Estonia and Latvia. The operation therefore leads to horizontal overlaps in Estonia and Latvia. Their activities do not overlap in Lithuania. In addition, the parties are not active in the wholesale of daily consumer goods in any of the Baltic Countries.

9. Since the market investigation did not show significant differences with regard to the definition of product markets and geographic markets in Estonia and Latvia, the following assessment applies to both countries.

A. THE RET ATT, MARKET FOR DAILY CONSUMER GOODS 1. Product market

10. The parties submit that the relevant product market should include all sources of supply of daily consumer goods, i.e. hypermarkets, supermarkets, self-service stores, discount outlets, specialised shops, neighbourhood stores, kiosks, petrol station shops, open and closed markets (particularly important in the Baltic Countries), i.e. all retail sales which fall under the national statistics of total market with food predominating. This would be due to the special characteristics of the market structure in the Baltic Countries with a high number of open markets (functional all year around), small neighbourhood stores and specialised outlets.

11. The Commission has, however, traditionally considered that there is a separate product market in the retail of daily consumer goods mainly carried out by retail outlets such as supermarkets, hypermarkets and discount chains. Previous decisions argue that these retailers are active in different markets than other retailers such as specialised outlets, service stations and others.

12. According to the Commission market investigation, it seems that the range (food and non-food products) and variety (branded products, non-branded products, specialities) of products in these traditional outlets would be narrower than in hypermarkets and supermarkets. With regard to open markets, some products which are typically sold in hypermarkets or supermarkets, are not sold at all or only partly at the open market.

13. Furthermore, there is a clear trend in the Baltic States towards an increase in shopping in supermarkets and hypermarkets. Nowadays the Baltic consumers value more and more the broad selection in supermarkets and hypermarkets and the fact, that they can get all their products under the same roof, which saves them time as well. It must be noted, however, that the consumption pattern remains different from the one found in other EEA countries with the average purchase per visit at a smaller level and more frequent visits.

14. It is, however, not necessary to decide on the exact delineation of the product market, since the operation does not give rise to competition problems under any product market definition.

2. Geographic market

15. In previous Commission cases the geographic market for the retail sale of daily consumer goods has been delineated, according to demand side arguments, by the boundaries of a territory where the outlets can be reached easily by consumers (radius of approximately 10 to 30 minutes driving time). This territory may also be larger (e.g. regional or national) if different local areas are connected in such a way that they result in overlapping circles. The delineation of each local area can only be undertaken on a case by case basis by taking into account specific local circumstances.

16. The parties submit, that both Estonia and Latvia have relatively small territories and that the major players on the market for retail of daily consumer goods in the Baltic Countries operate according to concepts based on national chain of stores (prices, quality, service level, advertising and promotions are, to a large extent decided at national level). Accordingly, Parties contend that markets are national.

17. The market investigation confirmed that most grocery retailers (hypermarkets, supermarkets, discounters) are active on a national level and make decisions on product range, assortment and advertising on a national level. On the other hand differences were mentioned with regard to price levels within Estonia and Latvia.

18. However, since the transaction does not give rise to competition problems under any market definition, the exact geographic scope of the market can be left open.

B. THE PROCUREMENT MARKET FOR DAILY CONSUMER GOODS 1. Product market

19. The procurement market involves the sale of daily consumer goods by producers of such goods for example to retail chains, restaurants, specialised shops and catering companies. In previous Commission decisions the question has been left open whether the product market should be further delineated according to the above distribution channels. In addition, the Commission has found in previous cases that each product group constitutes an individual market as it is not, or only imperfectly, substitutable for other products or product groups.

20. The parties submit, however, that the market for procurement of daily consumer goods constitutes one single product market since the pattern of demand for each product group is broadly similar in the Baltic Countries.

21. On the basis of the market investigation it seems that a delimitation of the procurement market according to product categories would be appropriate. Most suppliers and competitors confirmed that product categories should be formed according to the previous case law in the following way: Consumer goods:(l) Liquids (including soft drinks, beer and cider, water, still wine, sparkling wine and spirits) (2) hardware store products (including cleaning products, sponges, toilet paper and kitchen rolls, deodorants, insecticides, household soaps, washing products, water softeners, washing-up liquids, dishwasher products, all-purpose cleaning products/detergents) (3) perfumes/hygiene products (paper or cotton hygiene/sanitary products, perfumes, health-care products) (4) dry groceries (including breakfast products, sweet biscuits, cakes, confectionery, desserts, cake-making ingredients, baby foods, soups, pasta, starch-containing foodstuffs, preserves and prepared foods, condiments and seasonings, aperitif products, pet foods)(5) parapharmaceutical products (6) perishable goods sold on a self-service basis (dairy products and eggs, poultry, frozen foods, ice cream, factory-made pastries and bread). Traditional fresh products: (7) delicatessen (prepared meat products and seafood) (8) fish (9) fruit and vegetables (10) fresh bread and cakes (11) meat. Miscellaneous goods: (12) do-it-yourself (13) home decorating (14) culture (15) toys/leisure/relaxation (16) gardening (17) motor car accessories. Electrical goods/photography/home movies/audio: (18) large domestic electrical appliances (19) small domestic electrical appliances (20) photography/movie-making (21) hi-fi/audio (22) TV/video. Textiles (23)

22. According to the results of the market investigation it would not seem possible for producers of one product category to switch easily into making other products. Hence the flexibility of production seems to be rather low. Furthermore, some suppliers seem to sell their products differently according to the channel of distribution.

23. However, it is not necessary to decide on the exact market definition of the procurement market for daily consumer goods, since the operation does not give rise to competition concerns under any product market definition.

2. Geographic market

24. The parties suggest that the geographic market would most likely be EU-wide, due to the fact that they procure most of the products either from local suppliers or from local importers of multinational companies. The Parties however admit, that imported products still needed to be adjusted to the requirements of the respective country of retail e.g. in terms of packaging and language.

25. According to the results of the market investigation, suppliers would seem to negotiate on a national level, sometimes even on a regional or local level, but never on an international level. Some suppliers are only active in one Baltic State, others also in other Baltic States.

26.The exact definition of the geographic market can, however, be left open, since the concentration does not give rise to competition concerns under any geographic market definition.

VI. COMPETITIVE ASSESSMENT

27. The operation leads to horizontal overlaps in the daily consumer goods markets and the procurement markets in both Estonia and Latvia, regardless of the exact geographic market definition. On the contrary, as already mentioned above, the parties are not active in the wholesale of daily consumer goods in any of the Baltic Countries. In addition, their activities do not overlap in Lithuania as Kesko is not at all present. ICA is present in Lithuania but its market share does not exceed [10-20]% according to any alternative product market.

28. Regarding the Estonian and Latvian markets, the market investigation confirms the parties’ submission that there is no comprehensive collected official third party data available concerning the size of the total markets and market shares. The parties therefore submitted market share information based on their own sales information, best estimates and internal business intelligence. However, the market investigation broadly confirms the market share information given by the parties. The assessment regarding the Estonian and Latvian markets is set out below, according to the worst case scenario. The market shares would be at lower level if a wider product market definition was considered.

A. ESTONIA

1. Retail market for daily consumer goods

29. According to the market investigation the Estonian retail market for daily consumer goods continues to be in a state of transition. In fact, it has been estimated that the Estonian retail would still grow annually by at least [5-15]% during the next 3-5 years. According to some estimates the market would grow even by [25-35]%, mostly due to the growth of purchase power. According to the Statistical Office of Estonia, [45-55]% of daily consumer goods sold in Estonia in 2003 were sold through retail chains. This share of retail chains has been estimated to grow. It should also be mentioned that according to market studies available, the Estonian retail sector is reported to be very competitive.

30. Kesko operates in Estonia with 6 hypermarkets, 1 supermarket and 44 discount stores, whereas ICA has only 5 supermarkets. The combined market shares on the retail market for daily consumer goods in Estonia are shown in the table below. Whatever the product market, the overlap is lower than [0-5]%.

31. On the basis of regional market definitions, 2 regions would be affected by the merger. In the Tallinn region, the parties would have combined market shares of [40-50]% with an overlap of less than [0-10]%, facing Selver with [15-25]% and Prisma with [10-20]%. In the Narva region, the combined market share of the parties would be below [35-45]% (ICA: [20-30]%, Kesko [5-15]%) facing ETK with [25-35]% and Selver with [5-15]%.

32. On the basis of local market definitions (catchment area of 10-30 minutes by car), 2 cities (Tallinn, Johvi) would be affected by the merger. The combined market shares of the parties would be below [40-55]% in Tallinn (overlap of [0-10]%) facing Selver with a market share of [20-30]% and Prisma with a market share of [15-25]%. In Johvi the combined market share would be below [45-55]% (overlap of [10-20]%), facing ETK with [30-40]% and OG Elektra with a significant market share ([10-20]%).

33. These market positions do not give rise to concerns for the following reasons.

34. First, apart from the Johvi local market, the overlaps in all hypothetical geographic and product markets are minimal, and ICA's and Kesko's positions appear to be rather complementary with ICA focussing on supermarkets and Kesko on hypermarkets and discount stores.

35. Second, the retailing industry appears to be growing rapidly with all players having significant expansion plans. According to the market investigation, many of the competitors that are already present in Estonia, continue to open up more outlets. As already mentioned above, the Estonian retail market for daily consumer goods is growing strongly, at least approximately by [5-15]% per year within the next 3-5 years, partly due to the growing incomes.

36. Regarding the parties’ existing main competitors, the Selver retail chain is a fast growing domestic chain established in 1995, and it operates supermarkets and hypermarkets. ETK comprises of a large number of different consumer-owned cooperatives and its largest retail chains are the Maksimarket and Konsum chains. OG Elektra AS, which is a large supermarket and neighbourhood store chain, competes with the parties especially in the discount segment in Tallinn and in eastern Estonia. Stockmann AS, which belongs to the Finnish Stockmann Group, competes with the parties in Tallinn.

37. Furthermore, in 2004 the Lithuanian VP-Market Group (T-Market stores), which is the largest retail supermarket chain in the Baltic States, entered the Estonian retail market strongly and it is currently expanding its operations. It has publicly announced that it would open a new store in every town in Estonia with a population of 10 000 or more. VP-Market has already opened 6 T-Market stores in Estonia in 2004, and according to an Estonian newspaper, they are planning to open 15 T-Markets altogether in 2004 and 40 during the next three years.VP-Market competes with Kesko in the discount store segment.

38. Another strong potential entrant, Lidl (Germany), has also publicly announced that it would start its operations in Estonia in the near future and that it would be interested in having at least 8 stores there. It has already established two companies in Estonia; Lidl Eesti Oii and Lidl Uii, and is currently searching suitable locations and recruiting personnel in Estonia. Also the name of a third strong potential entrant, Metro (Germany), has been brought up by competitors.

39. Third, the market investigation confirms that there are almost non-existing barriers to entry, both from a legal point of view and also from the perspective of the availability of suitable locations or plots to establish new stores. The market investigation further confirms that the elements related to the establishment of new stores are inherent to a normal store building process in all EU Member States and do not constitute a particular barrier to entry. According to the parties and the market investigation, there have been a lot of significant entries in the Estonian retail market for daily consumer goods. Furthermore, as already mentioned, several new competitors are about to enter the market in the near future.

40. Therefore with the exception of the Johvi area, the concentration does not appear to bring any significant change to the conditions of competition.

41. Johvi is a small town located in the eastern part of Estonia with approximately 12.000 inhabitants. Even assuming that the Johvi area could be considered to present all the characteristics of a distinct market, and to constitute a substantial part of the common market, the concentration would not lead to serious doubts in that area for the following reasons. ICA operates one supermarket and Kesko two discount stores in the Johvi region. The parties are faced with competition especially from OG Elektra (8 discount stores) and ETK (1 discount store). In addition chain retailing is less developed in Johvi. Due to the lower than average household income, people still appear to prefer to do a large part of their purchases in the open market or at neighbourhood stores. It may therefore be considered that at least in this area the traditional shop formats will continue to provide a competitive constraint to chain retail. This conclusion would be supported by the findings of the GAIN report, according to which the independent groceries and open market are considered serious competitors to retail stores in the Estonian rural areas and small towns. In any event, as explained above, given the growth in the demand for such retailing, it can be expected that new retailing outlets will be built in the near future. In fact, according to an Estonian newspaper, for example Lidl plans to open up an outlet in Johvi in the near future.

42. For the above reasons, therefore, it can be concluded that the combination of ICA's and Kesko's retailing activities in Estonia would not lead to unilateral or coordinated effects which would significantly impede effective competition. With regard to these activities therefore, the concentration does not raise serious doubts as to its compatibility with the common market.

2. Procurement market

43. According to the parties the combined market share on the procurement markets in Estonia are [15-25]% (Kesko: [10-20]%; ICA: [0-5]%) in the product category “liquids”, [10-20]% (Kesko: [10-20]%; ICA: [0-5]%) in the product category “fish market”, [15- 25]% (Kesko: [15-25]%, ICA: [0-5]%) in the category “fresh fruit and vegetables”, [15- 25]% (Kesko: [15-25]%, ICA [0-5]%) in the category “bread and fresh pastery”, [15- 25]% (Kesko: [15-25]%, ICA [0-5]%) in the category “butcher shop's products” and [15-25]% (Kesko: [10-20]%, ICA [0-5]%) in the category “dairy products”. In the product category “dry groceries” Kesko has a market share of [10-20]%, but there is no horizontal overlap. In all other product categories (consumer goods, bazaar, electronics, textile) the combined market shares do not exceed [10-20]%.

44. The information received in the Commission market investigation does not contradict these estimates. The increased sales volume to the new entity might eventually improve the parties’ position as regards the prices offered by the suppliers. However, on the basis of the market investigation results, it may be considered that the transaction would not lead to any significant change in the parties’ negotiating position on sales conditions with suppliers.

45. Considering the strong positions of other competitors in the Estonian retail market for daily consumer products, the transaction would not have a negative impact on competitors’ access to supplies either.

B. LATVIA

1. Retail market for daily consumer goods

46. The results of the market investigation confirm that the Latvian retail market for daily consumer goods is in a state of rapid transition where supermarkets emerge from mid nineties. Indeed, in 2002, this market grew by 20% and the retail sales turnover surged by 148% between 1997 and 2002. The retail market is still fragmented with a large number of independent retailers and some foreign players, mostly coming from the Nordic countries. In general, it has been estimated that the Latvian retail market for daily consumer goods would grow even stronger than the Estonian market. In fact, according to market surveys, retail sales are expected to grow in Latvia by approximately 30% until 2007.

47. Kesko operates in Latvia with 5 hypermarkets and 11 discount stores, whereas ICA is the second largest operator in Latvia with 4 Rimi hypermarkets and 33 Rimi supermarkets. On the basis of a national market definition the combined market shares for 2003 are shown in the table below.

48. The parties would have a combined market share of around [30-40]%, with an overlap of [0-10]%. Based on the information submitted by the parties, ICA's market share has declined from [35-45]% in 2001 to [25-35]% in 2002 and [20-30]% in 2003. Kesko, on the other hand, has increased its share of the market from [0-5]% in 2001 to [0-10]% in 2003.

49. The largest competitors of the parties are VP Market with [25-35]%, Nelda and Mego with [5-15]% each. VP Market's market share has increased rapidly from [0-5]% in 2001 to [20-30]% in 2002 and [25-35]% in 2003. Nelda's market share has also increased over the last tree years.

50. On the basis of a regional market definition parties have overlapping activities in four regions: Riga, Kurzeme, Vidzeme and Latagake. However, only 2 regions would be affected by the notified transaction. In the Riga region, the parties would have a combined market share of around [35-45]%,with an overlap of less than [0-10]%, facing VP Market with [15-25]% and Nelda with [5-15]%. In the Kurzeme region, the combined share of the parties would be of around [30-40]% ([15-25]%+[5-15]%) facing VP Market, Mego and Beta with significant market shares.

51. On the basis of local market definitions (catchment area of 10-30 minutes by car), 3 cities (Riga, Liepaja and Ventspils) would be affected by the merger. The combined market shares would be below [35-45]% in Riga (overlap of [0-10]%) facing VP Market and Nelda with significant market shares. In the other two cities the combined market shares are below [35-45]% or even [25-35]% and the combined entity would be facing a larger competitor.

52. These market positions do not give rise to concerns for the following reasons.

53. First, the concentration does not bring a significant increase in market shares in any of the potential relevant markets because the overlaps in all hypothetical geographic and product markets are minimal. Furthermore, ICA's and Kesko's positions appear to be rather complementary with ICA focussing on supermarkets and Kesko on hypermarkets and discount stores.

54. Second, the retailing industry appears to be growing rapidly with all players having significant expansion plans. According to the market investigation, many of the competitors that are already present in Latvia, continue to open up more outlets. As already mentioned above, the Latvian retail market for daily consumer goods is expected to grow strongly, approximately by 30% until 2007, partly due to the growing incomes.

55. The biggest competitor is the Lithuanian company, VP Market, which is the largest retail supermarket chain in the Baltic States. It entered the Latvian retail market in 2001 with the T-Market discount stores and Maxima supermarkets and has become the largest retailer in Latvia in 2004 with 83 stores (10 Maxima and 73 T-Market). VP Market plans to open new stores in Latvia next year.

56. Regarding other strong competitors, Nelda is one of the largest trading networks in Latvia, including supermarkets, shopping centres, cafes and restaurants. It was established in 1991 and owns 17 supermarkets all located in or around Riga. Nelda supermarket chain has grown significantly over the past fine years and is planning to expand further. Mego is a Latvian company which owns 33 stores, mainly in Riga and is planning to expand further. Beta is the fifth largest food retailer in Latvia with 16 supermarkets and was established in 1998. Others competitors are: Elvi (7 supermarket stores), TopGroup (8 supermarket stores) and Sky (2 supermarkets). In addition to these competitors, it is also worth to mention the impact of LLC Baltstore, which is the Latvian association of trade companies. It was registered in March 2004 and is a cooperative network aimed at pooling the resources of smaller retailers through joint purchasing, marketing and logistics. At present it unites 21 enterprises with 306 shops, Mego among the bigger ones involved. Baltstore targets to become the largest retailer in Latvia.

57. Third, the market investigation confirmed the parties’ view that there are almost non­existing barriers to entry, both from a legal point of view and also from the perspective of the availability of suitable locations or plots to establish new stores. It may be concluded that the elements related to the establishment of new stores are inherent to a normal store building process in all EU Member States and do not constitute a particular barrier to entry.

58. According to the parties and the Commission market investigation, several significant entries have recently taken place in the Latvian retail market for daily consumer goods. The most significant is the entry of VP Market. It developed its position from zero to a market leader within less than tree years. Moreover, the market investigation has confirmed the parties' view that new actors are expected to enter the market by end of 2005, such as the German company Lidl or the Lithuanian company IKI. Lidl has been placing advertisements to recruit staff and is searching for available locations and it has acquired a location for a distribution facility near Riga.

59. For the above reasons, therefore, it can be concluded that the combination of ICA's and Kesko's retailing activities in Latvia would not lead to unilateral or coordinated effects which would significantly impede effective competition. With regard to these activities, therefore, the concentration does not raise serious doubts as to its compatibility with the common market.

2. Procurement market

60. According to the information submitted by the parties, the combined market share of the parties in the procurement market in Latvia would be around [15-25]% (ICA: [10-20]%; Kesko: [0-5]%) in the product category “fresh fruit and vegetables”, and [15-25]% in the product category “dairy products” (ICA: [15-25]%; Kesko: [0-5]%). In all other product categories (for example consumer goods, bazaar, electronics, textile) the combined market share has not been estimated to exceed [10-20]%.

61. The information received in the Commission market investigation does not contest these estimates. The increased sales volume to the new entity might eventually improve the parties’ position as regards the prices offered by the suppliers. However, on the basis of the market investigation results, it may be considered the transaction would not lead to any significant change in the parties’ negotiating position on sales conditions with suppliers.

62. Considering the strong positions of other competitors in the Latvian retail market for daily consumer products, the transaction would not have a negative impact on competitors’ access to supplies either.

C. POTENTIAL RISK OF COORDINATION

63. The transaction would not lead to a coordination of the competitive behaviour of Kesko and ICA in the sense of Art. 2(4) of the ECMR. Firstly, Kesko and ICA transfer all their current activities in the Baltic States to the proposed JV and they do not retain any other activities in these countries. Secondly, apart from the Baltic States, Kesko and ICA are not active in the same markets: Kesko, except for a very marginal activity in Sweden in the hardware sector, operates in the daily consumer goods markets only in Finland, where it holds a market share of approximately [30-40]% in retail and wholesale; ICA, which is active mainly on the Swedish and Norwegian retail markets for daily consumer goods, where it has a market share of about [30-40]% and [20-30]% respectively, does not have any activity in Finland. Thirdly, the relative size of the proposed JV is small, less than [0-10]% relative to the size of the retained activities. Therefore, their incentive to coordinate their competitive behaviour in other markets than the Baltic States will not change as a consequence of the proposed operation.

VII. CONCLUSION

64. For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the common market and with the EEA Agreement. This decision is adopted in application of Article 6(l)(b) of Council Regulation (EC) No 139/2004.

1. OJL24, 29.1.2004 p. 1

2. OJL24, 29.1.2004 p. 1

3. ICA Forbundet is a Swedish non-profit organisation of more than 2000 small and medium-sized ICA retailers. Through its subsidiary ICA Forbundet AB owns 30% of the shares in ICA. ^[1] Canica, a Norwegian limited company, has a 20% stake in ICA. It does not have activities in food retailing other than through the JV with ICA and Ahold.^[1] The joint venture between Ahold, ICA Forbundet and Canica was cleared by the Commission on 6 April 2000 (Case Comp/M. 1832- Ahold/ICA Forbundet/Canica).^[1] Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Notice on the calculation of turnover (OJ C66, 2.3.1998, p25). To the extent that figures include turnover for the period before 1.1.1999, they are calculated on the basis of average ECU exchange rates and translated into EUR on a one-for-one basis.

^4. Whether discounters are to be included in the product market was left open in COMP/M. 1684- Carrefour/Promodes par. 12 and COMP/M. 1221-Rewe/Meinl par. 17 (Decision 99/674/EC-OJ L 274,23.10.99). E.g. Commission's decisions in cases IV/M.784 Kesko/Tuko par. 19, 20 (Decision 97/277/EC-OJ LI 10, 26.4.97); COMP/M. 1221-Rewe/Meinl par. 12, 16; COMP/M. 1684-Carrefour/Promodes par. 9.^[1]       Imports by international producers via national suppliers is, however, an argument in favour of national procurement markets, see COMP/M. 1221-Rewe/Meinl, par. 84.^[1] E.g. decision in cases IV/M.1085-Promodes/Catteau par. 14, COMP/M. 1221-Rewe/Meinl par. 18 and COMP/M. 1684-Carrefour/Promodes par. 24.^[1] E.g. see COMP/M. 1221-Rewe/Meinl par. 18 and COMP/M. 1684-Carrefour/Promodes par. 25.^[1]     E.g. see COMP/M. 1221-Rewe/Meinl par. 81: strong indication for procurement markets according to distribution channels.^[1] E.g. see decisions in cases COMP/M. 1684-Carrefour/Promodes par. 16 and COMP/M. 1221-Rewe/Meinl par.77.^[1]See: Retailing in Emerging Markets - Europe, Retail Intelligence, February 2004, published by Mintel International Group Lt, p. 51^[1] Aripaev Online, May 31, 2004 ^[1] See for example M+B Planet Retail June 1, 2004 and Eesti Paevaleht Online January 12, 2004 ^[1] GAIN Report, Global Agriculture Information Network, 12/31/2002, Estonia^[1] Aripaev Online, March 24, 2004 and Aripaev July 7, 2004.

^5. The Parties make reference to case IV/M.784 - Kesko/Tuko; however, in this case it was only decided that it was not necessary to assess the impact on each individual product market since the pattern of demand was broadly similar and concentrated.

^6. See decision in case COMP/M. 1684-Carrefour/Promodes par. 16.^[1] Mintel; Retailing in emerging markets - Europe, Retail Intelligence, Febmary 2004, p.84 (www.mintel.com)^[1] The Latvian newspaper Dienas Bizness 27.10.2004