31997D0277

COMMISSION DECISION of 20 November 1996 declaring a concentration to be incompatible with the common market (Case No IV/M.784 - Kesko/Tuko) Council Regulation (EEC) No 4064/89 (Only the English text is authentic) (Text with EEA relevance) (97/277/EC)

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community,

Having regard to the Agreement on the European Economic Area, and in particular Article 57 (2) (a) thereof,

Having regard to Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings (1), as amended by the Act of Accession of Austria, Finland and Sweden, and in particular Articles 8 (3) and 22 thereof,

Having regard to the Commission decision of 26 July 1996 to initiate proceedings in this case,

Having given the undertakings concerned the opportunity to make known their views on the objections raised by the Commission,

Having regard to the opinion of the Advisory Committee on Concentrations (2),

Whereas:

1. The procedure under consideration concerns a request from the Finnish Office of Free Competition pursuant to Article 22 of Regulation (EEC) No 4064/89 (the Merger Regulation) to examine the acquisition of Tuko Oy by Kesko Oy. The request was received by the Commission on 26 June 1996. The operation was made known to the Office of Free Competition on 27 May 1996 by means of a press release sent from Kesko Oy. The request pursuant to Article 22 has therefore been made within the one-month period provided for in Article 22 (4).

2. After examination of the abovementioned request the Commission concluded that the request was admissible within the meaning of Article 22 of the Merger Regulation and that the concentration raised serious doubts as to its compatibility with the common market. By decision of 26 July 1996, the Commission accordingly initiated proceedings pursuant to Article 6 (1) (c) of the Merger Regulation.

I. THE PARTIES

3. Kesko Oy (Kesko) is a limited company incorporated in Finland. The share capital of Kesko is divided into exclusive shares and ordinary shares. Its ordinary shares are quoted on the Helsinki Stock Exchange. Holders of ordinary shares are entitled to one vote on a holding of between 1 and 10 000 shares and an additional vote for each subsequent 10 000 shares. No holder of ordinary shares controls more than 7,08 % of Kesko's share capital (corresponding to 0,16 % of the votes).

4. The exclusive shares are, in principle, only held by Kesko retailers (K retailers), their pension fund and other organs close to the K retailers. The total number of exclusive shares make up 38 % of Kesko's total capital, but, since they carry six votes for the first 100 shares held and an additional vote for each subsequent 100 shares, the collective of K retailers in effect control the majority of votes in Kesko. This arrangement, whilst not conferring on the K retailers joint control within the meaning of Article 3 of the Merger Regulation, in effect precludes any third party from acquiring control of Kesko. According to the company charter of Kesko, all members of its Supervisory Board shall be K retailers. The Supervisory Board nominates all other decision-making and executive organs of Kesko.

5. Kesko and the K retailers who are, legally, independent entrepreneurs, are commonly referred to as 'the K block`. All K retailers are contractually bound to Kesko. They are active in the retail sales of daily consumer goods and/or speciality goods. Kesko's activities are also within the fields of daily consumer goods and speciality goods, as well as in the provision of numerous services relating to those activities.

6. Tuko Oy (Tuko) is a Finnish limited company, active in the wholesale and retail of daily consumer goods and speciality goods. In addition to its wholly-owned retail outlets, Tuko has contractual agreements with a large number of, legally, independent T retailers. Tuko and the T retailers are commonly referred to as 'the T block`.

II. THE OPERATION

7. On 27 May 1996 Kesko acquired 56,3 % of Tuko's share capital, thereby obtaining 59,3 % of the voting rights, and direct control of Tuko.

III. CONCENTRATION

8. The operation constitutes a concentration within the meaning of Article 3 of the Merger Regulation.

IV. NO COMMUNITY DIMENSION

9. On the basis of the figures provided by the Office of Free Competition, the combined world-wide turnover of Kesko and Tuko exceeds ECU 5 billion. However, both Kesko and Tuko achieve more than two-thirds of their respective Community-wide turnover in Finland. It follows that the concentration has no Community dimension within the meaning of Article 1.

V. EFFECT ON TRADE BETWEEN MEMBER STATES

10. Kesko has argued that it is not within the Commission's competence to assess its acquisition of Tuko. In Kesko's view, the Commission would be excluded from assessing concentrations following a request pursuant to Article 22 (3) when the undertakings concerned realize more than two-thirds of their Community-wide turnover in the same Member State. However, Kesko's argument stems from a misinterpretation of the Interpretative Notes adopted at the same time as the Merger Regulation, which deal only with the intention of the Commission not to apply its residual powers pursuant to Article 89 of the Treaty of Rome if certain quantitative criteria are not fulfilled.

11. Article 22 (3) of the Merger Regulation requires that an effect on trade between Member States is shown. The acquisition of Tuko by Kesko will create foreclosure effects for new entrants, including potential entrants from other Member States, in particular on the Finnish markets for daily consumer goods. In addition, a large amount (about 30 %) of the products sold by both Kesko and Tuko originates outside Finland. The transaction will also affect trade between Member States in that suppliers from other Member States will, in effect, require access to Kesko's distribution channels to secure sufficient marketing of their products in Finland.

12. Moreover both companies are members of several international purchasing organizations, together with similar companies in other Member States. Since the spring of 1996, Kesko has also expanded its operations by opening retail outlets in Sweden.

13. It follows from the above that the change in the structure of the Finnish retail and wholesale markets for daily consumer goods will have an appreciable influence, directly or indirectly, actually or potentially, on the pattern of trade between Member States (3).

VI. ASSESSMENT PURSUANT TO ARTICLE 2 OF THE MERGER REGULATION

A. RELEVANT MARKETS

14. Kesko has argued that the current operation should be assessed at the wholesale rather than the retail level. The reason for this would be that the K retailers and T retailers are independent from the Kesko and Tuko companies and their central units. The assertion is based on the grounds that Kesko and Tuko basically provide purchasing, marketing, informatics, and other support services to the retailers, who nevertheless retain a large freedom of choice concerning supply sources, pricing policy, and other commercial matters, and indeed compete with each other as well as with other retailers.

15. Kesko's above argument cannot, however, be accepted. The assessment of the competitive impact of any concentration has to be based on all structural elements relating to the concentration, the parties thereto and the markets concerned. In this respect it is particularly appropriate to analyse the internal relationships of the Kesko group in the light of the major reorganization of this group, which was implemented in 1995. This analysis, see paragraphs 39 to 66, leads to the conclusion that the operation should be assessed at retail level since it is appropriate to consider Kesko, including the K retailers, as a centrally planned, structural feature of the Finnish retail market.

16. In addition to the assessment at retail level, an assessment will also be necessary at wholesale level, as regards cash-and-carry sales, as well as on the procurement markets for daily consumer goods.

17. According to the investigation, the concentration will primarily have an impact on the following markets:

(1) the retail market for daily consumer goods;

(2) the market for cash-and-carry sales of daily consumer goods;

(3) the markets for procurement of daily consumer goods.

1. The retail market for daily consumer goods

I. Relevant product market

18. The range of goods concerned includes principally food, drink, tobacco and non-food household consumables (that is, cleaning products, toiletries, disposable paper products, and healthcare products). A common characteristic of these products is that they all form part of a 'basket of daily consumer goods` that consumers expect to find in a supermarket environment. As the Finnish grocery retail sector has developed from small specialized stores (butchers, bakers, etc.) to the larger supermarket stores, consumer behaviour has developed accordingly. Therefore competition in Finland today takes place between supermarkets and other stores that are able to provide the kind of wide selection that enables the consumer to purchase most of the household necessities in a 'one-stop shop`, with attendant conveniences, such as parking facilities, trolleys, etc. (See also previous Commission decisions within this sector) (4).

19. Although the basket of goods supplied by different supermarkets may vary considerably in physical size, some smaller types of outlets, including specialized stores (which are relatively uncommon in Finland), kiosks and petrol stations, do not offer a service which is in direct competition with the basket of goods concept. Petrol stations have a selling area much smaller than even the smallest Kesko and Tuko outlets, and the sale of daily consumer goods at petrol stations is heavily focused on dry goods (about 70 to 80 % of total sales), whereas traditional daily consumer goods outlets achieve about 50 % of their sales from fresh produce.

20. In conclusion, the relevant market consists of the provision of a basket of fresh and dry food-stuffs and non-food household consumables sold in a supermarket environment. The market does not include sales at specialized stores, kiosks and petrol stations. Instead, these outlets provide a service that is complementary to those of supermarkets.

II. Relevant geographic market

21. Retailing professionals consider the catchment area for a supermarket to be delimited by a boundary within which the supermarket can be reached by car in no more than about 20 minutes. From the consumers' viewpoint, a 20-minute car journey might offer a choice of one, two, or more supermarkets and smaller outlets which also offer a 'basket of goods`, depending on his and their location. It is clearly impracticable to define exactly all the possible combinations and permutations, which are a function of the density and distribution both of the population of consumers and of the network of supermarkets. Furthermore, the degree of overlapping which occurs between supermarket catchment areas, together with the distribution of the consumer population, will not only determine the competitive interactions between geographically proximate supermarkets, but will also, to some extent, have a 'knock-on` or 'chain-reaction` effect on more distant supermarkets. Moreover, many important parameters of competition, such as range of products, sources of such products, quality, service level (opening hours, etc), advertising, promotion and prices (e.g. on campaign products) are not decided on the local level, but on a regional or national level. Such decisions are generally implemented at a number of retail outlets within a certain region. In view of the above, it is necessary and appropriate to aggregate what, from the viewpoint of any single consumer, may be a very local market.

22. Aggregation of sales figures can be made on, in principle, an indefinite number of different levels. As far as Finland is concerned, the widest possible geographic market is national, for reasons of, for example, language differences and consumer purchase preferences.

23. Both Kesko and Tuko (as well as their main competitors) operate their retail outlets according to national chain concepts, for which joint marketing and pricing decisions are taken. Moreover, customer loyalty schemes, such as the K-advantage card, are applicable all over Finland. Even though these factors indicate a relatively wide geographic market, it is not necessary for the assessment of the operation to decide finally on the exact scope of the relevant geographic market, given that the combined market share of Kesko and Tuko is of the same order whether assessed on national, regional or local level (see paragraph 106).

2. The market for cash-and-carry sales of daily consumer goods

I. Relevant product market

24. The products handled by wholesalers of daily consumer goods, naturally, mainly reflect the sales of such products at retail level and the use of such products by industrial households. The Commission's investigation has shown that in the Finnish market there are significant differences in 'classic wholesale` sales, and cash-and-carry sales.

25. Kesko has argued that the 'classic wholesale` market should be subdivided into three different markets: negotiations with suppliers, central invoicing and logistics. However, since Kesko's activities within 'classic wholesaling` vis-à-vis external customers have been of minor importance (about 1 %), and the concentration therefore does not raise competitive problems within this field, it is not necessary to make any assessment on Kesko's arguments in this respect.

26. Both Kesko and Tuko achieve significant turnovers from their sales in cash-and-carry outlets. As the term implies, cash and carry, as a means of wholesaling, is different from 'classic wholesaling`, in that customers provide the transport themselves and pay for the goods in cash rather than on credit. Cash and carry is therefore advantageous for customers who are less able to predict long-term demand, who do not have the physical facilities to hold large stocks for a period of time, who do not have the financial resources to guarantee the credit required for large bulk purchases and who therefore prefer to buy smaller quantities on a more frequent basis. Such customers are generally smaller companies, whereas larger customers tend to use wholesaling, as Kesko itself has confirmed in its reply to the Commission's Statement of Objections.

27. The distinct function of cash-and-carry outlets is reflected by the number of items available from, on the one hand cash-and-carry outlets, which normally stock about 10 000 different items, and 'classic wholesale` units, which normally have a much smaller number of items in stock (except for the central warehouses, in Helsinki, of Kesko, Tuko and Inex Partners). It is also reflected by the fact that Kesko has organized its cash-and-carry activities in a different subsidiary.

28. The main customer group of cash-and-carry outlets consists of industrial households (hotels, restaurants, hospitals, etc). The majority of these customers are small and medium-sized companies, which are, for the reasons mentioned in paragraph 26, largely unable to switch to 'classic wholesaling`. According to Tuko's annual report for 1995, its total sales through cash and carry equalled those of 'classic wholesale`, however, the ratio of the number of cash and carry customers to wholesale customers was approximately four to one. This shows that the average cash-and-carry customer purchases only about one-quarter of the amount purchased by the average 'classic wholesale` customer. The main product group purchased by these customers is dry-food products, which account for 60 to 70 % of total sales of Kesko and Tuko within this market, the remaining part being made up of fresh food and non-food products.

29. Given the distinct characteristics of cash-and-carry services and given that a substantial majority of cash and carry customers are in practice captive customers, even in the event of a rise in prices of the goods sold, the limited financial and storage capacities of these customers would make it uneconomic for them to switch to 'classic` wholesaling.

30. Kesko has stated that a number of its customers use classic wholesaling and cash-and-carry services 'in parallel`, and draw the conclusion that a rise in cash-and-carry prices would cause these customers to switch entirely to classic wholesaling, thus in practice rendering such a price rise unprofitable. The argument is fallacious in that it confuses complementarity with substitutability. Given the distinct differences between classic wholesaling and cash and carry as already outlined, it is clear that the limited number of customers who use both types of service regard them as complementary rather than substitutable. For example, a customer who finds it appropriate to use classic wholesaling to supply some of his requirements may have no alternative to using cash-and-carry services to meet the rest of his requirements (for example, for particular types of goods, or when his storage facilities are currently fully utilized, or when urgent 'topping up` is required to meet a sudden surge in demand). Therefore, the existence of a limited number of 'dual usage` customers does not indicate that a price rise in cash and carry would cause such customers to switch to classic wholesaling to an extent which would render a cash-and-carry price rise unprofitable, and thereby 'shielding` captive customers of cash and carry.

31. For the above reasons, sales from cash-and-carry outlets constitute a relevant product market that is separate from that of other wholesaling activities.

II. Relevant geographic market

32. The customers of cash-and-carry outlets are professional customers, mainly industrial households and, to a lesser extent, retailers. As such they are able to purchase within a relatively wide area, even though they will have to cater for the transport of the goods themselves. Evidence shows that customers of cash and carry buy regionally. Given the skewed population of Finland, cash-and-carry customers in northern Finland will generally have to travel further for their supplies than customers in south-western Finland, on the other hand regions are larger in northern Finland. It is therefore considered appropriate to assess the impact of the operation on a regional level.

3. Markets for procurement of daily consumer goods

I. Relevant product markets

33. The markets involve the sale of daily consumer goods by producers of such goods to customers such as wholesalers, retailers, and other enterprises. Not all producers will supply the full range of daily consumer goods; some may specialize, for example, in individual products or product groups, such as fresh products, or dry food or non-food products.

34. Again, from the demand side, corporate customers such as large supermarket chains will probably attempt to obtain the best possible conditions for their purchases of individual products, such as meat, canned food, or any other products. Each product or product group constitutes an individual product market as it is not, or imperfectly, substitutable for other products or product groups from the point of view of the demand side. However, since, in Finland, the pattern of demand for each of the product groups is broadly similar and concentrated (see paragraph 150), it is not necessary to assess the impact of the concentration on each individual product or group of products. It is sufficient to assess the impact of the increased buying power, brought about by the new demand structure, across the whole range of daily consumer goods.

35. It should be acknowledged that, especially for producers of non-food items, distribution channels other than supermarkets, e.g. pharmacies, are available. However, these are unimportant in relation to supermarket and cash-and-carry channels (see paragraphs 146 to 153) and would not affect the assessment of this case.

II. Relevant geographic market

36. About 70 % of the products sold by Kesko and Tuko consists of domestically produced goods. The manufacturers of these products achieve most, if not all, of their turnover in Finland. In view of this, these suppliers have stated that they regard Finland as the relevant geographic market.

37. Of the 30 % of imported products sold by Kesko and Tuko only a relatively small part, about 5 percentage points, is actually imported by the companies themselves. The remaining part is purchased from Finnish subsidiaries of international producers or from companies specializing in trade with imported goods. As regards such goods, and given that the present concentration affects the demand side of these markets, it is appropriate to define the geographic markets with particular reference to the ability of suppliers to switch their means of channelling goods to final consumers. As regards the manufacturers in question (much of whose goods are specially adapted for sale on the Finnish market, e.g. packaging, languages, etc.), access to the demand side, i.e. retail and cash-and-carry customers, can only be gained via Finnish retail/cash-and-carry distribution channels. Therefore as international manufacturers cannot switch to non-Finnish distribution channels it can be concluded that the relevant geographic market is Finland.

B. EFFECTS OF THE CONCENTRATION

38. The three daily consumer goods markets described above are vertically inter-connected in such a way that the position of Kesko and Tuko on the retail market for daily consumer goods creates the basis for their positions on the procurement markets (e.g. vis-à-vis suppliers of daily consumer goods). Conversely their position at the procurement level significantly affects their position at retail level. The market for cash-and-carry sales is also connected to their position on the procurement markets, in that Kesko and Tuko, from buying significant amounts for their retail operations, will have a further competitive advantage over the smaller, local cash-and-carry operators.

1. Retail of daily consumer goods

I. Structure of supply

1. Voluntary retail chains in Finland

39. Traditionally the Finnish retail market has been characterized by the existence of and important role played by voluntary retail chains. However, in the field of daily consumer goods there are now only two such chains in Finland; the K and T blocks. One of the main objects of these chains is to make buying and promotion possible on a larger scale than is possible for individual retailers. A common characteristic for these retail chains is that the benefit of the cooperation is secured through contracts. These contracts, while in principle of a horizontal nature (i.e. between each of the retailers of the K and T block respectively) are, for reasons of simplicity and surveillance, concluded between the individual retailer and the central organs of the group.

40. The contracts between individual retailers and the central organ focus on standardizing the behaviour of individual retailers, and thus limiting their independence, in areas such as purchasing, identification, promotion and sales policy, which are of key importance to the achievement of the purposes of the group. The level on which individual behaviour needs to be standardized for the system to be successfully implemented will depend on the competitive relationship, and thus inter-dependence, between the individual units of the chain. It is therefore consistent that Tuko and Kesko, in particular following its reorganization (see below), have standardized the behaviour of individual retailers to a much greater extent than, for example, international joint purchasing organizations or other similar organs, which do not to the same extent involve companies that in the absence of the cooperation would be in direct competition with one another.

41. Since the achievement of the group's purposes is dependent on the individual retailer's adhesion to the joint policies, as set out in the contracts, Kesko has been entrusted with internal power mechanisms to ensure individual retailer adherence to the common objectives. These internal power mechanisms include incentive creating schemes (e.g. the payment of bonuses and provision of services related to the business of the retailers) as well as dis-incentive schemes (e.g. the removal of benefits, exclusion or the threat thereof). Such dis-incentive schemes will be very effective where, as in the case of Kesko, individual members have made significant investments in the organization and are to a large extent dependent on the central organs for their business operations.

2. Nature and structure of the K block

42. From a company law perspective, the K retailers are independent legal entities, and the entrepreneurs themselves carry the economic risk of their business. For the reasons stated below, this cannot, however, be taken as a decisive criterion for assessing whether Kesko's acquisition will affect the retail market.

(a) Common interest

43. As described in paragraph 4, the K retailers together hold the majority of votes in Kesko. Moreover, all members of the Kesko Supervisory Board, which nominates all other decision-making and executive organs of Kesko, are K retailers. This arrangement reflects the strong common interest of the K retailers.

(b) Agreements between Kesko and the K retailers

44. The K retailer agreement specifies the general rights and obligations of the K retailers, whereas the chain agreement defines the rights and obligations related to the membership of one of Kesko's five sub-chains (see paragraph 47). According to the K retailer agreement the retailer undertakes to 'carry on, as its main area of activity, K retail which must be profitable` and, further to 'attempt to take full advantage of the benefits of joint purchasing of the K group and its private label products. The K retailer shall not, without justification, treat Kesko less favourably than other suppliers`. The chain agreements state further that the K retailer and chain agreements together constitute the basis for cooperation between the retailer and Kesko; and that the main purpose of these agreements is to promote trade in goods between the parties (i.e. the retailer and Kesko). According to the chain agreement, the K retailer is bound by decisions of the chain's Board of Directors concerning marketing, products to be entered into the basic selection and retail prices of campaign products (5). The result of this, in 1995, was that the K retailers in the five chains obtain between [75 and 100] % (6) of their total purchases through Kesko (see paragraph 72).

45. The K retailer obligations further include using the Kesko logotypes. The K retailers do not pay any consideration for the use of such logotypes, slogans or other intellectual property rights owned by Kesko, nor do they pay any consideration for some of the support services provided by Kesko (e.g. marketing assistance). Instead the K retailers pay an indirect consideration for such services through the margins which are added on when goods are transferred from Kesko's chain control unit to the retailer. Therefore any retailer wishing to purchase, for example, marketing assistance from a third party instead of Kesko would, in practice, be paying twice for these services. Thus, it must be concluded that the K retailers are intended to use Kesko as their only provider of such support services.

46. Moreover, the K retailers receive bonuses and rebates from Kesko based on volumes purchased through Kesko. The fact that these rebates increase according to the amounts purchased provides a typical incentive for retailers to remain loyal to the group strategy.

(c) Organization of Kesko

47. Kesko, as the central organ of the K group, is responsible for the corporate planning of the group's activities. In 1995 a major reorganization programme was implemented in the group's organizational structure. The main focus of this reorganization was to remodel existing K retailers into five national Kesko sub-chains, each of which would consist of retail outlets with common characteristics in terms of identification (logotype), size of sales area, basic product assortment and which would cooperate between themselves in joint advertising, campaigns, etc. (see paragraphs 67 to 72 for a detailed description of these chains).

48. To support this retail reform, five matching chain control units were created within Kesko's foodstuffs division. These chain control units form, together with a Board of Directors made up of retailers from the chain, the nationwide management of the chain to which it corresponds. The responsibilities of the chain control unit, which is totally operated by Kesko employees, include coordinating activities of the retailers in the chain concerning purchasing, marketing and sales policies, e.g. joint campaigns. The chain control units also serve as Kesko's profit centres (i.e. by calculating Kesko's profit margin on to all jointly purchased goods by each sub-chain before resale to the retailers).

49. All chain control units are part of Kesko's foodstuffs division. As such they share common management structures and report to the same manager (one of Kesko's executive Vice-Presidents).

50. One important and integral part of the reorganization is that the K stores of all five chains will be equipped with up-to-date information systems suitable for use in the Kesko network. These computer systems will be placed in the retail stores, but will continue to be owned by Kesko. Following completion of this project, the possibilities for rapidly transmitting information within the group will be significantly improved. The fact that Kesko has publicly stated its commitment to increased use of information technology is in line with the development in other retail groups throughout Europe. For example, the installation of check-out scanners connected to a central computer provides a basis for increased coordination of stock control, product range selection and flow of price information.

K retailers decisions on sourcing

51. Kesko has argued that K retailers take their decisions on sourcing independently from Kesko. To support this view, Kesko has indicated that only [30 to 50] %, as an average of the stores of all five chains, of the K retailers total supplies are purchased from Kesko's wholesale function, whereas [30 to 50] % is supplied directly from producers to the retailers and merely invoiced by Kesko.

52. The Commission's investigation, however, strongly indicates that producers, when supplying their products, perceive Kesko and the K retailers as an integrated entity, irrespective of the fact that the actual contact point within the entity may differ according to circumstances, and irrespective of the sales channel employed (whether classic wholesaling or central invoicing). This is supported by the following facts:

- all purchases by K retailers which are invoiced by Kesko are based on invoicing agreements concluded between Kesko and its suppliers, i.e. it is not possible for a K retailer to use this sales channel without the active involvement of Kesko,

- in relation to all invoicing transactions, Kesko does take legal possession of the goods in question, which are then resold to the retailer that ordered the goods. As such these transactions are included in Kesko's annual income statement as sales and not separated from Kesko's 'classic wholesale` sales,

- moreover, invoicing agreements concluded between Kesko and its suppliers generally specify the marketing support fees and/or other rebates, based on sales volumes, granted by the supplier to Kesko. These fees and rebates, which are the main method by which suppliers differentiate between their customers, are based on all purchases of the K group, i.e. including sales to Kesko in its capacity as a wholesaler as well as purchases made directly from K retailers based on invoicing agreements. Suppliers would not agree to pay such support if they did not believe that Kesko is in a position to ascertain the fulfilment of agreements (e.g. on the achievement of set volume targets or to carry out agreed promotional campaigns at store level),

- although Kesko has stated that '. . . it is not necessary in the routine work (of the Kesko personnel handling the invoices) to pay attention to individual products and their prices specified on invoices` it has to be concluded that Kesko, as it handles about [. . .] million invoice transactions every year (most, but not all, of which concern transactions involving K retailers), will have the possibility to gain enormous amounts of information concerning prices and other commercial terms applied by individual suppliers. Moreover, Kesko will only be able to ascertain that marketing fees and bonuses are received according to its agreements with the suppliers, if total amounts are followed.

53. In view of the above, the Commission cannot accept that the [30 to 50] % of total K retailers' supplies which are invoiced, but not physically delivered by Kesko, should be regarded as sourcing which is independent of Kesko. On the contrary, it is clear that Kesko has a central role even in these relations. It can therefore be concluded that whether certain supplies will pass through Kesko's warehouse facilities ('classic wholesale`) or be delivered directly to the retailer is basically a logistical question, depending on the type of good involved, the size of the order, etc.

K retailers' decisions on marketing and pricing

54. Kesko has also argued that the independence of K retailers vis-à-vis Kesko is demonstrated by the retailers' freedom as regards their individual commercial behaviour on the market, i.e. concerning advertising and pricing. To support these allegations Kesko has supplied three studies concerning price dispersion and advertising behaviour.

55. The Commission has identified several weaknesses in the methodology employed in these studies, as well as in the presentation of the data on which they are based (e.g. lack of clarity regarding selection criteria in terms of stores, products, geographic distribution and the period of the studies). These methodological problems are not, however, of vital importance given that it can be seen in two of the studies that a common feature is that the alleged individual freedom is demonstrated by comparing prices and advertising behaviour of the different Kesko chains, rather than individual behaviour within those chains. As the five Kesko chains have been created by Kesko to join all stores that present similar characteristics and defined chain policies in the same chain, it would be an expected result that the pricing and advertising behaviour differs between the chains.

56. If the studies are viewed with respect to the individual Kesko chains, a very high degree of concerted behaviour is found. Concerning advertising, the studies show that the amount spent on joint advertising within each chain is between [50 and 100] %. The study then continues to state that '. . . retailers from a chain which are not competing with each other will generally place joint advertisements rather than advertising separately`. Similarly, the study that purposes to show differences in advertised prices over a five-day period actually confirms that K retailers within the same chain showed a high degree of conformity.

57. In view of the above, and given Kesko's role in organizing the K retailers into the five different chains, it has to be concluded that the differences in terms of advertising and pricing behaviour that exist between the chains is part of a centrally planned strategy for the whole Kesko group.

(d) Ownership of business premises

58. As has been stated above, Kesko is responsible for the development and maintenance of the network of outlets of the K block. One area of particular strategic importance to Kesko is the ownership of business premises on which the retailing activities is conducted. Indeed according to Kesko's balance sheet, its property investments amount to almost two-thirds of its total fixed assets.

59. The strategic importance Kesko has attributed to being the owner of the retail business premises can be demonstrated by the fact that all retail outlets planned and/or built by Kesko continue to be owned by Kesko, but are leased to K retailers. Kesko's interest in maintaining the ownership of the business premises is particularly clear as regards larger outlets. More specifically, Kesko owns the business premises of [. . .] Citymarket and Rimi stores, [. . .] % of the Supermarkets and [. . .] % of the K markets. In the Neighbourhood store chain [. . .] % operate on premises owned by Kesko. Thus, the majority ( > 60 %) of the total turnover of all K retailers is realized from operations where Kesko owns the business premises. The K retailers operating on premises owned by Kesko have entered into a collaboration agreement with Kesko which sets out the principles of Kesko's business premises operations and the fee, calculated from the turnover or profit margin of the retailer, which Kesko receives as rent.

60. While no K retailer can, without the approval of Kesko, transfer (sell or sub-lease) his business as a going concern, the retailers that operate from business premises owned by Kesko have even more limited possibilities of transferring their business.

61. K retailers are not given an exclusive territory, in which they will be the only K retailer. In fact, the retailer's agreement with Kesko obliges him to accept that other retail outlets may be set up in the same area. The collaboration agreement can be terminated by Kesko for 'a valid reason relating to the development of the network`.

(e) Financial commitment

62. The obligations of a K retailer include owning a minimum of [. . .] exclusive shares in Kesko or, where greater, shares representing the equivalent of [. . .] weeks' purchases from Kesko (including amounts invoiced by Kesko). The average number of shares held by K retailers is [. . .], representing a stock market value of ECU [. . .]. These shares are pawned to Vähittäiskaupan Takaus Oy, VT, and used as a collateral for the retailers' purchases from/through Kesko. VT is [. . .] % owned directly by Kesko, and [. . .] % owned by the K retailers association and individual retailers, the remaining shares being held by other organs close to, and partly owned by Kesko (these organs are also the most important shareholders in Kesko). According to Finnish law, Kesko itself would not be allowed to accept its own shares as collateral. However, given the shareholding structure of VT, it is clear that, from the retailers' point of view, the shares are in effect pawned to Kesko.

3. Conclusion

63. Kesko, as the central organ of the Kesko group has responsibility for the planning and execution of changes to the overall group structure. Following the recent reorganization, it has organized the K retailers in five national Kesko sub-chains, each with its own specific characteristics, and each corresponding to a control unit within the Kesko company. Therefore, it has to be concluded that the horizontal cooperation and lack of competition within each of these five Kesko chains constitutes a structural feature of the Kesko group, and, as such of the Finnish retail market. The same is also true for whatever competition that may seem to exist between the five chains, since the main elements of that competition have, in fact, been centrally planned by Kesko.

64. It follows from the above that although the K retailers bear the financial risk of their business operations, they have through the agreements concluded with Kesko accepted to adhere to the overall principles planned by Kesko for the whole group. The fact that the K retailers make a major investment in Kesko through the acquisition of shares in the company, receive a number of support services from Kesko, but are not given an exclusive territory, which means that Kesko, at any particular time, may allow a new outlet to be set up in the same area, creates a strong combination of incentives and dis-incentives for K retailers, and in particular those who rent their business premises from Kesko, to remain loyal to Kesko and the Kesko chain in which the retailer has been placed.

65. Most of the above analysis concerning the relationship between Kesko and the K retailers is applicable also to the relationship between Tuko and the T retailers. However, it is not necessary to assess finally the position of T retailers prior to the concentration. The fact that Kesko would control Tuko would confer on it the ability to organize the various T retailers in a similar way as it has organized the K retailers in five different sub-chains. Indeed, Kesko would be in an even stronger position vis-à-vis the T retailers than Tuko, since the operation will in effect make it impossible for the T retailers to join a competing voluntary chain. (The transaction would, in the same way, also further increase the dependency of K retailers on Kesko.)

66. In view of the abovementioned elements, i.e. agreements within the K group, the organizational characteristics, ownership of business premises and financial commitment, it is appropriate to consider Kesko, including the K retailers, as a centrally-planned, structural feature of the Finnish retail market, Kesko's assertion that the competitive impact of the operation should not be assessed at the retail level cannot be accepted. Similarly, it would be appropriate to consider the T retailers as an integral part of the Kesko group following the operation.

(i) National chains operated by Kesko and Tuko

Kesko

67. The smallest Kesko shops are included in the Neighbourhood Store chain (K shops); there are 1 115 such shops in Finland. Their average sales area is [. . .] m² and the average number of items sold in K shops is between [. . .]. The average annual turnover of a K shop is about ECU [. . .]. According to Kesko the basic range of products decided by the chain's Board of Directors covered [. . .] % of all sales in 1995, and [. . .] % of all sales resulted from joint campaigns.

68. There are 226 Kesko Supermarkets (KK shops) in Finland. These have an average sales area of [. . .] m² and sell between [. . .] items. The average annual turnover of a KK shop is about ECU [. . .]. According to Kesko, the basic range of products decided by the chain's Board of Directors covered [. . .] % of all sales in 1995, and [. . .] % of all sales resulted from joint campaigns.

69. The Kesko Superstores (KKK shops), of which there are 78 in Finland, are, on average [. . .] m² and sell between [. . .] items. These shops have an average turnover of ECU [. . .]. According to Kesko the basic range of products decided by the chain's Board of Directors covered [. . .] % of all sales in 1995, and [. . .] % of all sales resulted from joint campaigns.

70. The largest shops are the Kesko Citymarkets (KKKK shops), of which there are 38 in Finland. In addition to daily consumer goods, these also have a speciality goods section, which is normally operated by Kesko itself. The daily consumer goods sections of Citymarkets are, on average, [. . .] m², sell between [. . .] items and have an average turnover of ECU [. . .]. According to Kesko the basic range of products decided by the chain's Board of Directors covered [. . .] % of all sales in 1995, and [. . .] % of all sales resulted from joint campaigns.

71. Finally, there are 19 discount Rimi Stores. Their average sales area is [. . .] m² and they sell [. . .] items. Their average turnover is ECU [. . .]. In the Rimi stores [. . .] % of the products are included in the basic range of products decided by the chain's Board of Directors. [. . .] % of all sales resulted from joint campaigns. According to Kesko's 1994 annual report, these stores are based on cost-efficiency and uniformly inexpensive prices. It is noteworthy that the agreements between Rimi retailers and Kesko contain basically the same obligations as Kesko's agreement with other K retailers.

72. The sources of supply were similar for all chains. Between [. . .] % were purchased independently from Kesko, whereas the remaining [. . .] % were purchased through Kesko, i.e. [. . .] % from Kesko's 'classic wholesale` and about [. . .] % invoiced through Kesko (Kesko's involvement in the invoicing transactions have been analysed in paragraphs 51 to 53). Although Kesko has not been able to supply exact figures, it is clear that Kesko's wholesale unit represents the major source of all chains for dry food and nonfood articles, whereas invoicing is mainly related to fresh produce.

Tuko

73. Following a recent reorganization, the T retailers are organized in three major groups.

74. The most important of the three groups is the Spar chain, which consists of 125 Sparstores, 216 Sparmarkets and 34 SuperSpar stores. The Sparstores are, on average [. . .] m², sell [. . .] items and have turnovers of ECU [. . .]. The corresponding figures for Sparmarkets are [. . .] m², [. . .] items and ECU [. . .], and for SuperSpar stores [. . .] m², [. . .] items and ECU [. . .].

75. Tuko also owns and operates 30 Anttila department stores. The daily consumer goods sections of these have an average sales area of [. . .] m², sell [. . .] items and have a turnover of ECU [. . .].

76. The smallest outlets within the Tuko group are the Tarmo stores. This group consists of stores that used to have T retailer agreements, but which were unsuitable for conversion into the Spar concept. Their relation to Tuko is based on 'the Tarmo promise`, under which Tuko, free of charge, grants to the shops the use of the Tarmo concept developed by Tuko. In return, the stores must, inter alia, stock a basic product range, participate in joint marketing and use advertising material decided by Tuko. There are in total 565 Tarmo stores in Finland. Their average size is [. . .] m² and they sell between [. . .] items. Their average turnover is ECU [. . .].

(ii) Cooperative organizations

77. In the Finnish daily consumer goods market, there are three cooperative retail groups: Suomen Osuuskauppojen Keskuskunta (SOK), Tradeka Oy (Tradeka) and Osuusliike Elanto (Elanto). The cooperative retail groups are based on cooperative societies which are owned by, in principle, members/consumers. Each cooperative society operates chains of retail outlets that are fully owned by it. The managers of the retail outlets of these groups are employees of the group.

SOK

78. SOK operates three different national chains of outlets: Prisma, S-market/Citysokos and Sale/Alepa. The outlets cooperate nationally and regionally in 23 regional cooperatives. The regional Board of Directors, which always include, at least, one representative of SOK, decide on marketing and pricing policy for all chains. Price differences between regions are mainly due to transport costs, which usually means that prices are higher in northern Finland. Although it is very rare that conflicts arise between the different regional cooperatives, the board of SOK, can settle such disputes. SOK is also responsible for planning and building new outlets. About 85 % of all outlets have been set up by SOK, the remaining part mainly consists of smaller and older stores.

79. The Prisma chain consists of 28 large outlets. Although the average selling area and number of items sold by these outlets is somewhat smaller than the Kesko Citymarkets, their average turnover is of the same order as that of the Citymarkets.

80. The S-market chain consists of 260 outlets. These outlets are in the same category as the KK shops in terms of selling area, number of items and average turnover.

81. The third chain operated by SOK is the discount Sale/Alepa stores. There are 220 such outlets. They are somewhat smaller than the Kesko Rimi stores, but nevertheless in the same category.

82. In addition to its activities in retail of daily consumer goods, SOK owns 50 % of Inex Partners Oy. The remaining 50 % of this company, which is active in the wholesale of daily consumer goods, is owned by Tradeka (see below). SOK is also active in the hotel and restaurant sector.

Tradeka and Elanto

83. Tradeka and Elanto are both cooperative organizations. Whereas Tradeka is active basically all over Finland, Elanto's activities are concentrated to the greater Helsinki area. Both groups are, due to financial difficulties, undergoing reorganization in accordance with the Finnish Act on Reorganization of Companies (No 47/1993). This will affect both companies until the end of 2003.

84. Partly as a result of their reorganization programmes, Tradeka and Elanto have, in 1996, entered into a far-reaching cooperation, covering the coordination of outlets, purchasing, selections, customer loyalty schemes, personnel and administration of information technology. Due to this cooperation, it is appropriate to assess both companies as part of the same group.

85. Tradeka operates three different chains of daily consumer-goods outlets. The largest outlets are the Euromarkets, of which there are 21 in Finland. These outlets are slightly smaller than the Kesko Super stores (KKK), in terms of sales area, number of items sold and average turnover. The 147 Valintatalo markets (28 of which are operated by Elanto) are smaller units, basically in the same range as the Tuko Sparmarkets, and the 479 Siwa stores (45 of which are operated by Elanto) are basically in the same range as the Tarmo stores. Three of Tradeka's outlets are operated from business premises owned by Tuko.

86. In addition to operating the Valintatalo and Siwa outlets in the Helsinki area, Elanto also operates three hypermarket Maxi outlets and 23 Marketta outlets. The latter are somewhat smaller than the K shops (K).

(iii) Stockmann, Wihuri and Sentra

87. One of the largest private companies in the field of retail of daily consumer goods is Oy Stockmann Ab (Stockmann), which has five department stores and 17 Sesto discount stores in southern Finland, and a total turnover in sales of daily consumer goods of around ECU 200 million. Although Stockmann does not carry the Kesko logos, it has entered into K retailer agreements with Kesko and use Kesko as its main supplier. Stockmann also sells private label product of Kesko and Tuko. Further, there is an agreement between Kesko and Stockmann on cross-ownership.

88. Wihuri Oy is another privately owned company. It operates two large Eurospar outlets and 114 smaller Ruokavarasto outlets in south-western Finland. In 1995 the total sales of these amounted to between ECU 150 and 200 million. Wihuri was one of the companies that originally founded Tuko and it owned a large block of shares in the company until, as part of the concentration presently assessed, this was sold to Kesko. Even though Wihuri is also active as a wholesaler, Tuko is Wihuri's main supplier of daily consumer goods, and Wihuri has concluded a long-term supply contract with Tuko.

89. Sentra Oy runs one large Eurospar outlet and 27 smaller Prima and Rabatti daily consumer outlets, all in the area of Uusimaa in southern Finland. Its total turnover is around ECU 100 million. As Wihuri, Sentra used to be a large shareholder in Tuko. Moreover, Tuko is still Sentra's main supplier of daily consumer goods, including private label products, and Sentra has concluded a long-term supply contract with Tuko. In addition, Sentra leases almost half of its business premises from Tuko. These lease contracts are connected to the abovementioned supply contract.

90. Sentra and Wihuri, together with Tuko and Oy Hallman (a company whose outlets are part of the Spar chain operated by Tuko), own Eurospar Oy, which is the franchisor of the Eurospar trademark in Finland.

91. The investigation has revealed that Stockmann is commonly referred to as being part of the Kesko group, and that Wihuri and Sentra are similarly referred to as being part of the Tuko group (see paragraph 97).

(iv) Other companies on the market

92. According to Nielsen Finland (Newsletter 1/96) there were 1 286 other private daily consumer-goods shops in Finland in 1995. Many of the shops in this category have been using Tuko, and to some extent Kesko, as their main supplier. The combined market share of this group was just over 6 %, which represented a decrease by 0,7 % units since 1994.

II. Market shares

(a) National level

93. Even before the acquisition of Tuko by Kesko, the Finnish retail of daily consumer goods was one of the most concentrated in the European Union. Only in the Netherlands, Denmark, and Sweden does any single group of daily consumer goods retailers have more than 30 % of the national market, and the median figure for all Member States is of the order of 18 % (7). Following the concentration the K block would account for at least 55 %, see paragraph 98, of all sales of daily consumer goods in Finland. National sales of daily consumer goods in 1995 are illustrated in Table 1.

>TABLE>

94. As can be seen from the above table the combined national sales of Kesko and Tuko was almost three times larger than that of their largest competitor, SOK. According to the Nielsen register, the market share of the companies indicated in Table 1, and in particular that of Kesko, has been stable over the last 10 years. The four main groups have reported their actual sales in daily consumer goods to the Commission. These figures indicate that the total size of the market may be about 15 % smaller than that reported by Nielsen, which in its turn might indicate that the figures reported to Nielsen include some sales of other products as well. However, as all four groups have reported lower actual figures to the Commission, the use of these figures does not alter the relative positions of Kesko and Tuko vis-à-vis SOK and Tradeka/Elanto.

(i) Kesko's argument on market shares

95. In its reply to the Commission's statement pursuant to Article 18, Kesko has argued that the combined market share of Kesko and Tuko at retail level would not be 60 %, but 49 %. Kesko claims that this finding can be supported by more recent figures from Nielsen, relating to retail sales in weeks 1 to 34 of 1996, and by deducting the market share of four retail organizations (Stockmann, Sentra, Wihuri and Tarmo), which although included in the Nielsen figures for Kesko and Tuko respectively have weaker links to them.

(ii) Appropriate reference period

96. First, the Nielsen data on which Kesko bases its allegations applies a different methodology (e.g. by including kiosks) from the Nielsen data previously supplied to the Commission. Kesko has not explained the reasons for this. Secondly, figures representing only 34 selected weeks cannot be presumed to have the same degree of reliability as figures representing a full year, which would fully take into account existing seasonal variations. In addition, the particular period chosen by Kesko reflects an abnormal period for Kesko itself, as it was undergoing major reorganizational measures over this period. As regards Tuko, the period chosen by Kesko will also show abnormal results as it represents the period when Kesko was in the process of taking over the ownership of Tuko. In view of this, any assessment of the possible effects of excluding the four abovementioned retail organizations have to take as its point of departure the abovementioned figures for 1995.

(iii) Inclusion of Stockmann, Wihuri and Sentra

97. According to Nielsen, the indicated data has been collected by means of telephone interviews, where each outlet is asked to identify whether it belongs to one of the four main groups, or should be included in the group 'others`. It is thus clear that all four groups have been included in the Nielsen register due to their own statements to that effect. Moreover, as Kesko (as well as Tuko) are on the consultative committee of Nielsen's register of retailers of daily consumer goods, it is therefore also clear that Kesko and Tuko have considered it appropriate to include the four retail organization in their respective figures.

98. Given the above, and the existing links between Kesko/Tuko and each of the four retail groups as described above (Stockmann's K retailer agreement and cross-share holding with Kesko (see paragraph 87), Tuko's ownership of business premises used by Sentra (see paragraphs 89 and 90), Tuko's active role behind the operations of the Tarmo chain (see paragraph 76)), the Commission would continue to be of the opinion that Kesko, post-merger, would have a significant influence over these retail groups, especially as their choice of wholesale supplier, which until now has mainly consisted of Kesko and Tuko, would in effect be removed by the operation. It is, however, not necessary to finally decide on this question as, using sales figures for 1995 supplied by Stockmann, Wihuri and Sentra (8), even excluding these companies from the market share of Kesko and Tuko, the combined market share would still exceed 55 % on a national basis. Thus even if these companies were to be excluded from Kesko's post-merger market share, it would still be almost three times as high as that of its closest competitor, SOK. Similarly, the effect of excluding these four groups would have very limited effect on a regional level (see paragraph 101).

(b) Regional level

99. If the sales are broken down to the 16 statistical areas used by Nielsen to monitor retail sales in Finland, the combined Kesko/Tuko share varies between [40 and 90] %. In all such areas the overlap between Kesko and Tuko is significant, as Tuko's share of the sales varies between [10 and 30] %. At the same time SOK's sales represent between [0 and 40] % in these areas, and the combined sales of Tradeka/Elanto vary between [0 and 20] %. The share of the group 'others` varies between [0 and 20] %. (These regional statistics are provided as Annex 1 (9)).

100. The four major groups have supplied their actual sales figures in each of the 16 areas used by Nielsen. Again, the actual figures are lower than those used by Nielsen. The result is, however, very similar. Although the Commission does not have information on the actual sales of the group 'others`, the exclusion of this group seems, if anything, to increase the combined Kesko/Tuko share vis-à-vis SOK and Tradeka/Elanto.

101. As has been explained above, see paragraph 95, Kesko has argued that the sales of Stockmann, Sentra, Wihuri and Tarmo should be excluded from its retail market share. Although the Commission continues to be of the opinion that Kesko, post-merger, would have a significant influence over these companies, this question can be left open as their exclusion would not significantly affect the assessment. On a regional level the three first mentioned retail organizations are only active in five out of 16 regions in Finland. Their exclusion could therefore only have an impact in these regions. However, even in these five regions the combined market share of Kesko and Tuko would still be between [50 and 70] % even if the three organizations were to be excluded. Moreover, in all of these five regions Kesko's post-merger market share would be at least three times higher than the second largest competitor. The effect of excluding the Tarmo chain would similarly have very minor effects as this chain, according to Kesko, has a national market share of [. . .] % spread over about 500 outlets all over Finland.

(c) Local level

102. The Nielsen statistics are available also on a very local level, where Finland is divided into 439 areas. According to SOK, these areas are too small to even represent catchment areas, at least for larger outlets. The Commission does not have access to relevant turnover figures at this level for Stockmann, Wihuri and Sentra, as these have been included in the figures of Kesko and Tuko respectively. Given the limited activity of these three companies as explained above (see paragraphs 97, 98 and 101), it can be concluded that this does not significantly affect the reliability of the figures presented below.

103. The Nielsen statistics show that the combined Kesko and Tuko share of the sales is over [. . .] % in 294 (67 %) of the 439 areas, and that the combined sales exceeds [. . .] % in 49 of the areas. These statistics also indicate that there are 104 areas where either Kesko or Tuko achieved no sales. Out of those 104 areas, Kesko or Tuko achieve singly more than [. . .] % of the sales in 44 areas. Finally, the statistics show that there were 12 areas where neither Kesko nor Tuko achieved any sales. In view of the widespread nature of their operations, it can be presumed that the areas where there is no overlap between Kesko and Tuko cover a very minor part of the Finnish population.

104. As an example of the function of these local statistics, the Uusimaa area, which is the region surrounding Helsinki, is broken down in these statistics to 35 different areas. It is clear that these 35 areas are too small to represent catchment areas in this region since, inter alia, they fail to take into account that a major part of the population in Uusimaa work in the central regions of Helsinki and thus have a good opportunity to do their shopping in the close vicinity of their place of work, or while driving between home and work. In any case, since these local areas might provide an indication of the combined strength of Kesko and Tuko at a very local level, the statistical figures for Uusimaa is presented below. The Uusimaa area should, if anything, understate the position of Kesko and Tuko since the Helsinki area is the most densely populated area where the number of competitors is the greatest (e.g. Elanto is only active there).

105. Within the Uusimaa area the combined Kesko and Tuko share of the sales is over [. . .] % in 29 of the 35 local areas. It is over [. . .] % in 23 of the areas, and over [. . .] % in 17. Finally, in 12 areas their share is over [. . .] %, and in three areas it is over [. . .] %.

III. Other factors

106. The combined market share of Kesko and Tuko of at least 55 %, whether assessed at local, regional or national level, creates a presumption of dominance. This presumption is further strengthened by Kesko's and Tuko's position as regards large retail outlets, their control of a significant part of all business premises suited for retail outlets in Finland, their customer loyalty schemes and private label products, their distribution systems and, not least, by the position of their central organs as buyers of daily consumer goods. All these factors, which contribute to raising significant barriers to entry on the Finnish markets for retail and wholesale of daily consumer goods will be examined below.

(i) Large retail outlets

107. The largest unit outlets, the hypermarkets, are normally located in off-centre locations and provide free parking space for their customers. Hypermarkets are generally considered to have significantly greater catchment areas than smaller outlets. Such hypermarkets are of particular importance, because the activity of a hypermarket will have a competitive impact on a larger number of other outlets than the smaller outlets.

108. There is a clear trend in the Finnish retail market for daily consumer goods that the hypermarkets are gaining sales at the expense of smaller retail outlets. In 1995, sales of hypermarkets grew by over ECU 140 million according to the 'Kehittyvä Kauppa` magazine, published by the K retailer association. It is notable that Kesko and Tuko accounted for two-thirds of the total increase in the hypermarkets. At the same time sales of all other outlets fell, with the greatest fall in terms of value (minus ECU 194 million) experienced by outlets with selling areas below 400 m².

109. In the Kehittyvä Kauppa magazine, hypermarkets are defined as outlets with a selling area exceeding 2 500 m². In a study conducted by KPMG Management Consulting (10) on the Finnish market for retail of daily consumer goods, hypermarkets in Finland are defined as outlets with a selling area exceeding 1 000 m².

110. Kesko argues that the definition of large retail outlets with a sales area exceeding 1 000 m² is probably out of date, and that the figure 2 500 m² is more appropriate. However, even Kesko itself has been inconsistent in this respect and used various figures to divide stores according to their size. In a study on price dispersion attached to Kesko's written reply to the Commission's Statement of Objections, Kesko itself refers to 1 000 m². Furthermore, at the oral hearing held in the case on 8 October 1996, Kesko's presentation material used a sales area for daily consumer goods of 1 500 m² for hypermarkets. Moreover, in its 1994 annual report (page 33), Kesko considers both K superstores and Citymarkets as 'large retail units`.

111. Furthermore, the Commission has received information on the average selling area of the outlets in the different chains operated by Kesko, Tuko and their competitors. According to this information, none of the retail chains operating in Finland has an average selling area in excess of 2 500 m² (Kesko's Citymarket chain being closest with an average selling area of [. . .] m²).

112. For the reasons mentioned above, and as the Finnish population is rather small and dispersed, it is not unreasonable to assess the advantages resulting from having a strong position as regards large retail outlets on the basis of selling areas of 1 000 m².

113. Account being taken of outlets with a selling area exceeding 1 000 m², it is clear that the position of Kesko and Tuko is very strong. This is illustrated in Table 2.

>TABLE>

114. Apart from the fact that the average selling area of all outlets indicated in Table 2 is above 1 000 m², these outlets also share common characteristics as regards the average number of items sold ( > 5 000) and their average yearly turnover ( > ECU 5 million). Within the four main groups, the chain that 'comes closest` to qualifying to these criteria is the Tuko Spar Anttila outlets, which apart from having an average selling area of 905 m² are well above the other criteria. The second largest SOK outlets, the S-markets, of which there are 260 in Finland, are well below all three criteria. These outlets are instead in the same category as the 226 Kesko K markets. The second largest Tradeka/Elanto outlets, the Valintatalo chain, are even smaller.

115. In view of the above, and given the trend in Finland toward the growing importance of larger outlets, the stronger position of Kesko and Tuko regarding these outlets will increase further the dominance created by their combined market shares as described above.

(ii) Control of business premises suitable for retail outlets

116. The possibility to acquire or lease business premises suitable for the operation of retail of daily consumer goods is of vital importance for the profitability of such operations. Both Kesko and Tuko have over a long period built up extensive experience in the area of business premises management. As has been explained above, Kesko and Tuko own a significant share of all business premises within their respective group. In addition, Tuko owns 19 business premises operated by competitors. It should be noted that the percentage of business premises that are owned by Kesko and Tuko is especially high as regards the larger retail outlets, which are of key importance to the group. Ownership of these business premises guarantees that these locations will remain within the group, even if the retailer presently operating them was to terminate his business operations.

117. The setting up of a retail outlet for daily consumer goods requires a number of permits from local authorities, the most important of which is to get the approval of altering the city plan. According to the competitors of Kesko and Tuko, such approvals are getting more and more difficult to acquire, especially for larger outlets. In areas which have not yet been city planned, which is often the case for locations suitable for hypermarkets, a company wanting to set up such an outlet has to start the whole process of getting the local authorities to draw up a city plan. Such a procedure can take up to three years in Finland. Moreover, the administrative procedure involved in the city planning allows for third parties to appeal against the setting up of new retail outlets. According to one competitor, Kesko has on several occasions used this possibility to delay its setting up of new retail outlets. If an appeal is lodged, the whole process can take up to five years.

118. Both Kesko and Tuko control a significant number of business premises suited for the retail of daily consumer goods, and have extensive experience in the management of such premises. This will further serve to maintain and increase the strength of the dominant position held by Kesko and Tuko on the retail market.

(iii) Customer loyalty schemes

119. Kesko operates the very successful K-advantage card, which exists as a charge or a credit card. In total, there are close to 600 000 K-advantage cards issued. Tuko has, so far not been very active in this area, as Tuko has issued just over 2 000 Sparcards.

120. The K-advantage card can be used in a total of about 3 000 shops in Finland, including all K shops (i.e. supermarkets and all speciality goods outlets). In addition to creating incentives for customer loyalty, these cards also serve as an important marketing tool for Kesko. Through the information received on customers' purchasing habits, marketing can be timed and directed to target groups.

121. Kesko has argued that the K block has the same number of or fewer cardholders than its competitors and that several persons in a household can have cards. This argument is, however, irrelevant as the Commission is not assessing Kesko's position in a market for loyalty cards. The key issue relating to such cards is the potential to use the flow of information regarding customer's purchase habits. Taking into consideration the very high market share of Kesko following the acquisition of Tuko, the possibility of Kesko to use such information will clearly be improved and will be at a level which none of its competitors can match.

122. Kesko states further in its reply to the Commission's Statement of Objections that information on consumers' purchasing habits cannot be used because of the diversity of the computer systems of the K retailers. However, in its 1994 annual report Kesko declares its decision 'to provide the K stores with up-to-date information systems suitable for use in the K network`, and to make a large investment in the K retailers' information systems and to lease equipment to the K retailers. Further, in its 1995 annual report Kesko announces that two new companies 'Tietokesko Oy and K-linkki Oy, which were organized into subsidiaries and are responsible for Kesko's and the K stores' information technology`, started their operations on 1 September 1995. The function of K-linkki Oy is 'to provide the K-Group's retail stores with information technology services appropriate for chain operations`.

123. In addition to the abovementioned, the chain agreements between Kesko and the K retailer further state that 'The competitive advantage of the chain is strengthened by compatible information systems and by the centralized decision-making of the chain's retailers effected on the basis of the information received therefrom`. The same agreement states further that 'The retailer's information systems are in accordance with the chain's needs and with the K group's information systems architecture, as well as compatible with the systems of the chain. The retailer uses the chain's common product grouping and keeps the updatings and codes of his information system up-to-date all the time.` Moreover, under the collaboration agreement between Kesko and the K retailer, Kesko offers to the retailer certain services, including planning of information system solutions as well as giving training and advice in their use.

124. It is clear from the above, that the development and compatibility of the K retailers' and Kesko's information systems, for, for example, the common use of information, play an important role in Kesko's recent reorganization.

125. Following the acquisition of Tuko, the coverage of the K-advantage cards could be expanded to cover also the Tuko outlets. Since Tuko has issued relatively few Sparcards, many customers used to shopping in the Tuko outlets would be expected to acquire such cards. This potential increase in the number of issued K-advantage cards, when realized, will give Kesko a significant advantage in terms of knowledge of customer purchasing habits, and allow its marketing efforts to increase the loyalty of its customers.

(iv) Private brands

126. Kesko has stated that sales in the different chains sales of private label products (Pirkka) accounted for between [. . .] % of all sales in 1995. In addition, Kesko sells, jointly with ICA of Sweden and Hakon Gruppen of Norway, products under other private labels (Diva and Menu). According to information received by the Commission, Kesko's private labels account for 20 to 30 % of its total sales in many product groups. In the Tuko Spar chain [. . .] % of total sales were accounted for by its Eldorado products. It is unclear whether products sold under the Spar and Smart brands are included in this figure indicated for Tuko. In addition, as has been mentioned earlier, the private brand products of Kesko and Tuko are sold by Stockmann and possibly others. None of the competitors of Kesko and Tuko have managed to establish such successful private brand products.

127. According to the 1995 annual report of the Spar organization, private label products are a vital element of the product mix and continue to account for a continuing proportion of total sales. Further, according to the same source, private-label products are used to encourage consumers to make repeat visits.

128. Although it is probably necessary for retailers to stock at least one premium brand product in each product group, private label products offer several advantages to Kesko and Tuko in their retail operations. One such advantage is that these products are not sold by other retailers, which are perceived of as competitors. Although private-label products are generally sold at lower retail price than branded products, the margins on these products can in fact be set at a higher level compared to branded products, which are available from competing retailers as well.

129. However, according to producers of branded products, private label products of Kesko and Tuko are used as a negotiation tool vis-à-vis their suppliers (see below).

130. In conclusion, the strong position of the private label products of Kesko and Tuko will provide advantages in terms of customer loyalty and possibly the ability to price a greater proportion of its sales without having the need to take into account the reaction of its competitors in the same way as with branded products.

(v) Distribution systems

131. Kesko has a subsidiary, Kesped Oy, active in, inter alia, transportation of daily consumer goods. Tuko's transport activities are carried out by Finnfrost Oy, a joint venture between Tuko and Inex Partners. According to information received by the Commission, these two systems offer the only available distribution systems for frozen foods in Finland.

132. The acquisition of Tuko will therefore increase Kesko's possibilities to influence the conditions under which daily consumer goods are transported in Finland. This will increase the dependency vis-à-vis Kesko of competing third party retailers, industrial households and, at least, small and medium-sized producers of frozen foods that use the transport services of either of the two companies. As frozen products constitute an important and integral part of the basket of daily consumer goods, Kesko's control over this distribution system will constitute a significant barrier to entry on the retail market.

(vi) Advantages resulting from increased buying power

133. The position of Kesko following the acquisition of Tuko, where an agreement with Kesko would be the only way for a supplier to guarantee shelf space in retail outlets representing at least 55 % of the Finnish market will create a unique purchasing power in the hands of Kesko. This will be further strengthened by their position on the cash-and-carry market (see also below paragraphs 146 to 153). Moreover, the strength created by this gate-keeper effect would be further increased by the combination of the successful Kesko and Tuko private label products, which could be used as a further negotiating tool vis-à-vis suppliers to achieve additional concessions on, inter alia, reduced prices or increased marketing support.

134. Post-merger, Kesko will thus be able to obtain lower prices from producers to an extent that none of its competitors will be able to match. This will act as a further dis-incentive for Kesko's competitors to actively compete, in particular, on prices against Kesko. It will also make new entry on the market more difficult and as such act as a significant barrier to entry on the market.

135. Therefore, Kesko's purchasing power would allow it the ability to act, to a significant extent, independently of its competitors on the retail market.

(vii) Conclusion

136. Following the acquisition of Tuko, Kesko would hold a market share of at least 55 % of the Finnish market for retail of daily consumer goods, whether assessed at local, regional or national level. This position will be further strengthened by the special factors explained above, which would raise barriers to entry and make it more difficult for existing competitors to compete actively on the market.

137. These strengths of Kesko makes it extremely unlikely that SOK, Tradeka/Elanto or any of the smaller private competitors of Kesko would be able to restrict Kesko's behaviour in the market in any significant way, and in particular if it were to increase consumer prices. The Commission's investigation shows that Kesko's competitors are fully aware of the dominant position that would be held by Kesko following the acquisition, as well as of the several ways in which Kesko could use this position against them.

138. For the above reasons, the Commission has come to the conclusion that the concentration would lead to the creation or strengthening of a dominant position on the Finnish market for retail of daily consumer goods through which effective competition would be significantly impeded.

2. Cash-and-carry market

I. Structure of supply in cash-and-carry market

139. The Finnish market for cash-and-carry sales of daily consumer goods is very concentrated. In this respect, it should be noted that a large majority of cash and carry customers are captive to this form of distribution (see paragraphs 24 to 31). Kesko and Tuko operate 26 and 30 such outlets respectively in Finland. Neither Inex Partners or its sister company, Meira Nova, is active in the cash-and-carry business. Apart from the outlets operated by Kesko and Tuko, there are only 11 other cash-and-carry outlets in Finland, seven of them operated by Wihuri, and two each operated by Heinon Tukku and Veljekset Halonen Oy (Halonen). The geographic dispersion of cash-and-carry outlets operated by these companies is illustrated in Table 3.

>TABLE>

140. It should be noted that all three of the local cash-and-carry companies have links to Tuko, since Tuko invoices part of the sales to their customers. In addition, Tuko is the main supplier of Wihuri's retail operation.

II. Market shares

141. The total cash-and-carry sales of Kesko in Finland in 1995 was ECU 295 million, [. . .] % of which was made up from sales to industrial households. Over the same period, Tuko achieved total sales of ECU 197 million, [. . .] % of which was made up from sales to industrial households. Thus, the combined sales of Kesko and Tuko on this market was ECU 492 million. This was almost four times as much as the combined sales of the only other three companies active in the cash-and-carry market. Measured at national level, the combined market share of Kesko and Tuko would reach almost 80 %.

142. However, given that the cash-and-carry customers have to transport the purchased goods themselves, the national market share of Kesko and Tuko cannot provide more than an indication of their combined strength within this field.

143. On a regional level, i.e. within the statistical areas indicated in Table 3, the combined market share of Kesko and Tuko would exceed [. . .] % in all areas. In fact, the combined market share would exceed [. . .] % in all areas except in Uusimaa, where Heinon Tukku has a relatively strong position. Out of the 16 regions, Kesko would be the only cash-and-carry operator in nine following the concentration. In seven of these nine areas, this position is created by the acquisition of Tuko.

III. Conclusion

144. The acquisition of Tuko by Kesko would create a dominant supply structure on the Finnish market for cash-and-carry sales of daily consumer goods. The combined market share would be between [50 and 100] % in all regions. In addition this position would be strengthened by Kesko's buying power in the same way as explained above concerning the retail market (see paragraphs 133 to 135). It is therefore extremely unlikely that Wihuri, Heinon Tukku or Halonen, which are all linked to Tuko, would be able to restrict Kesko's actions if it were to increase prices. The investigation has shown that the local cash-and-carry companies would have very limited possibilities of restricting Kesko from abusing its strong position following the concentration.

145. For the above reasons, the Commission has come to the conclusion that the concentration would lead to the creation or strengthening of a dominant position on the Finnish market for cash and carry of daily consumer goods through which effective competition would be significantly impeded.

3. Markets for procurement of daily consumer goods

146. The most important distribution channel for daily consumer goods is clearly through retail supermarkets. As stated above (paragraph 106) the combined national share of Kesko and Tuko of the retail market for daily consumer goods is at least 55 %. The second most important channel is sales at cash-and-carry outlets. In this market, the combined national market share of Kesko and Tuko is about 80 %. Kesko's and Tuko's total sales in the cash-and-carry market equal more than 25 % of their retail sales. This guarantees Kesko an extremely powerful negotiating position vis-à-vis the producers of daily consumer goods.

147. Although distribution channels other than supermarkets, e.g. pharmacies, are available to some producers, mainly in the non-food sector it is clear that distribution channels other than those dominated by Kesko and Tuko are not viable alternative distribution channels for the majority of producers.

148. Kesko has stated that as far as central invoicing is concerned, 'all terms and prices of products are negotiated directly between the supplier and the retailer`. However, the Commission's investigation strongly indicates that most or all negotiations between producers and Kesko involve personnel employed by Kesko and its associated companies, rather than K retailers. This applies to negotiations both on product prices and marketing support fees. The latter are particularly important since in Finland they are the means by which producers grant concessions to important customers. In any event the Commission maintains its contention in paragraph 52 that producers perceive Kesko as an integrated entity irrespective of contact point or sales channel. Therefore, a producer will feel himself to be in a weaker negotiating position if he is convinced that his negotiating stance will become known throughout the K block.

149. In a wider sense, the very economic rationale of the K block is increased purchasing power, a fact proclaimed in Kesko's 1995 annual report where it is stated that the aim of its recent reorganization is to increase Kesko's purchasing power.

150. As an indication of the magnitude of the buying power of Kesko and Tuko, the majority of the suppliers (including several major multinationals) who replied to the Commission's investigation indicated that they depend on Kesko and Tuko for approximately 50 to 75 % of their total sales in Finland. The dependency of different suppliers will differ according to the nature and size of their business and consumer perception regarding their products. Thus, although some very large producers of highly regarded brand products may have some countervailing power vis-à-vis Kesko, it will be of vital importance for most small and medium-sized producers to maintain sales through Kesko at the present level.

151. Kesko will not be dependent on any individual supplier to the same extent. Kesko has provided a series of statistics purporting to show its dependence on producers in terms of percentages of individual products purchased from individual producers. However, the percentages are not particularly high, exceeding 50 % for one supplier in only four of 10 product groups cited. In any event, for any one product group Kesko will normally be able to switch from one supplier to another, since there will nearly always be an alternative large producer with adequate capacity to meet increased short-term demand. The producers, on the other hand, will not have a similar possibility to switch from Kesko post-merger, since Kesko's retailing competitors will not have the capacity (with only about 40 % of the market) and additional retailing capacity can only be brought on-stream in the medium or even long term.

152. Moreover, recent developments in retailing will enhance the buying power of the merged entity. In particular, private label development is a key element in the power wielded by retailers vis-à-vis branded daily consumer-goods producers. It enables retailers, who are inevitably privy to commercially sensitive details regarding the branded goods producers' product launches and promotional strategies, to act as competitors as well as key customers of the producers. This privileged position increases the leverage enjoyed by retailers over branded-goods producers. Both Kesko and Tuko already market successful private label products. Again, a new dimension in retailers' control over information is emerging with the increased use of store payment cards and 'loyalty` cards, which enable retailers to make targeted appeals to individual consumers based on known patterns of past purchases and socio-demographic profiles. The potential combination and expansion of Kesko's and Tuko's card schemes will enhance their negotiating power vis-à-vis producers, who do not enjoy the same immediate access to information as consumer behaviour.

153. For these reasons, and given that the concentration would lead to dominant positions on the retail and cash-and-carry markets, the Commission is of the opinion that the increased buying power of Kesko would further reinforce the dominant position of Kesko on the retail and cash-and-carry markets. In particular Kesko would be able to use its buying power to employ different strategies, the long-term effects of which would be to further weaken the position of its competitors.

C. POTENTIAL COMPETITION

154. There has not been any foreign new entry to any of the markets discussed above. It is very likely that one of the main reasons for this is that Kesko even before the acquisition of Tuko enjoyed a very strong buying position as the most important customer of the majority of all producers of daily consumer goods. It would also seem that Finland's geographic position reduces the scope for most potential entrants in achieving economies of scale and/or scope from an expansion of business operations to Finland.

155. Whereas the acquisition of Tuko might have offered a window of opportunity for entering the Finnish market, there are in effect no other companies in Finland, the acquisition of which could be used to achieve a similar result. It can thus be presumed that any foreign company wishing to enter the Finnish market will have to be prepared to invest significant amounts in finding and developing new retail outlets, marketing of a new brand name, etc. Such investments would to a large extent not be recoverable if the entry showed to be unsuccessful.

156. According to Kesko, food retailing is undergoing a process of internationalization, demonstrated by the fact that several western European companies have expanded their operations outside their country of origin. In view of this the Commission has contacted all companies mentioned by Kesko as potential entrants on the Finnish markets, as well as several companies known to have expanded retail operations abroad. According to these companies, the most important criteria in assessing whether to enter a new market relate to existing competition, availability of local partners (suppliers, wholesalers) and access to competitively priced goods, access to sufficient number of good store locations and local spending power.

157. The strengthening of Kesko's position on the markets described above makes it very doubtful that a potential new entrant would find any of the abovementioned criteria fulfilled. Prior to the concentration, a new entrant could have established a cooperation with Tuko as a wholesaler. Following the concentration, a new entrant would most certainly find it very difficult to find local wholesalers and/or suppliers to cooperate with and, in particular, to secure supplies at prices comparable to those of Kesko. A potential new entrant would experience difficulties in securing access to store location, given, inter alia, the existing possibilities under Finnish law to appeal city planning decisions (see paragraph 117).

158. The increase in Kesko's buying power will constitute one of the most significant barriers to foreign entry, since one of the key questions for any potential entrant will be to secure access to daily consumer goods at prices which allows such an entrant to effectively compete with Kesko. This would apply also to the market for cash-and-carry sales.

159. Even in the hypothetical scenario where a foreign company would try to enter the Finnish market for retail of daily consumer goods, it is unlikely that such an entry would have any impact on the position of Kesko. This can be demonstrated by an example from Sweden, a market that in many respects has similar characteristics as the Finnish market. The Norwegian company Rema 1000, stated by Kesko as one of the most likely entrants to the Finnish market, did enter the Swedish market in 1991 with their discount store concept. According to the company itself, its market share in Sweden, after five years of operations, is below 1 %.

160. Theoretically, the most logical potential entrants on the Finnish market would seem to be companies with strong positions in Sweden or Norway, since such companies are situated closest to the Finnish market both in terms of geography and in market structure. However, Kesko has extensive cooperation with the Swedish market leader, ICA, as well as with the Norwegian market leader, Hakon-gruppen. This cooperation includes joint purchasing and cooperation concerning private labels products, e.g. 'Diva`. Similarly, Tuko has extensive cooperation with Swedish Dagab, Danish Dagrofa and Norwegian Unil, which also involves the joint ownership of the 'Eldorado` private labels.

161. In conclusion, in the past there has been no entry on any of the markets where Kesko would hold a dominant position following the concentration. It is also likely that Kesko, when acquiring Tuko, had realized that the acquisition of Tuko would serve as a window of opportunity to enter the Finnish market at reasonably high risk. It has therefore to be concluded that the concentration increases barriers to entry, and that it is extremely unlikely that any entry would be possible following completion of the acquisition.

VII. UNDERTAKINGS PROPOSED BY KESKO

162. By letter of 23 October 1996, Kesko submitted proposals for undertakings in order to remove the doubts raised as to the compatibility of the concentration with the common market. Kesko's proposal contained undertakings to:

(1) sell about 55 % of Tuko's previous activities relating to wholesaling of daily consumer goods and all Tuko's Cash-and-Carry outlets; and

(2) sell [. . .] retail outlets ([. . .] Spar stores owned by Tuko and [. . .]).

Kesko proposed to complete the said undertakings before the end of year 1997.

163. According to Kesko the proposed divestiture would form the basis for the creation of a new block of retailers on the market (Newco). In addition to the [. . .] retail stores mentioned above, this new 'block` would consist of four retail groups; Stockmann, Wihuri, Sentra and Tarmo (see paragraphs 76 and 87 to 91). The combined market share of Newco would, according to Kesko, be about 11 %.

Market position of Newco

164. Newco would consist of five different retailing organizations (each with a national market share of about 2 %). Of these five organizations only the [. . .] previous Tuko and [. . .] would be legally incorporated in Newco. The other four organizations are Stockmann, Wihuri, Sentra and Tarmo, which are a heterogenous group of retailing organizations, whose connections to Newco would be weaker than they were with Kesko/Tuko before the concentration and which would only consist of sales agreements. Moreover, Newco would be regionally skewed, since most of its outlets would be in the south-west of Finland.

165. Newco would be at a competitive disadvantage vis-á-vis, in particular, Kesko since it would be very weak as regards large size outlets and since it would not have any joint private label or customer loyalty scheme. Newco would also be at a competitive disadvantage as concerns logistics and distribution to its outlets. Furthermore, the majority of the [. . .] Spar stores that would be sold to Newco have not been operating profitably. Lastly, Newco would have a competitive disadvantage as concerns purchasing as it would not be able to conclude purchasing agreements for more than, at most, one-fifth of the volume of Kesko. This would also have a negative impact on Newco's cash-and-carry activities.

166. In view of the above none of the representatives of producers, competitors or customers contacted by the Commission were of the opinion that Newco would be able to act as an integrated competitor in the medium to long term. Moreover, all companies contacted replied that they would not be interested in acquiring Newco.

Market position of Kesko following the proposed measures

167. In spite of Kesko's proposal, the concentration would still mean that the K retailers and the Spar retailers, which constitute the essential elements in the concentration, would be combined in the same organization, with a market share approaching 50 %. The [. . .] Spar stores to be divested would represent only some 9 % of Tuko's turnover at retail level. The corresponding figure for the [. . .] would be some 2 %.

168. In this respect it has to be borne in mind that, prior to the concentration, Kesko already had the highest national market share of any retailer of daily consumer goods in the EU, and faced competition from, mainly, Tuko and SOK, each with a market share of approximately 20 %. Even if Kesko's proposal were to be implemented, Kesko would have a market share 2,5 times larger than that of SOK. Moreover, Newco would, for the reasons indicated above, be a much weaker competitor than Tuko has been in the past.

169. All representatives of producers, competitors or customers contacted by the Commission were of the opinion that, since the concentration would still significantly increase Kesko's market share as well as reduce the competitive strength of the other retailers on the market, the operation would, despite the proposed remedies, result in a dominant position. One competitor indicated that, due to the inherent weakness of Newco, it is likely that Kesko would be able to regain most of the market share seemingly transferred to Newco in the medium term.

Overall comment

170. The main competitive concerns identified by the Commission concerns the retail market, which is the main market, in terms of turnover as well as direct impact on Finnish consumers. For the reasons stated above, Kesko's proposal is clearly insufficient to deal with the problems identified on this market.

171. Given that Kesko has not offered any undertaking able to address the problems identified on the retail market, this would mean that the problems identified would remain on the retail market. It is also doubtful whether the proposal could remove the competitive concerns on the cash-and-carry market, since Newco would not have the same possibility as Tuko to compete with Kesko concerning the achievement of advantageous purchasing conditions.

172. Therefore, it has to be concluded that the remedies proposed by Kesko are not capable of removing the dominant position, in particular on the retail market for daily consumer goods in Finland. Therefore, the concentration cannot be authorized on the basis of the proposals made by Kesko.

VIII. CONCLUSION

173. As a result of the concentration, Kesko Oy would acquire a dominant position on the Finnish markets for retail of daily consumer goods and cash-and-carry sales of daily consumer goods. As the concentration has already been implemented, the Commission will, in a separate decision based on Article 8 (4) of Regulation (EEC) No 4064/89, adopt appropriate measures in order to restore conditions of effective competition,

HAS ADOPTED THIS DECISION:

Article 1

The concentration by which Kesko Oy acquired sole control of Tuko Oy by purchase of shares is declared incompatible with the common market and with the functioning of the EEA Agreement.

Article 2

This Decision is addressed to:

Kesko Oy

Satamakatu 3

FIN-00160 Helsinki.

Done at Brussels, 20 November 1996.

For the Commission

Karel VAN MIERT

Member of the Commission

(1) OJ No L 395, 30. 12. 1989, p. 1; corrected version OJ No L 257, 21. 9. 1990, p. 13.

(2) OJ No C 130, 26. 4. 1997.

(3) See judgment of the Court of Justice of the EC in Case 56/65, Société Technique Minière v. Maschinenbau Ulm, [1966] ECR, p. 235.

(4) For example Case No IV/M.242 - Promodes/BRMC.

(5) According to the Rimi Store chain agreement, the Board can also fix prices for the basic selection.

(6) In the published version of this Decision, some information has been omitted or replaced by ranges, pursuant to Article 17 (2) of Regulation (EEC) No 4064/89 concerning non-disclosure of business secrets.

(7) According to statistics provided by AIM (1995).

(8) According to Kesko, the national market share of the Tarmo stores is [. . .] %.

(9) Annex witheld from public version for confidentiality reasons.

(10) 'Päivittäistavarakaupan kansainvälinen tehokkuusvertailu`, Helsinki 25 October 1991.