EN

Case No COMP/M.2662 - DANISH CROWN / STEFF-HOULBERG

Only the English text is available and authentic.

REGULATION (EEC) No 4064/89 MERGER PROCEDURE

Article 9 Date: 14/02/2002

COMMISSION OF THE EUROPEAN COMMUNITIES

Brussels, 14/02/2002

In the published version of this decision, some information has been omitted pursuant to Article 17(2) of Council Regulation (EEC) No 4064/89 concerning non-disclosure of business secrets and other confidential information. The omissions are shown thus [Ö]. Where possible the information omitted has been replaced by ranges of figures or a general description.

PUBLIC VERSION

MERGER PROCEDURE ARTICLE 9(3)(b) DECISION

Decision relating to the referral of the Case No COMP/M.2662 Danish Crown/Steff-Houlberg to Danish Competition Authority, pursuant to Article 9 of Regulation 4064/89

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Having regard to the Treaty establishing the European Community,

Having regard to Council Regulation (EEC) No. 4064/89 of 21 December 1989, on the control of concentrations between undertakings, as amended by Council Regulation (EC) No 1310/97 of 30 June 1997(together, ëthe Merger Regulationí), and in particular article 9(3) thereof,

Having regard to the notification made by Danish Crown AmbA and Steff-Houlberg Amba on 21 December 2001, pursuant to article 4 of the said Regulation,

Whereas:

1.1. On 21 December 2001 the Commission received a notification of a proposed concentration by which the co-operative Danish Crown AmbA merges with the co-operative Steff-Houlberg AmbA.

2.2. By a letter dated 28 December 2001 as supplemented by a letter dated 8 February 2002 the Danish Competition Authority requested the proposed concentration to be

1OJ L 395, 30.12.1989, p.1; corrected version OJ L 257, 21.9.1990, p.13

2OJ L 180, 9.7.1997, p.1; corrigendum OJ L 40, 13.12.1998, p.17

Commission europÈenne, B-1049 Bruxelles / Europese Commissie, B-1049 Brussel - Belgium. Telephone: (32-2) 299 11 11.

referred to the competent Danish Competition Authority with a view to assessing it under Danish national competition law, pursuant to article 9(2)(a) of the Merger Regulation (the ìRequestî).

I THE PARTIES

3.3. Danish Crown AmbA (Danish Crown) is the largest Danish co-operative slaughterhouse and has in total 20,525 members (farmers) who supply live animals to the co-operative. Danish Crown is active in the slaughter of pigs and cattle, meat trading and meat processing. Its main activities in the EU are in the UK, Denmark, Germany, France and Italy. Danish Crown is also active worldwide including USA, Japan and Eastern Europe. Danish Crown is the largest slaughterhouse in Europe.

4.4. Steff-Houlberg Amba (Steff-Houlberg) is the second largest co-operative slaughterhouse in Denmark. Steff-Houlberg has 2,366 members who supply live pigs to the slaughterhouse. The co-operative activities cover the slaughtering of pigs, purchasing of fresh beef, meat processing and meat trading. Its main activities in the EU are in the UK, Denmark, Sweden, Germany, France and Italy. Steff-Houlberg is the seventh largest slaughterhouse in Europe.

5.5. The parties are also active in the collection of abattoir by-products through Daka Amba, the processing and sale of natural casings from animals slaughtered through DAT-Schaub, and the supply of spices and food ingredients, packaging material, workersí clothes and auxiliary tools for the meat and food industry through SFK. DAT-Schaub and SFK were already before the merger controlled by Danish Crown.

6.6. Danish Crown is also active in the sale of hides from cattle through Scan-Hide and operates a marginal business slaughtering of lams and horses.

II THE OPERATION

7.7. On 28 November 2001 the Boards of Representatives of Danish Crown and Steff-Houlberg approved the proposed merger. The operation is conditional upon approval from all relevant competition authorities. Following the operation, Steff-Houlberg will cease to exist as a legal entity; Danish Crown will take over all its activities, assets and liabilities. The merged entity will be controlled by 23,000 members of whom no individual members will have a controlling interest. Control will be exercised through the democratically elected decision-making bodies of the merged co-operative ñ the Board of Representatives and the Board of Directors.

2.1 Concentration

8.8. The proposed concentration is a full merger within the meaning of Article 3(1)(a) of the Merger Regulation by which Danish Crown merges with Steff-Houlberg.

2.2 Community dimension

9.9. The undertakings concerned have a combined aggregate world-wide turnover of more than Ä5 billion (Danish Crown Ä5.4 billion, Steff-Houlberg Ä709 million) and each

3Turnover calculated in accordance with Article 5(1) of the Merger Regulation and the Commission Notice on the calculation of turnover (OJ C66, 2.3.1998, p. 25). To the extent that figures include turnover for

10.have a Community-wide turnover in excess of Ä250 million (Danish Crown Ä3.5 billion, Steff-Houlberg Ä609 million), but they do not achieve more than two-thirds of their aggregate Community-wide turnover within one and the same Member State. The notified operation therefore has a Community dimension.

2.3 Background

The Co-operative Structure

11.10. Danish Crown and Steff-Houlberg are organised as co-operatives. This implies a structure based on the principle that all costs and profits are shared between the members. However, the economic responsibility of each farmer towards the co-operative is limited according to the Articles of Association. The yearly profit of the co-operative is split between the members as a residual payment based on their supply to the co-operative. For the financial year 2000/2001, the residual payment for Danish Crown was around Ä0.13 per kilogram of pig meat delivered and it was Ä0.04 higher than that of Steff-Houlberg. The residual payment is only paid to members of the co-operative. However, the co-operative is able to buy pigs from both co-operative members and non-members. The three large Danish co-operatives Danish Crown, Steff-Houlberg and Tican account for 95% of pig slaughtering in Denmark.

12.11. The members of the co-operative are the farmers. They are both suppliers and co-owners of the co-operative. The largest members will supply significantly below 1 % of total supplies to the merged entity. The farmers both own and control the slaughterhouses processing their products.

13.12. Danish co-operatives are in general not subject to any specific statutory company law. The relevant provisions governing the activities of co-operatives are laid down in their Articles of Association. In general, members have an obligation to supply all their animals intended for slaughtering to the co-operative. The co-operative likewise has an obligation to buy 100% of the memberís production. However, due to the Commissionís decision in case COMP/M. 1313 Danish Crown/Vestjyske Slagterier (the ìPrevious Caseî), members of Danish Crown are allowed to supply up to 15% of their weekly supply to other slaughterhouses as set out in Article 5.4 of the Articles of Association, (the split delivery rule). To make use of this rule the member must notify Danish Crown in writing six weeks in advance, and must give Danish Crown three month notice in writing if it wishes to cease making use of the split deliveries. The split delivery rule took effect on 1 October 2001.

14.13. Every farmer can become a member of a co-operative for a particular species (cattle or pigs) by giving three months notice. In order to leave a co-operative the member must also give notice. The notice period for Danish Crown is 12 months and that of Steff-Houlberg 24 months.

III THE REQUEST FOR REFERRAL

the period before 1.1.1999, they are calculated on the basis of average ECU exchange rates and translated into EUR on a one-for-one basis.

4Commissionís decision COMP/M.1313 of 9 March 1999, Official Journal L 020, 25/01/2000 p.1-36.

15.14. By letter of the 28 December 2001 as supplemented by a letter dated 8 February 2002 the Danish Competition Authority, according to Article 9(2)(a) of the Merger Regulation, requested the Commission to refer the case to Denmark. On a preliminary assessment the Danish Competition Authority has identified five product markets, where, in its view, the merger threatens to create or strengthen a dominant position:

1)1) The purchase of live pigs for slaughtering

2)2) The sale of fresh pork for human consumption

3)3) The sale of fresh pork for further processing

4)4) The production and sale of processed meat for human consumption, and

5)5) Collection of abattoir by-products.

16.15. Furthermore, the Danish Competition Authority has pointed to the fact that the co-operation of Danish slaughterhouses in Danske Slagterier must be taken into account. The Danish Competition Authority contend that these markets are distinct markets and all limited to Denmark.

IV ASSESSMENT PURSUANT TO ARTICLE 9 OF THE MERGER REGULATION

17.16. As noted above, the Danish Competition Authority has made a request under Article 9 paragraph (2)(a) seeking a referral of the notified concentration on the basis that the concentration in question threatens to create or to strengthen a dominant position as a result of which effective competition will be significantly impeded on a market within that Member State, which presents all the characteristics of a distinct market.

18.17. Pursuant to Article 9(3) of the Merger Regulation, the Commission assesses whether, having regard to the market for the products or services in question and the geographical reference market, there is such a distinct market within the Member State seeking the referral and that such a threat to competition exists. The Commission has then the possibility to either itself deal with the case or refer whole or part of the case to the competent authorities of the Member State concerned with a view to the application of that Stateís national competition law.

19.18. The Commission has therefore sought to assess whether the relevant product markets identified by the Danish Competition Authority in the Request present the characteristics of distinct markets within Denmark and whether the concentration would threaten to create or strengthen a dominant position as a result of which effective competition will be significantly impeded on the markets in question. Furthermore, the Commission has investigated whether further markets exist in Denmark or abroad in order to assess the possibility and scope of referral.

Market definition

20.19. The Commission has previously analysed the Danish pork industry in the Previous Case. In the Previous Case the Commission identified the following relevant product markets: 1) the purchase of live pigs and cattle, respectively, for slaughtering constitutes two separate relevant product markets. 2) The market for fresh pork and beef, respectively, can be subdivided into sales to industrial processors and for human

21.consumption. 3) The market for fresh pork for human consumption can be further subdivided into retail and catering as distinct markets. In the retail sector fresh meat can be subdivided into sales to the final consumer through supermarkets, butchers and farm sales. However, it was left open whether supermarkets and butchers constitute distinct markets. 4) Finally, the Commission defined a market as regards the abattoir-by products consisting of the high-risk material.

20. As noted above, the Danish Competition Authority has identified the following relevant markets: 1) The purchase of live pigs for slaughtering, 2) The sale of fresh pork for human consumption, 3) The sale of fresh pork for further processing, 4) The production and sale of processed meat for human consumption, and 5) Collection of abattoir by-products. The Commission considers that the following markets should be identified.

4.1 The purchase of live pigs for slaughtering

4.1.1 Product market

21.21. The main activity of the parties is the slaughtering of pigs and processing of pig meat. Danish Crown is active in the slaughtering of both pigs and cattle, whereas Steff-Houlberg is only active in the slaughtering of pigs. There is no supply-side substitutability between slaughtering pigs and cattle as it would take both time and cost for a farmer to shift production from the breeding of one species of animals to another. Furthermore, it is not possible to use the killing lines of other species to slaughter pigs without significant cost. The Commission therefore concludes that the purchase of live pigs for slaughtering is a relevant product market. This conclusion conforms to that in the Previous Case and is not contested by the parties.

4.1.2 Geographical market

22.22. In the Previous Case it was found, that the geographical market definition for the purchase of live pigs for slaughtering was at most Denmark. This was based on the fact that 90% of live pigs originate from within 120 km from the slaughterhouses. The co-operative buying obligations for Danish slaughterhouses and the supply arrangements for farmers were also thought to provide a significant obstacle to the export of Danish live pigs.

23.23. The parties argue that there are well-founded arguments supporting the view that the geographical scope of the market for the purchase of live pigs for slaughtering should be defined as Northern Europe. However, they have taken notice of the Commissionís view that the market is limited to Denmark and have chosen not to disagree in this case.

24.24. Notwithstanding this, the parties claim that price correlation analysis indicates that the market is larger than national. Based on price correlation the relevant geographic market would include Northern Europe (Denmark, Sweden, Northern Germany and the Netherlands). The parties have provided data showing the price correlation between Denmark, Germany, Sweden and Netherlands. The data show that correlation exists, but at the same time, major differences occur. Firstly, the fluctuation in correlation across time is including a time period of 4 years. That amount of data is not robust for a correlation test. Secondly, there exist big differences in correlation across Member States, which make the analysis impossible to rely on. An example is the correlation between German and Swedish pig prices. In 1998 the correlation was

5

25.0,85953, in 1999 0,47524, in 2000 0,90668 and in 2001 0,32614. Such variations are not compatible with the statement that the northern part of Germany and Sweden belong to the same Northern European market.

26.25. However, as argued in the Previous Case, the price correlation between Danish and foreign pig prices is based on how the price quotation is made. The pre-calculation system starting point is the sales departmentís assessment of the expected sale price of each individual cut. This could include the sale of tenderloin in Japan, the sale of ribs in the United States and the sale of pork loin in Denmark. The sales departments weekly assessment of sales prices is (after a deduction of a profit margin) used to calculate a ìcost priceî of each of the primal cuts: legs, fore-ends and middles. The same price information is used to calculate weekly quotations for the live pigs (weighted with the composition of different cuts made in each week, and adjusted in some cases due to market situation).

27.26. It can be concluded that the price of a slaughter pig is calculated based on sales on different national markets. However, it can not be concluded that the price is primarily based on the prices paid by other slaughterhouses to their suppliers. In conclusion, this does not indicate that the Danish market for live pigs is a part of a wider Northern European market for the purchase of live pigs for slaughter. In fact, the pre-calculation system is based on sales in different countries indicating that all markets are national in scope, based on Danish Crowns assessment of the possible sale to each individual market.

28.27. In addition, there has been relatively little change on the market since the Previous Case although transport distances may have increased due to the higher concentration of the market. Third parties have stated that the ìcatchment areaî, i.e. the distance from which slaughterhouses obtain their pigs, normally has a maximum distance of 250-300 km from the slaughterhouse. Pigs could be transported from further distances but this will only be the case in the event that the slaughterhouse is in serious shortage of pigs for slaughter.

29.28. Moreover, the co-operative structure is still considered to be a main factor discouraging exports and imports. Members of Steff-Houlberg are not allowed by the conditions set out in the Articles of Association to supply pigs to other slaughterhouses or for export. Whilst the undertaking given in the previous decision by Danish Crown concerning split deliveries (i.e. allowing members of a co-operative to deliver 15% of their weekly production of pigs to third-party slaughterhouses) was intended to address this problem it did not come into operation until 1 October 2001. Since then only [Ö] members out of a total of 20,525 of Danish Crown have used the possibility to make split deliveries primarily to private slaughterhouses in Denmark. Even if the 15% rule were extended to also include the members from Steff-Houlberg, it would have little effect on the overall market, based on the effect the rule has had on Danish Crown members.

30.29. The export of live pigs for slaughtering has remained low and amounts to 1.4% of the total Danish production of slaughter pigs and sows. This figure includes export from farmers outside the co-operatives, which account for the major part of the export, and export from the co-operatives. The third largest Danish co-operative Tican has not got the capacity to slaughter sows themselves, and has therefore exported their memberís sows to Germany. Third parties have informed the Commission, that there exists a shortage of Danish pigs for slaughter abroad, and this is due to the high percentage of pigs, that members have to deliver to the co-operatives.

31.30. The import of live slaughter pigs into Denmark is also negligible. In 1999 imports were 300 tonnes and in 1998 100 tonnes. The low level of import could be explained by the fact, that none of the three co-operatives (Danish Crown, Steff-Houlberg or Tican) import pigs for slaughter. This is partly due to the fact, that the three large co-operatives have an USDA authorisation allowing export of fresh pig meat to the USA. Furthermore, rules adopted by Danske Slagterier in order to avoid ìcontagious pig diseaseî make import difficult. The rules involve certain precautions to be observed. For example, the co-operatives should have facilities that allow foreign and local pigs to be separated. If the USDA-authorised slaughterhouse wishes to process meat of foreign origin, this must come from countries allowed to export to the US market.

32.31. The Commission therefore concludes that the geographical market for the purchase of live pigs for slaughtering is limited to Denmark. This geographic market presents, pursuant to Article 9(2)(a), all the characteristics of a distinct market.

4.2 The sale of fresh pork for human consumption

4.2.1 Product market

33.32. In the Previous Case the market for sale of fresh pork for human consumption was subdivided into catering, supermarkets, butchers and farm sales but supermarkets and butchers could be considered together. Based on this the retail market was split into sales form supermarkets, butchers and farmers (farm sales).

Retail and Catering

34.33. The distinction between the retail and catering markets was in the Previous Case based on the fact that the origin of the meat was less important for caterers than for retailers. There was no realistic possibility for arbitrage between the two markets in view of the differences in distribution and packaging. The retail market needs to be supplied at all times with the full assortment of different parts of fresh pork whereas caterers typically only need the cuts from the meals included on their menus. Finally the Commission found that wholesalers supplying caterers and retailers buy their meat cheaper than retail clients do.

35.34. The parties argue that the market for sale of fresh meat for human consumption should not be subdivided into retail and catering since there is full demand and supply-side substitutability. Firstly, the majority of the productsí characteristics are identical, both retailers and caterers buy large pieces of pork, which can be further cut in-house, likewise both retailers and caterers buy fresh pork in smaller pieces. Secondly, the parties have no possibility of discriminating between customers as many wholesalers are selling to both retailers and caterers. Thirdly the price-structure applied to retail and catering customers is almost identical. Fourthly, the arguments that retailerís need the full assortment and caterers do not is beside the point. And lastly, the parties submit that the product differentiation on the market for sale of fresh pork depends on

5 See section 4.3 for a discussion of the USDA regime.

36.how the carcasses are cut and that this can be changed immediately without any significant costs.

37.35. Whilst it is possible to find similar cuts which are sold to both retailers and caterers (the parties identified five such cuts) the Commission notes that retailers are far more likely to engage in in-house cutting than caterers. The parties stated in the notification that 85% of the volume supplied to supermarkets and 99% of that supplied to butchers required further cutting whereas only 20% of meat supplied to caterers would be cut further in-house. The demand for the majority of pieces does therefore not appear to be as similar as claimed by the parties.

38.36. The parties have identified nine wholesalers in Denmark who sell both to retailers and caterers, none of these are included in the partiesí list of the five largest Danish customers. Furthermore these companies account for less than [0-10]% of Danish Crownís Danish sales of fresh pork for human consumption by volume and less than [0-10]% of Steff-Houlbergís sales. It therefore appears that the parties are able to identify whether the customer is retail or a catering customer for their majority of sales. For example sales to the largest supermarkets account for [90-100]% of Steff-Houlbergís total Danish sales and over [70-80]% of Danish Crownís sales. Furthermore, Steff-Houlberg only has one catering customer. The Commission therefore does not regard minor sales to wholesalers who may supply both caterers and retailers as sufficient to ensure price competition between the two groups.

39.37. Danish Crown has one general basis price list for pork which does not differentiate between caterers, butchers and supermarkets. The parties argue that it is natural that prices will vary among different customers, but this will be decided by the bargaining power of each individual customers. Predictably the largest three supermarkets which have over 80% market share and account for over [70-80]% of Danish Crownís Danish sales top the list of customers receiving the largest discounts. Danish Crown claimed in the notification that it undertakes over [60-70]% of its sales both to retailers and to caterers at list price and therefore the price paid by the majority of customers is identical irrespective of whether they are retailers or caterers. On further questioning it materialised that only sales at campaign price were included in the remaining roughly [40-50]% not sold at list price and this figure did not include companies which obtained various discounts. If one were to consider these sales as being below the list price then over [90-100]% of sales are subject to various discounts. This suggests that the list price merely functions as a starting point for bargaining. Whilst the starting point is the same for both retail and catering, the discounts could vary quite considerably. The price list is thus not considered a relevant factor for considering the prices actually paid by customers in the retail and catering sectors and does not indicate that these customers should be considered part of the same market.

40.38. The proportion of non-Danish meat purchased by caterers is also much higher than that of retailers. Imports of fresh pork into Denmark have remained relatively stable in the last 3 years accounting for 11,500 tonnes in 2000/2001 or 9.7% of total volume. Of this around only 1,200 tonnes was bought by supermarkets accounting for approximately 1.4% of total sales whereas imports accounted for approximately 45% of catering sales. The origin of the meat therefore appears less important for caterers

than for retailers and it follows that there are more actual and potential suppliers. The conditions of competition would thus appear to be different for catering customers and retailers.

39.39. In view of the above the Commission, in line with the Previous Case concludes that the retail market should be considered separately from the catering market.

Supermarkets, butchers and farm sales

40.40. It is not necessary for the Commission to determine whether the retail market should be split further into butchers, supermarkets and farm sales. This is because the parties are not engaged in farm sales and their sales to butchers are small (less than [0-10]% of retail sales). It would make no difference to the analysis whether these were included in the definition or not.

4.2.2 Geographical market

Retail

41.41. In the Previous Case the geographical scope of the market was found to be Denmark. The Commission recognised that fresh pork is transported and traded internationally but found that barriers to imports into Denmark were too high for imports to have a competitive impact in the short-term on the Danish market for pork sold through supermarkets. Indeed imports accounted for less than 2% and consumer preferences for Danish pork were found to be high. This situation has not changed.

42.42. The parties are not convinced that this narrow definition is correct but have decided not to contest it in the notification. Despite this, the parties continue to provide a wide range of arguments of why the geographic scope of the market should be defined as wider than national.

43.43. The parties argue that the geographic scope of the market is EU-wide and maintain that there is high supply side substitutability. In particular it is maintained that barriers to import are not as high a stated in the Previous Case.

44.44. The parties informed the Commission that although there was high supply-side substitutability between fresh pork sold in different countries it was impossible to compare prices in different countries. Indeed the parties state in the notification that ëcomparable prices for fresh pork cannot be obtained. The best reflection of the movements in pork products is a price of movements for live pigsí. The difficulty in obtaining unbiased price statistics on meat at the retail or wholesale level is due to the fact, that the carcass can be cut in many different ways, and that the cuts thereby obtained can be further de-boned, de-fatted and trimmed in many different ways. Cutting specifications may vary between suppliers, customers and even individual butchers in the shop. However not only the value of each individual cut is influenced by cutting specifications, but the specifications of one cut affects the specifications and thereby value of all neighbouring cuts from the same carcass. Furthermore, it appears that the preferred cuts of meat also vary across national borders and therefore it is very difficult to find identical pieces of pork meat to compare.

9

45.45. Irrespective of these difficulties, the Commission asked the parties to make comparisons on four products (pork loin, pork collar, tenderloin and pork shoulder). For all but tenderloins the specifications differed between member states. These comparisons were said to show some correlation. However, the Commission does not share this view. It is by no means clear from the charts produced that prices were similar enough to suggest that there were restraining influence between countries. Indeed, several charts indicated quite large discrepancies.

46.46. The value per kg for all fresh pork supplied by the parties in the EU 2000/2001 was also quite different between countries. The Danish value is given as [Ä1.50-2.00] which is higher than the UK [Ä1.50-2.00] but significantly lower than Sweden [Ä3.50-4.00], Germany [Ä2.00-2.50] and the Netherlands [Ä2.00-2.50]. Over time it is also worth noting that whilst the value has increased for all these countries (apart from Germany) the increase is not in the same proportion.

47.47. A relevant geographic market comprises of the area in which the undertakings concerned are involved in the supply and demand of products and services, in which the conditions of competition are sufficiently homogeneous and which can be distinguished from neighbouring areas because the conditions of competition are appreciably different in those areas. If customers can buy at the same prices as customers located in other areas, such areas should be included in the geographic market definition. If, however, customers buying from suppliers located in other areas cannot buy at the same prices as the customers located in those areas, such areas should not be included in the geographic market definition. In other words, there can be situations where the fact that customers are being supplied from a plant in a certain area does not mean that these customers are getting the prevailing "market" prices in that area. This could, for instance, be the case when the suppliers deliver the products to the premises of the customers. If arbitrage between customers is not possible, suppliers can then charge different prices to customers in different areas. In a prospective analysis, as is carried out in investigations of concentrations, the possibility of future price discrimination by undertakings which through a concentration can achieve a very strong position among the suppliers that can supply to a certain area, may be a crucial factor in delineating the relevant geographic market to take into consideration.

7 Furthermore the parties explained that small changes in the mix of fresh meat products sale in the individual member states affects the average value per kg significantly. This was because the average figures include products varying in value from 4 to 40 DKK/kg.

49.49. On this basis and given in particular the inability of the parties themselves to compare prices of identical products and the differences in cuts between countries, the Commission considers that it is possible for a supplier of fresh pork for human consumption to price-discriminate between countries. A slaughterhouse will be able to identify which country it is selling to given that it must comply with the special cutting requirements of that country when cutting the meat. Furthermore, it is very unlikely that arbitrage between countries could occur. This is because firstly the product is cut to one particular country specification which may not be desired in the other country and secondly it would be very difficult for the retailer to identify any potential gains which could be made from arbitrage given the apparent difficulties in comparing products (and prices) between countries. Another obstacle to arbitrage may also arise from the desire for freshness and transportation costs.

50.50. As regards the Danish market in particular, the Commission has found that the reasons cited in the Previous Case for concluding that the market was not wider than Denmark remain valid. In particular, the Commission found that consumer preference for fresh pork of Danish origin remains high.

51.51. In light of the above the Commission therefore concludes that each national market should be considered separately taking account of the possibility of potential competition from imports in the competitive assessment. In particular the Commission concludes that there is a separate market for the supply of fresh pork for human consumption to retailers in Denmark. This market presents all the characteristics of a distinct market pursuant to Article 9(2)(a) of the Merger Regulation.

Catering

52.52. In the Previous Case the catering market was thought to be wider than Denmark but was left open as no competition concerns arose in this market. The same approach can be taken in this case. The parties have not defined the market for catering separately as they believe it should be considered in conjunction with the retail market.

4.3 The supply of fresh pork for further processing

4.3.1 Product market

53.53. Fresh pig meat (including frozen pig meat) is either sold for fresh meat consumption or sold to processors. In the Previous Case, the Commission concluded that fresh pork and fresh beef are separate relevant product markets for the purpose of the proposed merger. In the Previous Case, the Commission defined a separate product market for the sale of fresh pork and fresh beef meat to industrial processors. The parties do not contest this definition in the notification.

4.3.2 Geographic market

54.54. In the Previous Case, the Commissionís investigation showed that, in general, meat processors could and do source their meat regardless of its origin. The geographic market was, therefore wider than national. In the present case, the parties submit that the geographic market is not less than EU-wide. This is based on the large trade flows in the EU.

55.55. The market for fresh pork for further processing differs in a number of ways from the market for the supply of fresh pork for human consumption. Firstly, the market is characterised by larger trade flows across borders. Secondly, as the meat is further processed into a finished product before being sold to consumers, the origin of the meat seems to be of less importance to the final consumer.

56.56. In order to examine the magnitude of the market, the parties were asked to provide price information on pork products sold for further processing. However, according to the parties, a direct price comparison of the several hundred products for each party and each country is impossible to produce in practice. The parties claim, this is due to their pre-calculation system, as described in section 4.1.2.

57.57. Since the supply of fresh pork for further processing is produced according to buyersí specifications and since there is no possibility to compare the prices on different cuts, the market could be national in scope. However, imports of fresh pork for further processing do take place. The parties have not been able to estimate the size of such imports. Based on the statistics from Danske Slagterier the imports of pork carcasses and cuts for direct human consumption or further processing was 36,257 tonnes in 1999 and 43,874 tonnes in 2000. However, imports also include pork for direct human consumption and imports for the catering market. These import amounts to 11,200 tonnes in 1999/2000 and to 11,500 tonnes in 2000/2001. Based on a total market of 329,600 tonnes in 2000/2001 the import share could be estimated to be 9.8%. This figure has to be compared to the large Danish surplus of pig meat, where the majority is exported.

58.58. Some processors are importing fresh pork from abroad for further processing. The imports are primarily from Germany and other EU-countries. However, it is not possible to measure to which extent Danish processors use foreign pork in the supply of processed meat to the Danish market. This will be of importance in evaluating the down stream market of the supply of processed meat to the Danish market.

59.59. Despite the greater number of cross-border trade for pork for further processing compared to fresh pork for human consumption could indicate the market is wider than national, the Commission investigated whether consumer preferences for meat of a particular origin could extend to processed products. If this were true, processors would also need to purchase pork of particular origin to satisfy their consumersí demand. This is because the demand for pork for processing is derived from the downstream demand for processed meat products. The Commission found some evidence of consumer preference or regulatory rules, which make the origin of meat for processed products important.

60.60. Third parties claim that the consumers of Danish processed meat products have an expectation that Danish raw materials are used. This is perceived as a guarantee that high standards of hygiene and quality are maintained. Third parties further contend that stringent Danish regulatory demands aimed at protecting against salmonella infections also function as a barrier for foreign processors wishing to supply the Danish market.

61.61. An example of the importance of the porkís origin is the U.S. market. Some Danish slaughterhouses and processors are selling fresh pork (including frozen pork) products

on the US market. This requires the pig to be slaughtered in a slaughterhouse with USDA authorisation. In order to export to the United States, a foreign country must have a residue control program with standards equivalent to U.S. standards and the country must be free of certain animal diseases such as foot and mouth disease and swine fever etc. Export to the U.S. requires certificates to accompany imported meat, poultry and egg products to identify products by country and plants of origin, destination etc. Today, imports of fresh pork are only allowed into the US from the following EU-countries: Denmark, Finland, Ireland, UK (Northern Ireland) and Sweden. Furthermore Iceland and Norway are allowed to export to the U.S. market. Import is not, except from the mentioned countries, an alternative for the Danish producers of processed meat products that have the USA as export market. In order to maintain the right to export their products to the USA, the meat products must comply with the requirements set by the United States Department of Agriculture (USDA).

The USDA authorisation requires that the processors only use meat from other EU-authorised companies from countries authorised for US export. However, the meat has to origin from and be imported from countries which are US-authorised. Furthermore, during processing or storing the meat is not allowed to get in contact with meat that origin from a non-US authorised country. This means, those processors, processing foreign meat without US authorisation have to separate the processing of each meat product.

If processors cannot buy Danish fresh pork for further processing, they will not be able to export their products to the US market. In Denmark only Danish Crown, Steff-Houlberg, Tican and 2 private slaughterhouses including the former Koopmann have this USDA authorisation. Third parties have argued that the merger will reduce the number of suppliers of fresh pork with an USDA-authorisation.

The above data show that consumer preference for meat of specific origin, in particular Danish origin, can be a barrier for processors wishing to obtain their supply of pork from non-Danish slaughterhouses. However, the level of imports is not insignificant at 9.8%.

In the light of the foregoing, the Commissionís investigation has shown indications of a possible national market for the supply of fresh pork for further processing which would be limited to Denmark. If the market for the supply of fresh pork for further processing is to be regarded as wider than national, the concentration will not threaten to create or strengthen a dominant position. On the contrary, if the market is defined as Danish, there will be a threat to competition in the Danish market. Given that any competition concerns would be limited to Denmark, the Commission has reached the conclusion that the geographic market definition can be left open in the present case leaving the Danish Competition Authority to investigate this issue further.

4.4 The supply of processed meat

4.4.1 Product market

66.Processed meat products can be defined as meat from mammals or birds containing external ingredients such as salt or spices, being raw, dried smoked or cooked. This was the definition adopted in the Previous Case. The parties agree to this definition of the product market.

67.A relevant product market comprises all those products and/or services which are regarded as interchangeable or substitutable by the consumer by reason of the productsí characteristics, their prices and their intended use.

68.The various processed meat products vary in several dimensions such as the raw material used (pork, beef, poultry), ingredients (spices), water content, heat treatment (smoked or boiled), portion, packaging, temperature (chilled or canned). All processed meat products constitute a combination of this 7-dimension scheme.

69.From a demand side the products for this market satisfies a demand for a meal either as a full meal or meal ingredients (in a sandwich or salad). The parties contend, that the complete inter-substitutability of different meals and also on meals within or outside the home based on individual subjective criteria makes it without any meaning to try to narrow this market into segments. The Commission does not agree with this argument.

70.Processed meat products vary over a large range of different products from pre-packed bacon to ready-prepared dishes and components. From a demand side, consumers are not willing to substitute pork processed products with beef or poultry processed products. In addition, there is clearly no supply-side substitutability between pork-processed products and processed products using other meat. A supplier of bacon cannot obtain its supplies for raw pork from beef producers or poultry producers, as these suppliers are unable to switch their production. Thus, a wide range of the processed sandwich meat products produced for the Danish market are traditionally made with pork meat, such as most liver paste, salamis, cooked ham, fillet and cooked belly roll. Producing the products with meat other than pork is not a viable alternative. It is therefore reasonable to conclude that the product market for processed meats should be subdivided in processed pork products, processed beef products, processed poultry products etc. There is therefore a distinct product market for processed pork products.

71.This market should be further divided into particular processed products such as pork sausages or sliced bacon. From a demand side perspective consumers may be unwilling to substitute sausages with raw bacon at 5-10% permanent price increases. The characteristics of processed products make such substitutability unlikely. There is for example no demand substitutability, when making sandwiches (sm¯rrebr¯d), to substitute (cold cuts) ìpÂlÊgî with canned meat, since canned meat often requires further processing. P‚tÈ and liver paste may have different demand to cold cuts. Liver paste is in comparison to cold cuts made from different raw materials and it differs in taste and fat content from cold cuts, which normally are regarded as leaner. Furthermore, it could be argued that the consumers in Denmark regard liver paste, as a separate product market. The kilo price of cold cuts is normally higher than that of liver paste. The closest demand side substitute to liver paste is different p‚tÈs, which basically consist of the same raw materials.

72.Normally, supply-side substitutability would indicate the absence of separate sub-markets and would dictate a single market for all pork-processed products as a supplier of pork products could easily change his production to supply bacon, ham etc. However, in some instances supply-side substitutability may not change the above conclusion due to different consumer preferences and perceptions for different categories of products. The meatís origin may be more important when a consumer considers buying liver paste than ham (Danish origin may be particularly prized for traditional Danish delicacies). Thus, the meatís origin may be less important when buying a p‚tÈ, which is normally regarded as a foreign speciality, in comparison to buying traditional Danish delicacies.

73.Therefore, it may be necessary to sub-divide the processed pork market into separate product markets.

The parties have, on the request of the Commission, identified six categories of processed products, which may have some similarities. The categories are:

(a) Raw cured products

This group consists of products, which will be cooked by the consumer before eating. Some of the products may have received some heat treatment (i.e. by smoking) but will still by consumers be regarded as raw. The group consists of bacon in various types, slices and dices, and based on different parts of the carcass. Also larger pieces of ham and loin cured or smoked is included in this group.

(b) Processed meat for cold consumption (pÂlÊg)

This sub-group consists of products that will not be cooked by the consumer before eating, but will be consumed as it is ñ typically in slices or dices. Dried ham, cooked ham, ìk¯dp¯lseî and salami for sandwiches, which can also be used in salads and cold starters.

(c) Canned meat

The product group consist of fully preserved meats in selected metal cans or glasses which have a prolonged shelf life also at ambient temperatures. The product group includes canned ham, luncheon meat, cocktail sausages and semi-prepared meats with gravy and other ingredients.

(d) Sausages

The product group consists of sausages which are cooked and who are consumed directly as sausages typically after heating in water, frying or grilling. This segment also includes sausages for hotdog stands.

(e) P‚tÈs and pies

The group consists of pies, which are commuted and heat treated products (baked or boiled) often including ingredients such as liver, and in most cases spreadable. On the Danish sub-market liver paste is a large part of this sub-group.

(f) Ready-prepared dishes and components for such

This group ñ which also could be called convenience foods ñ consist of products which typically have a higher degree of preparation for consumption as a hot meal, either being the complete meal itself (i.e. a meat component and gravy, potatoes and vegetables) or being a significant component of this (such as meat balls ìfrikadellerî, burgers, schnitzels etc). Such products are often sold in frozen state but can also be sold as chilled products.

75.The parties have informed the Commission, that a large volume of processing is carried out in the individual supermarkets for sale in the deli-counter or at butcherís shops. However, it might be necessary to split the sub-product markets into customer groups, i.e. into to wholesale, catering and retail. A special category could include the sale of fast food including the sale from hotdog stands. This might constitute a separate product market different from the sale to final consumers through supermarkets.

76.The Commission considers that the above sub-divisions of the market are likely to be warranted, given the fact that these markets, as shown below, have a national dimension and the parties appear to have a strong market position therein (see below). An investigation of the precise effects of the merger on these markets is necessary.

4.4.2 Geographic market

In the Previous Case, as in the product market of pork for further processing, the geographic definition of the market for processed meat products was defined as wider than national. It was argued that consumer preferences seemed to be, less important than those relating to fresh meat are. Also the origin of the meat product was normally not indicated, and was therefore not known to the final consumer. This was unlike fresh pork products where origin was found to be important.

78.The parties maintain that the market for production and sale of processed meat products (which the parties define as including processed products from all types of meat such as pork, beef, poultry etc.) is at least EU-wide.

79.In order to assess the geographic market in the present case, the parties were requested to provide price data from different national markets, where they sell their processed meat products. This, they claimed, was not possible due to the fact, that the products sold on the different markets are not comparable, because of differentiation in products. The parties claim, due to differentiation in products (raw material dimension, fat content, water content etc.) it is very difficult to identify similar products, which are sold in the member states. The parties stressed those small differences in fat or water content will make the price fluctuate and provide misleading figures. The parties provided the Commission with an example of Danish Crownís direct (variable) production cost for cooked sliced ham. The index figure show that the cost of ham supplied to UK retail is index 100 compared to index 72 for cooked ham supplied to a major Danish retail chain. In addition there are many products, which are only produced for sale in one country, since the products differs in spicing, presentation and other minor process features.

80.The parties were however able to produce price comparisons on a few products including sliced bacon. Even these comparisons were said to involve products, which were not sufficiently similar to allow a true comparison. These (incomplete) price comparisons however show that considerable price variations exist on different markets.

81.The Commission disputes based on the information supplied by the parties, that the geographic market is EU-wide. However, it can not be ruled out that there are markets for individual product groups within the processed meat category, which are more wide ranging than others are. This may be due to different consumer preferences and perceptions for different categories of products. However, within each product category the manufacturer of a product will differentiate the product based on the preferences of the national market where the product is sold. This may be by simply adding more water or fat or it may be a separate recipe altogether supplied with reference to specific national consumer preferences.

Furthermore, some processed meat products are only sold on one national market. As above for fresh pork this allows the suppliers to differentiate between different national markets. There are separate identifiable markets with different tastes and preferences, which can be identified at the point of sale, and arbitrage is unlikely to take place, as the products do not satisfy the taste or other requirements of neighbouring markets. This indicates that the markets are national in scope.

The importance of consumer preferences is, however, difficult to assess. Virtually no import of sausages is taking place in Denmark. This is because the Danish origin of some products such as sausages seems important to Danish consumers. Danish retailers confirmed this. Furthermore, Danish sausages have a ìmÊrkevareî brand effect appealing to Danish consumers.

In conclusion, the geographic scope of an overall product market for the supply of pork-processed products (possibly split into the sub-markets described above) is considered as national in scope on the basis of the suppliersí ability to differentiate between different Member States. This is based on the fact that products are produced with reference to the specific national consumer preferences. In addition, as regards pork processed products, Danish origin appears to be an important factor for consumers in Denmark. The geographic market therefore presents pursuant to Article 9(2)(a) all the characteristics of a distinct market confined to Denmark.

4.5 Collection of abattoir by products in Denmark

4.5.1 Product market

85.The parties have identified a relevant product market for the collection of high-risk abattoir by-products. This activity concerns the collection, processing and marketing of inedible animal by-products from farms and the meat industry, especially from the slaughterhouses. The by-products are converted into different products such as bone meal, meat meal, animal fat etc. The parties are active in this market through Daka a co-operative involved in the collection of abattoir by products and high-risk material. Daka has recently acquired Kambas, formerly an integrated division of Steff-Houlberg.

86.Some abattoir by-products are high-risk materials, which may only be supplied to a high-risk renderer. The rendering plant must be licensed to that purpose according to Community legislation.

The relevant product market is the market for the collection of high-risk abattoir by-products since this material may only be supplied to a high-risk renderer. This is in conformity with the Previous Case and the parties agree to this definition.

4.5.2 Geographic market

9According to a third party a very limited import of sausages has taken place in 2000, to compensate for a deficit on the Danish market, due to shortage of sausages of Danish origin. The import was produced especially according to Danish consumer preferences.

Danish legislation prohibits cross-border transport of high-risk material and the market is therefore limited to Denmark by virtue of legal provisions. The parties agree that the relevant geographic market is limited to Denmark. As a result, the market presents all the characteristics of a distinct market confined to Denmark.

V. COMPETITIVE ASSESSMENT

5.1 The purchase of live pigs for slaughtering in Denmark

89.The parties estimate that they will have a market share of around 90% (74.1% Danish Crown, 15.4% Steff-Houlberg) of the Danish market for slaughtering of live pigs The proposed merger will increase the already strong position of Danish Crown by 15.4% and at the same time eliminate the next largest competitor in Denmark. The market shares on the Danish market for purchase of live pigs can be illustrated as follows:

Table 1: Purchase of live pigs for slaughtering. Market shares and number of pigs 2000.

2000

Market share Heads (%) 74.1

Heads

15,689,097

15.4

3,270,099

Danish Crown/Steff-Houlberg

89.5

18,959,196

Tican

5.5

1,155,474

Private slaughterhouses

3.7

780,033

Slaughterhouses abroad

1.3

282,808

Total

100.0

21,177,511

Actual competition

90.As shown in Table 1, the only actual competitors to the merging parties are Tican and the private slaughterhouses. Tican is the next largest competitor with 5.5% and its capacity is fully utilised and would be unable to accept any new members. It has a waiting list with farmers wanting to join the co-operative. In addition, the location and structure of Ticanís slaughterhouse does not allow for an increase in capacity. The largest private slaughterhouses slaughter around 200,000 pigs yearly. This is only a fraction of the amount of pigs slaughtered by Danish Crown.

91.Tican and the private slaughterhouses, will not be effective competitors to the merged entity. Private slaughterhouses have some spare capacity but will not be able to compete with the parties, as they do not achieve anywhere near the same capacity or economies of scale. The annual number of pigs slaughtered by the former Koopmann, for example, amounted to 200,000 pigs. Danish Crown slaughters 15,689,097 pigs annually.

92.The are some economies of scale in the slaughtering of pigs and the limited size of private slaughterhouses makes it impossible to engage in fierce competition with the merging parties on down stream markets. None of the private slaughterhouses have the necessary facilities (slaughterhouses, processing units, distribution facilities) to engage in competition with the parties. The high concentration of the Danish retail sector implies that a private slaughterhouse needs to concentrate on niche product markets and to follow the parties in price setting. Because of their small scale, no private slaughterhouse will be able to compete on price for the larger retail contracts.

93.Instead Tican and the private slaughterhouses follow a strategy based on local supplies, and Tican is also exporting a large proportion of its pig meat.

94.In addition, it is expected that the market share of the parties will rise in 2002. This is due to the fact that the private slaughterhouse Koopmann went bankrupt on 8 January 2002 (Koopmann purchased 200.000 live pigs in 2001). Koopmann bought a slaughterhouse from Danish Crown, based on Danish Crownís commitments in case IV/M.1313. The commitment to sell slaughter capacity to a competitor (Koopmann) to the parties (Danish Crown/Vestjyske Slagterier) and to Steff-Houlberg ñ in the market for sale of fresh pork through supermarkets ñ was designed to avoid the creation of a dominant position by Danish Crown/Vestjyske Slagterier by creating an important third alternative. This third alternative now appears to have disappeared and the proposed merger will lead to single firm dominance.

Split deliveries

95.It should be noted that the Statues of Association of Steff-Houlberg prohibit members to deliver pigs to other slaughterhouses. This makes it impossible for the members of Steff-Houlberg to supply other slaughterhouses in Denmark or abroad. Following the undertaking in the Previous Case members of Danish Crown have only recently been effectively able to deliver 15% of their weekly volume to third parties (split deliveries). However, the impact of the split delivery rule has been insignificant. So far only [<1]% of the members of Danish Crown have used the possibility. The rule is also expected to comprise members of Steff-Houlberg when joining Danish Crown. Even then, based on the present effect, it seems highly unlikely, that the split delivery rule will have any significant effect on the market.

96.The main method for the merged entityís members to deliver elsewhere is the split delivery rule. The possibility to make split deliveries to foreign slaughterhouses is however practically limited to the large members supplying a substantial amount of pigs a week and farmers outside the co-operatives. This is because a farmer must deliver a full truckload to minimise the transport costs. If a farmer, for example, decides that his minimum number of pigs for profitable delivery is 200 then this farmer must produce around 1300 pigs a week to make the split deliveries viable. Among the 10 largest suppliers of Danish Crown and Steff-Houlberg, only 2 farmers seem to supply more than 900 pigs a week. If a farmer takes up the possibility to make split deliveries, the farmer will not receive any residual payment for the pigs delivered to third parties. Furthermore, if the member is not able to deliver his full split supply to third parties, and instead delivers it to the co-operative, he will not receive any residual payment for the quantity reserved for split delivery. This places a large risk on the farmer wishing to deliver to other slaughterhouses.

Potential competition - exports

97.In theory foreign slaughterhouses could compete with the parties. However, more than 80% (taking into account the 15% rule) of the production from members from co-operative slaughterhouses is reserved for the Danish co-operatives, due to the obligation to supply according to the Articles of Association of the co-operatives. For the moment no competition exists between Danish co-operatives and foreign slaughterhouses. This is reflected in the small export and import of live pigs.

98.Exports of live pigs also pose some obstacles. For example, it is demanded that special veterinary rules be followed which include the obligation to ensure that a veterinarian supervises loading. In practice, due mainly to this veterinary rule and due to transport cost, it is only possible for private Danish slaughterhouses to receive split deliveries. As mentioned, foreign slaughterhouses are further limited in obtaining supplies of live pigs due to distance and transport costs.

Conclusion

99.It can be concluded, that post-merger Danish Crown will be purchasing around 90% of pigs slaughtered in Denmark. Farmers would have next to no choice of where to deliver their pigs (partly because the lack of other co-operatives with spare capacity or private slaughterhouses, and partly because of their supply obligation to the co-operative and the existing obstacles to exports).

The Commission concludes that the concentration threatens to create or strengthen a dominant position as a result of which effective competition would be significantly impeded on the market for purchase of live pigs for slaughtering in Denmark.

5.2 The sale of fresh pork for human consumption in Denmark

5.2.1 Retail

100.The new entity will combine the two largest suppliers of fresh pork to the retail market in Denmark with a combined market share of [50-70]%. The parties are actually the two closest competitors on the Danish market. By merging, the competition between the parties will be eliminated. For example, the parties have been engaging in head-on competition for supermarket contracts. The two largest supermarket contracts (Coop Danmark and Dansk Supermarked) lost by Steff-Houlberg in the last three years were won by Danish Crown. The parties have also given examples of where Steff-Houlberg won contracts from Danish Crown including the retail sellers Magasin and Rema 1000. It is clear that the merger will eliminate such competition in the future.

Table 2: The sale of fresh pork for human consumption to retailers, market shares and volume 2000.

Retail (2000)

Market share volume (%) Volume in 1,000 tonnes product weight [Ö]

[40-50]%

[10-20]%

[Ö]

Danish Crown/Steff-Houlberg

[50-70]%

[Ö]

Total

100

95.3

101. Despite the geographic scope of the market being defined as Denmark, when assessing the position of the parties within the Danish market, the Commission takes into account actual and potential competition which the parties face from suppliers outside the Danish market.

Actual competition

102.At present there is little actual competition from outside Denmark, as shown by the very low level of imports into the retail sector. The import of fresh pork has been very low constituting only 1.3% of the Danish retail market. Therefore when considering actual competition only alternative suppliers within Denmark are considered.

103.The smaller private slaughterhouses and Tican are operating in Denmark and the parties claim that they provide an effective competitive constraint. However, the merged entity will become by far the largest supplier to the retail market, it will be better placed to supply retailers through its extensive distribution network and its ability to supply a full product range. No other Danish supplier could supply the volume and breadth of products required by the supermarkets. If the supermarkets turn to these suppliers it is highly unlikely that these could be engaged on more than a regional basis. Furthermore, the merged entity will be far stronger financially than its competitors and none of the small private slaughterhouses including Tican would be able to engage in price competition with the merged entity due to its small size and capacity constrains. The existence of much smaller competitors in Denmark can not be relied upon as sufficient competitive restraint on a company, which controls around 90% of pigs, slaughtered in Denmark.

Potential Competition

104.To assess potential competition one must consider whether alternative suppliers would emerge if the merged entity were to increase its prices/decrease its product quality or service in Denmark. The parties argue in the notification that there is a high degree of supply side substitutability since slaughterhouses can readily change the way in which they are cutting carcasses to alter the supply of various of fresh pork meat cuts so that a foreign supplier could change its supplies to satisfy Danish demand. The Commission should therefore consider potential competition from slaughterhouses outside Denmark as a restraining factor on the merged entity.

105.In addition, the parties maintain that there are no barriers to import and focus on three issues identified by the Commission in the Previous Case as entry barriers: (a) access to distribution facilities, (b) the Danish DT 104 salmonella test, and (c) national preferences for Danish pork.

(a) Access to distribution facilities

106.The parties claimed that there no major distribution-related barriers to import. Any supplier could enter Denmark and negotiate an agreement on distribution with third parties such as independent hauliers. Besides the parties argue that some of the big Danish supermarkets distribute a significant part of the fresh pork themselves.

107.In the Previous Case, the Commission found that distribution posed some concern: "In Denmark pork sold through supermarkets is with few exceptions distributed directly from the slaughterhouses to the individual stores through the distribution systems of the parties. Amongst the Danish supermarkets only FDB (now Coop Danmark) would currently have the logistical facilities in terms of refrigeration capacity and trucks to distribute pork from its distribution centres. According to the parties, FDB at present undertakes own distribution for about [Ö]% of the pork sold

in FDB stores via its own distribution centres, while the remainder is accounted for by direct distribution from the slaughterhouses to the individual stores. Consequently, in Denmark the overwhelming volume of pork sold through supermarkets is distributed directly from the Danish slaughterhouses to the individual stores through the distribution systems of the slaughterhouses. The supermarkets, therefore, depend to a large extend on the slaughterhouses for distribution of fresh pork".

108.Today, the largest Danish retailer Coop Danmark receives [BUSINESS SECRET] of its fresh pork supplies from the parties distributed directly to its shops. Only [BUSINESS SECRET] is distributed by Coop Danmark itself.

109.As described in the Previous Case: "In terms of distribution the most obvious possibility for a non-Danish slaughterhouse to have its meat distributed in Denmark would be to deliver it to the distribution centres of major Danish supermarket chains. However, as set out above, the Danish supermarkets mainly rely on a system of direct distribution from the slaughterhouse to the individual stores. In reality, therefore, this possibility does not exist. For imports to have an impact on the market, distribution would have to be arranged. This would imply either the creation of a distribution system or the creation of additional distribution facilities by the supermarkets".

110.The parties have argued that there exists several hauliers available on the market to distribute pork from a non-Danish slaughterhouse. Furthermore, third parties could arrange distribution through for example Arla Foods, which already have a network distributing dairy products. However, the system is mainly focused on dairy products. Furthermore, Arla is presently to a large extent distributing fresh pork from the parties and it is uncertain, whether Arla has the capacity to distribute fresh pork form foreign slaughterhouses. As far as the wholesalers are concerned, those distribution facilities are more scattered and limited.

(b) The Danish DT 104 salmonella test

111.The parties argue that the control is carried out on a non-discriminatory monitoring basis comparable to the testing of Danish products and applies to both Danish and foreign suppliers. In addition the risk of product withdrawal is small and applies equally to Danish suppliers as to importers.

112.In the Previous Case the Commission found that: "The additional veterinary control for salmonella (DT 104) on imported pork makes it less attractive for non-Danish slaughterhouses to export to Denmark. More importantly, the main potential exporters such as Germany and the Netherlands currently are less advanced in actions aimed at a general reduction of all types of salmonella and they do not have specific veterinary controls for salmonella DT 104. Obviously, this increases the costs of exports to Denmark compared to exports to other countries. Furthermore, it involves an increased economic risk compared to that borne by Danish slaughterhouses, where general salmonella reduction plans are well advanced and specific DT 104 tests are routinely done. The importers thus face, in general, a greater risk that they will not pass those controls with the consequence that the whole shipment of pork, which may

already to some extent have been distributed to the stores, has to be withdrawn. The Swedish producers are quite confident that they have salmonella under control in their production. However, the largest Swedish producer, Swedish Meats, has stated at the oral Hearing that its potential exports to Denmark will be niche products in view of the fact that their production costs are higher than those of Danish producers".

113.Since the Previous Case, no change seems to have taken place, and the salmonella control is still regarded as a barrier to entry. Although, the salmonella control is not to be regarded as an insurmountable barrier to entry.

(c) National preferences for Danish pork

114.The parties submit that, whilst there still are some national preferences among Danish consumers there is no reliable evidence to sufficiently justify a separate Danish market for pork meat for human consumption. In any event the barrier to import is so low that potential imports constitute a strong potential competition. The parties also contend that the degree of national preferences for the end consumer would depend upon the price and that a small but permanent price increase would lead to an increased demand for slightly cheaper imported of products. Furthermore the differences in customer preferences between countries do not imply that imports are not possible.

115.The Commissionís market investigation, however, confirmed that Danish consumers still have strong preferences for pork of Danish origin. Most third parties said that there exist very strong consumer preferences for Danish pork which are based on the slaughter methods, the size of the pig, ability to deliver and the quality of the pork. Danish quality was perceived to be higher that of importers. There is also a perception that consumers would not accept foreign meat on the basis that so much is produced in Denmark. This is also confirmed by the fact that the retail sale of foreign pork is almost non-existing.

116.In addition to consumer preferences for Danish pork there is the branded pork which the Commission views as a section of the market where non-Danish suppliers may not be able to compete effectively with the parties even if they were to overcome the very strong consumer preferences described above. On the Danish market, the parties estimate that around 15% of fresh pork for human consumption (approximately 18,000 tonnes) are branded. Of this it is estimated that [0-10]% is sold under the partiesí brand and [0-10]% is sold as retailers private label. The brands are special types of pig meat that are typically of higher quality or originating from pigs reared under certain specified conditions, e.g. organic pigs. There are several different types of branded pigs in Denmark such as Antonius (Steff-Houlberg) which is a Danish pig of high quality, which is breed under conditions, which are preferable to those of non-branded pigs. Frilands Gris (Danish Crown) is also a high quality pig which has been bread outside. The Gourmet Gris is the branded pig of FDB, itís label states ëBedre trivsel, Dansk, bedre smagí (Better welfare, Danish, better taste) it is also of higher quality and has the requirement that the pig must have 30% more space in the paddock and more comfort in its lying area that non-branded pigs. It emerges that branded pigs

are marketed to be of Danish origin and given the investment and reputation vested in building these brands such as the Gourmet Gris it is highly unlikely that retailer would source such pigs elsewhere. It must therefore be assumed that 15%-19% of the fresh pork for human consumption in the retail sector which make up the high quality pork products is not in any event, subject to any price restraint from imports.

In view of this, it is difficult for the Commission to rely on the claim that Danish preferences would change if prices were to increase. There is no historical evidence of imports nor has the Commission been persuaded that consumer preferences are likely to change in the future. The fact that consumers are willing to consume imported dairy products or beef does not necessarily imply that they would behave in similar fashion with respect to pork meat. It is clear that the retailers believe that consumer preferences will remain strong for Danish pork and the Commission believes that such preferences still represent a large barrier to entry.

Buyer Power

118.The parties highlight that the retail market is highly concentrated. Three major retail buyers in Denmark have more than 80% market share of the retail pork sales segment. These retail chains are claimed to have significant bargaining power towards the slaughterhouses and would use this to prevent any increases in prices. The Commission accepts that the retail market is highly concentrated but finds that it is difficult for the retailers to use their bargaining position if there are no other suppliers on the market who can deliver the right quality, range and volume of products. Furthermore, the three major Danish supermarkets together only account for [50-60]% of the parties totals sale of fresh pork. Based on the parties total sale of fresh pork for direct human consumption Coop Danmark accounts for [Ö], Dansk Supermarked for [Ö] and Super Gros for [Ö]. In addition the three largest retailers in Denmark by large quantities of pork form the parties. In 2000 Coop Danmark bought [Ö] of their supplies of fresh pork from the parties. In the same period Dansk Supermarked bought [Ö] and Super Gros bought [Ö] of their fresh pork from the parties. On balance therefore it could be concluded that there would not be sufficient countervailing buyer power to the merged entity.

Conclusion

119.In conclusion, the merger threatens to create or strengthen a dominant position on the Danish market for supply of fresh pork sold through retail.

5.2.2 Catering

120.On the Danish market for the supply of fresh pork to caters, the parties combined market share will in 2000 be [10-20]% in Denmark. The market is supplied from wholesalers relying on supplies mainly from private Danish slaughterhouses. On this market, the transaction does not threaten to create or strengthen a dominant position as a result of which effective competition would be significantly impeded on a market in Denmark.

5.3 The supply of fresh pork for further processing

If all 18,000 tonnes were sold to retailers.

121.The parties will be the largest pig slaughterhouse in Europe. In comparison, the largest competitor will be the Dutch Dumeco/Sturko with a market share of [0-10]%. However, a number of competitors exist in the market with market shares between [0-10]% and [70-80]% of the market is split between other slaughterhouses in the EU. The parties have submitted data showing the production and sale of fresh meat for further processing in the EU. The parties will have a market share of [0-10]% in the EU.

122.On the relevant market of fresh pork for further processing in Denmark, the parties have provided the Commission with market share data. However the parties claim that, the statistics on the business-to-business market is very poor. The parties estimate their market share to be [60-70]% on the Danish market for the supply of fresh pork for further processing.

Table 3: Production and supply of fresh pork for further processing in Denmark 2000/2001.

Fresh pork

Market share volume (%) Volume 1,000 tonnes product weight

[40-50]%

[Ö]

[10-20]%

[Ö]

Danish Crown/Steff-Houlberg

[50-70]%

[Ö]

Total

100%

329.6

124.It can be seen from Table 3, that the parties will control [60-70]% of the Danish production and supply of fresh pork for further processing. Their competitors are Tican and private slaughterhouses, who only supply a minor part of the market.

125.As described above, this puts the parties in a situation where they are likely to control the Danish market. Third parties including processors will be dependent on the parties for the supply of fresh pork either for the supply of processed meat to the Danish market or in relation to export of fresh pork to the USA. The Danish processors exporting to the US are left with three alternatives. Firstly, they continue to buy their supply of fresh pork form the parties and maintain their USDA-authorisation and continue to export to the US. Secondly, they can give up their USDA authorisation at begin importing pork from non-US authorised third countries. Finally, they can separate their production of US and non-US products, which would in principle mean, that they would have to build new production facilities, so that a total separation of the US and non-US meats is possible.

126.The merger primarily will create an overlap on the market for fresh pork for further processing. The overlap on the market for fresh beef for further processing is low. Steff-Houlberg does not slaughter cattle, but instead buy beef meat on the market.

127.As previously described, the parties are vertically integrated from slaughtering of pigs to the supply of processed meat. In the market for the purchase of live pigs, the parties will have a market share of 89.5%. The parties are by far the largest suppliers of pork

for further processing in Denmark. They supply a number of processors such as manufacturers of sausages, canned meat, cold cuts, liver paste etc.

128.Due to the upstream market for purchase of live pigs for slaughtering, the parties will control more than [90-100]% of the market for the supply of Danish fresh pork for further processing. The domestic manufacturers of processed meat are to a large extent dependent on supplies from the parties. After the merger, there will only be one small competitor on this market, Tican. However, due to the small size and capacity constraints of Tican, compared to the parties, Tican will not be able to supply a sufficient number of Danish processors to constrain effectively the partiesí prices. Ticanís market share on this market is at most [0-10]%. The private slaughterhouses accounting for the remaining [20-30]% will like Tican, neither be able to constrain the parties position nor supply the major processors of fresh pork.

129.The competition problems identified above are limited to the Danish market for the supply of fresh pork for further processing. In this market, the merger threatens to create or strengthen a dominant position as a result of which competition could be significantly impeded in Denmark. As noted above, the concentration will not threaten to create or strengthen a dominant position in markets outside Denmark.

5.4 The supply of processed pork products

130.As noted above, the effects of the concentration will be limited to the Danish market. On the Danish market, the parties will have [40-50]%market shares of the overall processed meat market (that is including processed meat from all species). According to the product market definitions, the market share has to be sub-divided further. As shown below, the overall market can be further split and separate product markets may exist. The figures below include meat from all species, that is raw cured products would include raw cured ham, beef and poultry. A further split of the market into pork products only would produce higher market shares for the parties as can be seen from the table below.

Table 4: Possible sub-markets for processed pork products in Denmark.

Overall market

Partiesí market share(%)

Product markets

Raw cured products

[20-30]%

[90-100]% Pre-packed sliced bacon

Processed meat for cold consumption

[30-40]%

[70-80]% Product made from the pork loin (hamburgerryk).

[40-50]% Cooked belly roll (rullep¯lse).

Canned meat

[50-60]%

[~100]% Lean pork shoulder (mager bov)

[~100]% Canned liver paste

Sausages

[50-70]%

P‚tÈ and pies

[30-40]%

[90-100]% (Frikadeller) meat balls

[40-50]% burgers

a) Raw cured products

Table 5: The production and supply of raw cured products in Denmark in 2000/2001.

Raw cured products

Market share volume

Volumes tonnes product weight

[10-20]

[Ö]

[10-20]

[Ö]

Danish Crown/Steff-Houlberg

[20-30]

[Ö]

Eberhart, DK

[10-20]

[Ö]

Theilgaard, DK

[0-10]

[Ö]

Lepo, DK

[0-10]

[Ö]

Harryís r¯g

[0-10]

[Ö]

J¯rgen Rasmussen, DK

[0-10]

[Ö]

Kreatina, S

[0-10]

[Ö]

Others

[30-40]

[Ö]

Total market

100

23,000

of which import

< 5

500

133.The parties market share is estimated to be [20-30]% on the market for raw cured products. As it can be seen from table 5, the parties will be the largest supplier on this market. Most of the competitors are small sized Danish Competitors. From time to time import occurs from other countries. However, the parties estimate the import to be approximately 2%. The partiesí market share of sliced pre-packed bacon is approximately [80-90]% in Denmark.

134.It remains uncertain, whether sliced pre-packed bacon constitute a separate product market. However, indications show, that this might be the case. In any event, it can be left open, since it at most effects the Danish market.

b) Processed meats for cold consumption

135.This product market consists of different cuts for cold consumption in sandwiches ìsm¯rrebr¯dî and could also be used in salads.

Table 6: The parties and competitors market share of production and supply of processed meat for cold consumption in Denmark in 2000/2001.

Processed meat for cold consumption

Market share volume (%) Volumes tonnes product weight

[20-30]

[Ö]

[0-10]

[Ö]

Danish Crown/Steff-Houlberg

[20-40]

[Ö]

Defco, DK

[10-20]

[Ö]

G¯l, DK

[0-10]

[Ö]

3-stjernet, DK

[0-10]

[Ö]

Suhl, DK

[0-10]

[Ö]

Theilgaard, DK

[0-10]

[Ö]

SÿR-WI, DK

[0-10]

[Ö]

Bunkenborg (Skare), DK

[0-10]

[Ö]

Danpo, DK

?

Kreatina, S

?

Zimbo, D

?

Bˆklunder, D

?

Stockmeyer, D

?

Others

[20-30]

[Ö]

Total market

100.0

42,500

of which import

20-25

10,000

c) Canned meats

Table 7: Production and supply of canned processed meat in Denmark 2000/2001.

Canned processed meats

Market share volume (%) Volume tonnes product weight

[50-60]

[Ö]

[0-10]

[Ö]

Danish Crown/Steff-Houlberg

[50-70]

[Ö]

Bauvais, DK

[20-30]

[Ö]

Bˆklunder, D

[0-10]

[Ö]

Haugen-Gruppen S

?

Others

[0-10]

[Ö]

Total market

100.0

3,500

- of which import

15-20

600

d) Sausages

Table 8: Production and supply of sausages in Denmark 2000/2001.

Sausages

Market share volume (%) Volume tonnes product weight

[20-30]

[Ö]

[30-40]

[Ö]

Danish Crown/Steff-Houlberg

[50-70]

[Ö]

G¯l, DK

[10-20]

[Ö]

ÿ-p¯lser (Stryhnís), DK

[10-20]

[Ö]

Tican, DK

[0-10]

[Ö]

Theilgaard, DK

[0-10]

[Ö]

Danpo, DK

[0-10]

[Ö]

Bˆklunder, D

?

Rose Poultry, DK

?

Robert DamkjÊr, DK

?

Others

[0-10]

[Ö]

Total market

100.0

27,000

- of which import

< 5

500

e) P‚tÈs and pies

Table 9: Production and supply of p‚tÈs and pies in Denmark 2000/2001.

P‚tÈs and pies

Market share volume (%) Volume tonnes product weight

[10-20]

[Ö]

[10-20]

[Ö]

Danish Crown/Steff-Houlberg

[30-40]

[Ö]

Stryhnís, DK

[40-50]

[Ö]

TjÊreborg, DK

[10-20]

[Ö]

G¯l, DK

[0-10]

[Ö]

Gerlach Import, DK

[0-10]

[Ö]

Others

[0-10]

[Ö]

Total market

100.0

18,000

- of which import

5-10

1,400

f) Ready prepared dishes and components for such

Table 10: Production and supply of ready-prepared dishes and components in Denmark 2000/2001 (parties estimate).

Ready prepared dishes and components for such

Market share volume Volume tonnes Volume tonnes

dishes and corrected product product weight

(%)

[20-30]

[Ö]

[Ö]

[0-10]

[Ö]

[Ö]

[20-40]

[Ö]

[Ö]

Danpo, DK

[10-20]

[Ö]

[Ö]

Daloon

[0-10]

[Ö]

[Ö]

Inalca, I (McDonalds)

[0-10]

[Ö]

[Ö]

Ardo, B

[<1]

[Ö]

[Ö]

DafgÂrd, S

[<1]

[Ö]

[Ö]

Tavola/Perkins, L, B & F

[<1]

[Ö]

Schwans, UK &D

[<1]

[Ö]

Jensenís, DK

[<1]

[Ö]

Alberto, D

[<1]

[Ö]

Salomon, D (Burger King)

[0-10]

[Ö]

[Ö]

Defco, DK

[0-10]

[Ö]

[Ö]

Moy Pack, IRL & F

[0-10]

[Ö]

[Ö]

Tops, B

[<1]

[Ö]

[Ö]

Findus, S

[<1]

[Ö]

[Ö]

Wagner, D

[<1]

[Ö]

Dr. Oetker, D

[<1]

[Ö]

[Ö]

Ranch Master, D

[<1]

[Ö]

Top Food, DK

[<1]

[Ö]

Others

[30-40]

[Ö]

Total market

100.0

31,928

100.0

25,596

Actual competition

The parties will be the largest competitor on all defined markets. Furthermore, the parties will be vertically integrated from the supply of meat for further processing to the processed products. This is also only the case for Tican on the markets for sausages and ready prepared dishes and components and the sub-segments identified. However, Tican is only marginally represented on the market for the supply of sausages. The competitor Defco is supplying speciality cold cuts made from beef, poultry and pork, and it has a market share of [10-20]% on the market for cold cuts. On all other markets, Defco have a market share below [0-10]%. Stryhnís is a major competitor on the market for liver paste, where it is the largest supplier with a [40-50]% market share. However, Stryhnís are not active on other cold cuts product market. It is only present on the market for sausages through its company ÿ-p¯lser.

For sausages in Denmark, the parties will have a market share of [50-70]%. The largest competitor GÿL will have a market share of [10-20]% and the third largest competitor ÿ-p¯lser will have [10-20]%. All other competitors have a market share below 2%. Imports of sausages to Denmark are very low, and since the products are different according to consumer preferences, the market would constitute a separate product market. According to the retail sellers, Danish consumers prefer sausages of Danish origin. Furthermore, Danish sausages have a ìmÊrkevareî brand effect appealing to Danish consumers. Imports have taken place in 2000, but mostly to compensate for a deficit on the Danish market, due to shortage. On the market for ready-prepared dishes and components, separate product markets such as frikadeller (meatballs) could be identified, where the market share of the parties is [90-100]% .

146. Based on the limited product range, no competitors are able to engage in effective competition with the merging parties. On the contrary they are limited to focusing on their existing product range and are vulnerable to competition from the parties. They further lack the financial means and size of the parties. The parties will be the largest individual supplier on every market, except for the supply of liver paste. Furthermore, the Danish competitors will also, in varying degree be dependent on supply of fresh pork and beef from the parties as suppliers for their manufacture of processed meat.

Potential competition - imports

147. Foreign supplier could deliver processed meat to the Danish market and they do to a certain extent. If products are produced according to Danish consumer preference import could take place. In the short run, the supplier would have to incur the cost of changing production and provide with the special Danish products, such as cold cuts. No major obstacle to supply side substitution seems to exist, leaving aside the possible preference for products made from Danish meat.

148. However, the Commission has not been able to identify any foreign supplier, offering a product range comparable to the parties. Imports seem to be concentrated on foreign specialities such as Parma-ham, salamiís etc. However, on the market for ready prepared dishes and components for such, depending on the precise market definition, imports seem to play a more important role. However, based on the information given by the parties on the catering and fast food sub segment, imports do not seem significant.

149. In addition, imports have to be based on contracts with retailers on the Danish market. As explained earlier the Danish retail market is concentrated with the three major retail chains accounting for around 80% of the sale. Since some of the processed meat products such as sausages and cold cuts have a limited durability time due to freshness, it is important for a supplier to either have deliver directly to the retail shops or to the retailersí distribution centres.

150. If a foreign supplier is to be attractive to the retail chains, the supplier must be able to supply a wide range of processed products at a competitive price in comparison to the parties.

Brand names and advertising

151. The existence of strong brand names and advertising constitute an additional entry barrier for potential imports. The Danish markets for processed meats are characterised by a number of brand names. These brand names are intensively marketed on the Danish market. The marketing comprises products such as bacon, cold cuts, liver paste, sausages and ready prepared dishes. This includes brand names such as Stryhnís (liver paste), 3-stjernet (salami), G¯l-p¯lser (sausages). Also the largest Danish retailer Coop Danmark is marketing its private brand Goman (cold cuts).

152. The parties have various brand names. On the market for cold cuts Danish Crown use brand names such as P≈lÊkker, Tulip, Den Gr¯nne Slagter. The P≈lÊgger brand is market as made of Danish meat and is intensively branded on national television. After 8 months of sharp advertising in Denmark, 80% of the consumers know the brand name. Tulip is known conducting an advertising campaign including television-commercials and is built on the co-operation with the known chef Keith Floyd and Ida Davidsen, who is well known for making ìsm¯rrebr¯dî open sandwiches in Denmark. Steff-Houlberg is particularly well known for its sausages and also markets their cold cuts such as ìLuxury delicatessenî.

153. The brand name ìDen Gr¯nne Slagterî is another brand the parties advertise and is one of the best know brand names for cold cuts in Denmark and Sweden. It has a low fat content (3%) and is appealing to consumers with the desire for cold cuts with a low fat contend. The marketing of the ìDen Gr¯nne Slagterî is made through strong television-commercials, recipes and hangers.

154. Overall it can be concluded that brand name and advertising seem important in order to supply the market for processed meat. Especially products such as bacon, sausages, cold cuts and p‚tÈs and pies are regarded as brand products. The major suppliers are all involved in advertising on national television. It can be concluded, that the parties have several strong brand names on the Danish market enhancing their very strong position on the market.

Buyer power

155. As it was the case in relation to fresh pork sold in retail, it can be argued that the retail market is highly concentrated. The three major retail buyers in Denmark have a market share of more than 80% of the retail market segment. These retail chains are claimed to have significant bargaining power towards the suppliers of processed meat products and would use this to prevent any increases in prices. The Commission acknowledge the high concentration of the buying power but finds that it is difficult for the retailers to use this buying power if there are no other suppliers on the market who can deliver the right quality, range and volume of processed products. On balance therefore it would appear that there would not be sufficient countervailing buyer power to the merged entity.

Conclusion

156. In conclusion, the merger threatens to create or strengthen a dominant position as a result of which effective competition will be significantly impeded on several separate markets for processed meat in Denmark. No competition problems would arise outside the Danish market.

5.5 Collection of abattoir by products in Denmark

157. The parties had activities in this field through two entities, Daka and Kambas. The two entities have now merged. Daka is a co-operative involved in the collection of abattoir by products and high-risk material. Danish Crown owns the majority of shares in Daka but due to an undertaking given in the Previous Case Danish Crown has reduced its voting rights in Daka so that it does not have sole control over Daka. Daka has recently acquired Kambas, another co-operative collecting abattoir by-products, which was formerly an integrated division of Steff-Houlberg.

158. At present there are 3 companies including Daka approved by the government to handle high-risk material. On the market for collection of high-risk material Daka had in 2000/2001 a market share of 89%. Daka is organised as a co-operative, and its members have an obligation to supply Daka. All the major slaughterhouses in Denmark are members of Daka.

159. The merger between Danish Crown and Steff-Houlberg will result in Danish Crown achieving control of Daka as due to the merger, since Danish Crown will gain hold to Steff-Houlbergs interest in Daka. This would mean, that Danish Crown gains sole control of Daka. However, according to the decision in the Previous Case, Danish Crown is not allowed to gain control of Daka. This was in order to protect the small slaughterhouses depending on Daka. As a result, Danish Crown has pledged to dispose of Steff-Houlbergís interest in Daka, and not gain sole control of Daka. However, Danish Crown will reduce its influence in Daka prior to the implementation of the merger between Danish Crown and Steff-Houlberg in order for the merged entity not to have sole control over Daka.

160. Since the activity of Daka is limited to the Danish market, and the merger will result in Danish Crown gaining control of Daka, the merger threatens to create or strengthening a dominant position as a result of which effective competition will be significantly impeded on a market in Denmark, which represents all the characteristics of a distinct market, the aspects of the concentration regarding Daka, can be referred to Denmark. However, any changes made have to be in compliance with the commitments in the Previous Case.

5.6 The failing company defence

161. The parties maintain that, during the last three years, Steff-Houlbergís annual result have not been sufficient to enable Steff-Houlberg to pay a competitive residual payment to its farmers. In particular, Steff-Houlberg has not been able to match the residual payment of Danish Crown. In 2000/01, the residual payment of Steff-Houlberg was 0.30 DKK/kg (Ä0,04) lower than that of Danish Crown. They also claim that if the merger with Danish Crown was not announced a large number of suppliers would give notice to leave Steff-Houlberg with the aim to join Danish Crown. This led Steff-Houlberg to initiate merger negotiations with Danish Crown.

162. So far, no members have left Steff-Houlberg to join Danish Crown, but the parties maintain that this may be the likely consequence of the difference in the residual payment. The parties maintain that, if a large number of members left Steff-Houlberg, (accounting for approximately [Ö]% of its supply) it would lead to a rapid decline in revenues and ultimately it would result in the economic breakdown of Steff-Houlberg. It is argued that the only Danish slaughterhouse, which is able to absorb the bulk of the members, is Danish Crown, it is also the only slaughterhouse, which could utilise Steff-Houlbergís capacity in a viable fashion. [Ö].

163. Based on this, Danish Crown claims that the transaction would rescue Steff-Houlberg from impending bankruptcy. In the light of this, Danish Crown points out that the failing company defence applies. The parties argue that Steff-Houlberg satisfies the criteria identified in case COMP/M.2314 BASF/Eurodiol/Pantochim .

(a) the acquired undertaking would in the near future be forced out of the market if not taken over by another undertaking,

(b) there is no less anti-competitive alternative purchase, and

(c) the assets to be acquired would inevitably exit the market if not taken over by another undertaking.

164. The application of the concept of the ìrescue mergerî requires that the deterioration of the competitive structure through the merger is at least no worse than in the absence of the merger.

165. The parties argue, that in the absence of this merger Steff-Houlberg members will begin looking for an alternative co-operative due to the lower residual payments and that Danish Crown will by far be the most evident alternative. Such flight from Steff-Houlberg would inevitably lead to economic problems and result in an organic growth of Danish Crown who is obliged to accept new members.

166. The Commission does not agree with these arguments. The parties have informed the Commission that a [Ö]% reduction in the supply of pigs would be acceptable to Steff-Houlberg, a [Ö]% decline would affect its competitiveness seriously and a [Ö]% decline would impose a major risk on its viability. The parties have estimated, that a decline of [Ö]% of supplies would make the residual payment fall to DKK [Ö] less than Danish Crown (all other things remaining equal) which is equivalent to lower annual earnings of DKK [Ö] for an average member (supplying the average number of 1200 pigs per year) of Steff-Houlberg.

167. It has not, however, been sufficiently shown that mass exit would occur. For the supply of pigs to decline by [Ö]%, more than [Ö] farmers would have to leave Steff-Houlberg. Such exit is unprecedented. For example, when the merger between Vestjyske Slagterier and Danish Crown was announced, only around [Ö] farmers left. In addition, it is not clear that Danish Crown would have sufficient capacity to accommodate such a vast number of pigs without also acquiring assets for additional slaughtering capacity. The exit of members of Steff-Houlberg shows, that the members leaving Steff-Houlberg after giving the 2 year notice is very limited. In 1998/1999 [Ö] members left Steff-Houlberg of which [Ö] joined Danish Crown. In 1999/2000 [Ö] members left Steff-Houlberg but none joined Danish Crown. Of the [Ö] members leaving Steff-Houlberg in 2000/2001 only [Ö] joined Danish Crown.

18 Decision of 11 July 2001.

168. Looking at the residual payment of Steff-Houlberg in 1998/1999 it reached the same level as that of Danish Crown. [Ö]. In 2000/2001 the real difference between the two companies was DKr. 0.30 (Ä0.04) pr. kilo. However, the transfer of funds from the equity has not jeopardised the financial viability of Steff-Houlberg.

169. The claim that most farmers would go to Danish Crown is also unsubstantiated. There are 10 privately owned slaughterhouses slaughtering over 10,000 pigs per year that could absorb a significant proportion of Steff-Houlbergís members.

170. Another consideration is whether Steff-Houlberg would be forced out of the market in the first place. Steff-Houlberg is, at present, making profits and its financial situation is healthy. Furthermore, it is the seventh largest pig slaughterhouse in Europe and it is conceivable that alternative buyers, such as non-Danish slaughterhouses, could be considered as potential purchasers. In 2000/2001 Steff-Houlberg paid the third highest price for slaughter pigs in Europe. Only Danish Crown and Tican were paying higher prices. Furthermore, Steff-Houlberg owns the largest and most modern slaughterhouse in Northern Europe. The Commission does not consider that the parties have shown that there are no alternative buyers. Indeed, the company has not been put up for sale and little effort has been undertaken to find such alternative buyers.

171. The parties have also failed to show that the assets would inevitably exit the market if Steff-Houlberg were not taken over by Danish Crown. Even if Steff-Houlberg were to go out of business, as the parties claim, the Commission does not believe that the assets would be lost. In particular Danish Crown and other slaughterhouse would purchase the assets to satisfy the increased demand for slaughter capacity which would arise from Steff-Houlbergís failure.

172. Finally, the condition that the competitive structure through the merger is at least no worse than in the absence of the merger is also not satisfied. It is unlikely that, in the event of exit of Steff-Houlberg, Danish Crown would be the only slaughterhouse, which would gain ex-Steff-Houlberg members. Some members would be expected to go to private slaughterhouses. This could strengthen the position of the private slaughterhouses to the benefit of the competitive process. Such benefits would not be derived through the merger.

173. The Commission therefore considers that the proposed merger does not, on the basis of the information available, qualify for the failing company defence. In particular, it has not been sufficiently shown that members would leave to such a degree to lead Steff-Houlberg to bankruptcy. Nor has it been shown that no other less anti-competitive purchaser is available or that Danish Crown would, in the absence of the merger, gain all the members of Steff-Houlberg in any event. Finally, it is highly unlikely that the assets of Steff-Houlberg would exit the market should the merger fail to take place. It is therefore concluded that the rescue merger concept does not apply in this case.

174. As part of the Previous Case Danish Crown made commitments, in order to protect the smaller Danish co-operative slaughterhouses in their trade association, Danske Slagterier (Association of pig slaughterhouses). Danske Slagterier is among other things involved in the R&D in connection to the breeding of pigs. The present merger may, if approved, raise a conflict with these commitments. The issue is, therefore, a matter for the Danish Competition Authority to take into account in its investigation.

175. From the above it follows that the conditions to request a referral under article 9(2)(a) are met. The Commission also considers that, given the local scope of the markets affected by the transaction, the Danish national competition authorities are better placed to carry out a thorough investigation of the case, and that it is therefore appropriate for the Commission to exercise its discretion under article 9(3)(b) so as to refer partially the case to Denmark as regards the distinct markets identified above.

176. It is to be noted that this Decision is without prejudice to the commitments made in the Previous Case which remain valid.

177. On the same date as this decision, the Commission has also adopted a decision on the basis of Article 6(1)(b) of the Merger Regulation clearing the operation with regard to the markets for sale of fresh pork for further processing and processed pork products as regards markets outside Denmark.

178. Accordingly, the Commission has adopted this decision :

Article 1

The notified concentration of the merger between Danish Crown and Steff-Houlberg is referred in part to Denmark, pursuant to article 9(3)(b) of Regulation 4064/89.

Article 2

This decision is addressed to Denmark.

Done at Brussels,

For the Commission

Mario MONTI Member of the Commission

39

40