CCE, 5 décembre 2006, n° M.4389

THE COMMISSION OF THE EUROPEAN COMMUNITIES,

Dear Sir/Madam,

Subject : Case No COMP/M.4389 - WLR/BST

Notification of 27/10/2006 pursuant to Article 4 of Council Regulation No 139-2004 (1)

Publication in the Official Journal of the European Union No. C 276 page 4

1. On 27/10/2006 the Commission received a notification pursuant to Article 4, and following a referral pursuant to Article 4(5), of Council Regulation 139-2004 (the "Merger Regulation") with respect to a concentration leading to the acquisition of sole control by the American company WLR Recovery Fund III, LP ("WLR") of the German company BST Safety textiles Holding GmbH ("BST").

2. After examination of the notification, the Commission has concluded that the notified operation falls within the scope of Council Regulation (EEC) No 139-2004 and does not raise serious doubts as to its compatibility with the common market and with the EEA Agreement.

I. THE PARTIES

3. WLR is a private equity investment fund, ultimately controlled by a US citizen, Mr. Wilbur Ross. Of the companies controlled by WLR, one, the American company Safety Components International "SCI"), is active in the sector concerned by the transaction, that is, the manufacture of components for automotive airbag modules. In Europe, SCI manufactures cut and sewn airbag cushions ("CSC", see below).

4. BST is the holding company of a group of companies active in weaving technical fabrics including their design, production and distribution. In Europe, BST manufactures flat airbag fabric for use in CSC airbags, and also "one piece woven" airbags ("OPW", see below).

II. CONCENTRATION

5. WLR will acquire the entire share capital of BST, through an acquisition vehicle, BST Safety Textiles Acquisition GmbH, wholly-owned by WLR. The notified transaction therefore constitutes a concentration within the meaning of Article 3 (1) (b) of the Merger Regulation.

III. COMMUNITY DIMENSION

6. The concentration does not have Community dimension within the meaning of Article 1 of the Merger Regulation. On 25 August 2006 the notifying party informed the Commission in a reasoned submission pursuant to Article 4(5) of the EC Merger Regulation that the concentration was capable of being reviewed under the national competition laws of at least three Member States, namely Germany, Poland, Portugal and Spain, and requested the Commission to examine it. None of the Member States competent to examine the concentration indicated its disagreement with the request for referral within the period laid down by the Merger Regulation. As no Member State has objected to the referral, the transaction is to be reviewed by the Commission.

IV. ASSESSMENT

A. Market structure

7. The sector involved in the transaction is the manufacture of components for automotive airbag modules, which provide passenger protection in the event of a collision. An airbag module consists basically of a sensor, an inflator, and an inflatable bag or cushion, and it is the manufacture of the latter component in which the parties are active (the relative approximate costs of airbag module components are [10-20]% for the cushion, [40-50]% for the inflator, and [35-45]% for the sensor and other parts). Airbag modules are generally manufactured and assembled by so-called 'tier 1' companies for supply to OEM automotive manufacturers.

8. The various types of airbag modules include driver, front passenger and side modules. In airbag modules for "first impact" protection, the cushion must stay inflated as long as the impact lasts (1 second or less), whereas in some side airbag modules which are specially intended for "rollover" protection - (so-called "curtain" modules, usually installed in sports utility vehicles) - the cushion must stay inflated for as long as the car is assumed to roll over after the impact (approximately 10 seconds). Such 'curtain' modules require airbags which include a specifically designed pattern of separate inflatable compartments.

9. Airbag cushions are composed of fabric, which is woven from yarn which has appropriate technical qualities, and which represents approximately [40 to 80]% of the total cost of a cushion. CSC (see above) are manufactured by sewing together pieces of woven "flat fabric" which have been cut into the appropriate shape. OPW (see above) cannot be woven on flat fabric looms, and are manufactured by weaving the bag directly from yarn on relatively sophisticated and costly looms, thus obviating the need for cutting and sewing. OPW technology is particularly suited to the production of airbags which involve a complex pattern of separate inflatable chambers in one and the same cushion.

10. Within the EEA, there are four main suppliers of flat airbag fabric, BST, UTT (Germany), Milliken (US/UK), and Porcher/NCV (France). Recent market entrants are Takata (Japan) which has recently completed a weaving operation in Romania (an EU member as of January 2007), and Toray (Japan), which is soliciting for business with TRW and Autoliv (see below). As far as CSC are concerned, there are two main 'independent' suppliers, SCI itself and Aerazur (France). The two major tier 1 airbag module suppliers, TRW (US) and Autoliv (Sweden), are vertically integrated upstream into CSC, as are some smaller players such as the Japanese company Takata, and indeed Autoliv has significant in-house OPW production; all tier 1 companies use their inhouse CSC/OPW production exclusively for in-house consumption, and do not sell to third parties. The overall average annual growth rate for airbag units in the EEA is around 5%, and is expected to remain at this level over the next several years.

B. Product market definition

11. The parties submit that there are three relevant product markets, that is, the flat fabric market (which the Commission's investigation has confirmed is a distinct upstream market), and furthermore two distinct downstream markets, CSC and OPW.

12. Although CSC and OPW both perform the same basic function, as cushions in airbag modules, the Commission's investigation has confirmed that they are not in reality technical or economic substitutes.

13. Firstly, as regards front airbag modules, CSC may have two or three sides, which are sewn together, but it is not possible to produce three-sided bags using OPW technology; the majority ([70-90]%) of front passenger airbags are three-sided in order to offer greater volume when inflated, and are therefore CSC of necessity. Furthermore, although the remaining [10-30]% of front passenger airbags and all front driver airbags are two-sided, producing these with OPW technology would not be economically rational, given the significantly higher costs involved; the cost of a CSC is approximately € [3-15] whereas the cost of a OPW is between € [5-20], which reflects, principally, the differential capital costs of weaving looms (€50,000-100,000 and €230,000-250,000 respectively).

14. Secondly, as regards side airbag modules, both CSC and OPW can be, and are, used for 'first impact' protection (see above). However, where rollover protection (see above) is needed, OPW is the only technology which is technically and economically appropriate for the requisite, relatively complex, pattern of separate inflatable compartments. The sewn seams in CSC are not as airtight as are the woven seams in OPW; the stitches create small holes in the fabric, and the air can leave the airbag cushion rapidly in the area between the two (cut and sewn) fabric layers. CSC would have to be sealed on the sewing lines to prevent overly-rapid deflation, making them significantly more expensive to produce than OPW.

15. In short, CSC are more appropriate for relatively simple first-impact cushions, where additional labour costs for the cutting and sewing process are more than offset by the relatively inexpensively woven flat fabric used, whereas OPW are more appropriate for more complicated rollover applications, where the economic trade-off between labour costs and automation favours the latter. In the course of its investigation, the Commission found one or two limited examples of OPW used in front airbag modules, and of CSC used for rollover applications. However, the former case involved a module manufacturer with extensive in-house OPW facilities, which may well have altered the economics of production in this particular instance; the latter involved a module manufacturer with in-house CSC facilities in an extra-EEA country with low labour costs which could not be achieved in Europe.

16. The above distinction is reflected in the experience of the notifying parties themselves. SCI has no supply program for CSC side curtain airbag modules for rollover protection, and conversely BST has no supply program for OPW for front airbag modules, since neither company has ever been approached to develop such programs.

17. In conclusion, on the basis of its market investigation, the Commission has concluded that flat fabric, CSC, and OPW constitute distinct product markets.

C. Geographic market definition

18. The parties argue that the markets for flat airbag fabric, CSC and OPW are EEA-wide in scope. They base their views on the homogeneity of competitive conditions within the EEA and the fact that practically all customers and competitors of the parties have production facilities within the EEA. Furthermore the parties submit that they are not aware of any significant imports of flat airbag fabric into the EEA from outside the EEA, and that total imports of CSC are estimated to represent only around 13% of total CSC sales in the EEA.

19. The Commission's investigation has confirmed that the relevant geographic market for all three products is EEA-wide in scope. This is in line with the Commission's previous findings in the automotive components industry, and in the airbag sector in particular (2).

D. Competitive assessment

20. BST manufactures only flat fabric and OPW, and SCI only CSC, within the EEA, and therefore the proposed transaction would not give rise to horizontally affected markets, there being no overlap between the parties' activities in view of the distinct markets defined above.

21. The transaction would, however, give rise to vertically affected markets since BST manufactures flat airbag fabric, which is used in CSC manufacture, in which SCI is active (indeed there is an ongoing supply relationship between the two companies). The Commission's investigation therefore focused on the competitive impact of this vertical flat fabric/CSC relationship.

22. In particular, the Commission firstly investigated the possibility that the merged entity would be likely to foreclose downstream tier 1 airbag module manufacturers, that is, would be likely to restrict access to fabric or CSC products, thereby raising its downstream rivals' costs by making it harder for them to obtain supplies under similar prices and conditions as those obtaining pre-merger ("input foreclosure"). Secondly, the Commission investigated the possibility that the merged entity would be likely to foreclose its rivals in the upstream fabric market access to a sufficient customer base, thereby reducing their ability to compete ("output foreclosure").

1. Market position of the parties

23. On the downstream market for CSC, the two main 'independent' suppliers are SCI and Aerazur (with [40-45]% and [35-40]% of 2005 'independent' sales respectively). The Commission's investigation has confirmed that independent CSC producers are subject to strong competitive pressure from tier 1 module suppliers' own in-house production of CSC, which latter represented [55-60]% of total EEA production in 2005. On the basis of total CSC production the share of SCI is [15-20]%, the other largest producers being the tier 1 module producers TRW ([20-25]%) and Autoliv ([20-25]%), followed by Aerazur with [15-20]%.

24. On the upstream market for flat airbag fabrics, BST is the EEA-wide market leader with a [45-50]% market share in 2005. Other flat airbag fabric suppliers include UTT ([20-25]%), Milliken ([15-20]%) and NCV ([10-15]%). As already noted, BST currently supplies fabric to SCI, but does not supply fabric to the other main 'independent' CSC manufacturer Aerazur; that part of BST's fabric production which is not sold to SCI goes mainly to TRW and Autoliv.

25. The Commission's investigation therefore focused on the competitive impact of the proposed combination of BST's [45-50]% EEA flat fabric market share with SCI's [15-20]% share of EEA CSC production.

2. Input (supplier) foreclosure

26. The Commission's investigation revealed that the merged entity would not have an incentive to foreclose downstream fabric or CSC customers, for several reasons.

27. Firstly, even if the merged entity sourced fabric exclusively from BST and de-selected other current fabric suppliers, SCI is unable to absorb BST's total production of flat airbag fabric. SCI's flat airbag fabric demand amounted to approximately [...] m² million in 2005 (SCI's sales of CSC having decreased from €[90-120] million in 2003 to €[60-90] million in 2005), while BST's total EEA fabric output was [...] m² million. Therefore even if BST were chosen to cover all SCI's current needs, more than [50-70%] of BST production would still be available to other customers, and it would make no economic sense to adopt commercial practices, such as excessive pricing or difficult terms of delivery, which would detract from these profitable sales of flat fabric, where BST's current profit margin is of the order of [...]% (CSC margins are much lower, around [...]%).

28. Secondly, the main fabric customers of BST, other than SCI itself, are also SCI's CSC customers. Indeed TRW and Autoliv which represent respectively [30-60]% and [10-40]% of BST sales (3) of flat airbag fabric are also two of the five largest customers of SCI, and represent [10-30]% respectively of SCI's CSC sales (4). Thus any attempt by the merged entity to foreclose its customers on fabric supply would have a negative impact on demand from these same customers for CSC, and vice versa, so any such attempt would be commercially counter-productive.

29. In any event, the Commission's investigation has shown that the merged entity would not even be able to foreclose tier 1 airbag module manufacturers such as TRW and Autoliv since the latter have significant influence upwards through the fabric/CSC supply chain. Contracts between tier 1 module manufacturers and CSC manufacturers typically include a clause obliging the supplier to be competitive in price throughout the term of the supply agreement compared to the tier 1 firm's own in-house production; indeed, because of their own in-house CSC production, tier 1 manufacturers have a detailed knowledge of the cost structure of the independent CSC suppliers, and are thereby in a strong bargaining position (the parties have provided examples of tier 1 manufacturers switching airbag suppliers during the course of a production programme). Indeed, the Commission's investigation has shown that tier 1 module manufacturers, when outsourcing the production of CSC to 'independents' such as SCI, typically insist that details of fabric input prices are included in the CSC tender document, and furthermore exert a significant degree of influence over the choice of flat fabric supplier. The extent of this upwards influence is illustrated by the fact that Milliken UK is exiting the CSC market as of December 2006 because tier 1 module manufacturers were frequently insisting that Milliken used fabric other than its own in-house fabric.

30. During the course of its investigation the Commission received no complaints from tier 1 module manufacturers concerning the proposed transaction, and found no evidence of an inadequate degree of competition in the relevant markets, where supply programs are awarded on the basis of competitive bids. Some tier 1 module manufacturers even took the view that the merger would benefit customers, especially in the context of an automotive industry which is subject to considerable cost pressures throughout the whole supply chain, and where a trend towards global 'vehicle platform' strategies may be facilitated, ceteris paribus, by the emergence of vertically integrated component suppliers.

3. Output (customer) foreclosure

31. During the course of its investigation, the Commission received some expressions of concern from the main fabric competitors of BST (UTT, Milliken and NCV/Porcher), to the effect that the merged entity would foreclose access to a sufficient CSC customer base, thereby reducing their ability to compete. The Commission analysed these concerns in some detail, and found that the proposed transaction would not bring about such a degree of customer foreclosure that competition would be significantly impeded.

32. The Commission's investigation revealed that the merged entity would not be able to foreclose competitors on the fabric market, for a number of reasons.

33. Firstly, the Commission found that it is difficult for a CSC producer to switch a fabric supplier quickly, given ongoing production program commitments on both sides, and the need for certain certification procedures to be followed if switching is to occur. Nevertheless, even if post-merger SCI were to rapidly adopt a purchasing policy of sourcing exclusively from BST, only one competitor of BST currently depends for more than 10% of his flat fabric turnover on SCI. Moreover, most of BST's main competitors are large diversified companies who also produce a range of technical fabrics for applications other than airbag modules, and who could be expected to withstand a sudden drop in flat fabric sales without undergoing severe financial difficulties.

34. In the longer term, BST's supply of flat airbag fabrics to SCI currently represents [10-30]% (around [...] m²), of BST's total output, and around [35-55]% of SCI's total fabric demand. If SCI sourced exclusively from BST, this would bring only about [10-20]% of the total EEA flat fabric production (c.[...] m² in 2005) in-house. BST's weaving capacity is currently fully utilised, and could not be increased economically in the short term (expansion would lead to significant sunk costs for investment in building and machines); so the [10-20]% of total production that would be directed to SCI would necessarily leave unsatisfied demand for fabric from SCI's competitors, demand which could only be met by BST's competitors, principally UTT, Milliken, and NCV/Porcher.

35. In a wider perspective, SCI's purchases of flat airbag fabrics currently represent only [15-25]% of total EEA demand for flat airbag fabrics, so BST's competitors would still have access to a substantial customer base. As noted earlier, overall EEA demand for airbag products is currently growing at around 5% per annum, and is likely to continue to do so in view of increasing consumer emphasis on vehicle safety.

36. Again, the Commission found that not only do tier 1 module manufacturers exercise considerable influence on choice of fabric supplier even when outsourcing CSC production (see above), but they have a preference for multiple sourcing of fabric in order to maintain sufficient competitive alternative suppliers; this is particularly so in view of the fact that tier 1 firms are themselves purchasing fabrics for their internal CSC production.

37. In the course of the Commission's investigation, one of BST's fabric competitors suggested that the merged entity, with control of its major cost input, ie fabric, would opt to cut prices of CSC in order to expand its share of the CSC market; this would create a backlash demand for cheaper fabric from other CSC's producers trying to compete, which would in turn squeeze BST's fabric competitor's margins, perhaps causing some to exit the market.

38. However, the Commission found that the merged entity would not have an incentive to adopt such a strategy. Given the transparency of the fabric/CSC production chain (see above), if tier 1 module producers saw that SCI was enjoying a lower fabric transfer price from BST post-merger, they would themselves request a corresponding price cut from BST for their fabric prices for in-house production. As noted earlier, fabric margins are much higher than CSC margins (about [...]% as against [...]%), so that the overall trade-off of a low-price CSC strategy would be negative for the merged entity; any profits from increased volumes of CSC sales achieved on the back of lower prices for the [10-30]% of BST output that goes to SCI would be more than negated by lower margins on the [70-90]% of BST output sold to third parties.

4. Conclusion

39. In view of the above, the creation of foreclosure effects such as to significantly impede upstream or downstream competition can be excluded.

V. CONCLUSION

40. For the above reasons, the Commission has decided not to oppose the notified operation and to declare it compatible with the common market and with the EEA Agreement. This decision is adopted in application of Article 6(1)(b) of Council Regulation (EEC) No 139-2004.

Notes:

1 OJ L 24, 29.1.2004 p. 1.

2 See for example case COMP/M.3972 TRW Automotive/Dalphi Metal Espana.

3 Data in volume for 2005.

4 Data in volume for 2005.