Joint Venture Investigations After Publication of the Consolidated Jurisdictional Notice

Ted Henneberry and Douglas Lahnborg

Orrick Herrington & Sutcliffe LLP

This article describes the Commission’s assessment of joint ventures under the EC Merger Regulation (ECMR). Particular attention is given to jurisdictional issues in the Consolidated Jurisdictional Notice (the Notice) and cases decided after its publication in 2007. It concludes with an overview of the Commission’s substantive assessment of joint ventures. This often mirrors that of mergers and acquisitions, although occasionally, the Commission also analyses the competitive effects of joint ventures under article 81 of the EC Treaty.
Joint ventures cover a broad range of commercial arrangements where two or more parties reach a common understanding to achieve a specific goal, and typically share losses and profits. The complexity of joint ventures varies from simple collaborations such as technology firms cooperating with manufacturers to bring their ideas for products to market to the creation of complex corporate structures such as multi-national defence contractors pooling assets and employees to develop military aircraft. EU competition legislation covers the full range of joint ventures. It segments joint ventures into three categories: (i) acquisition of joint control by two or more companies of an existing company – for example two private equity firms taking over a publicly listed company; (ii) creation of a full-function joint venture set up to operate independently of its parents on a lasting basis – for example, a mobile handset joint venture such as SonyEricsson; and (iii) joint ventures closely linked to their parent companies and whose creation does not result in a structural change in the market – such as a joint R&D project. While the first and second categories require mandatory notification under the ECMR, assuming the turnover thresholds are met, category (iii) falls outside the ECMR and does not require notification to the Commission. Category (iii) joint ventures fall under article 81 of the EC Treaty and can be investigated under Regulation 1/2003.
Since the publication of the Notice in July 2007, over 40 joint ventures have been investigated by the Commission. All of these have been cleared by the Commission at Phase I. While the majority have resulted in clearances under the simplified procedure, some cases have been afforded more scrutiny and shed some light on the Commission’s application of the Notice.

Jurisdictional assessment

In 2007, the Commission published a Consolidated Jurisdictional Notice under the ECMR (the Notice). The Notice, in chief, clarifies and replaces four existing guidelines from 1998. This section describes the rules that determine whether a joint venture qualifies for ECMR investigation, with a focus on the Notice’s developments and clarifications of the principles set out in the old guidelines.

Acquisition of joint control in existing businesses

Transactions that constitute ‘concentrations’, as defined by the ECMR, require notification to the Commission. Acquisition of ‘joint control’ by two or more undertakings of another undertaking is one category of such concentrations.¹ As a result of the transaction, two (or more) undertakings – the parents – will control another undertaking – the joint venture. A change from sole to joint control also qualifies for investigation. The same is true for transactions replacing one controlling parent with another parent. A reduction, however, in the number of controlling parents will not qualify for investigation, if the transaction does not leave one of the joint venture partners with sole control. This section relates to acquisition of joint control in existing undertakings. Creation of newly established ‘full-function’ joint ventures, another category of concentration, is described in section 1(b) below.
Definition of joint control gives rise to complex and often finely balanced jurisdictional assessments and this is not the place to go through this in detail. Briefly, joint control exists where two or more undertakings have ‘the possibility of exercising decisive influence’ on another undertaking.² Joint control arises where two companies own 50 per cent each of the voting rights or where both parents have the right to appoint an equal number of board directors. In those situations, both parents can exercise decisive influence on the joint venture. Parents may also exercise decisive influence without equality in voting rights or right to appoint directors. This is the case where minority shareholders have veto rights over a joint venture’s strategic commercial behaviour. Standard minority shareholder protection, such as rights to reject changes in a joint venture’s articles and memorandum of association, or an increase or decrease in capital, or liquidation, is not sufficient to confer control. In practice, joint control tends to turn on whether the minority shareholder can veto the annual business plan, annual budget and appointment of senior management. A shareholder with all three veto rights most likely will exercise control; a shareholder with one or two veto rights might exercise control depending ultimately on whether the shareholder can block the functioning of the joint venture.
Another area that deserves further attention, given the Commission’s recent clarification of its policy, is joint control resulting from ‘commonality of interests’. Minority shareholders without veto rights control a joint venture if they will act together in exercising their voting rights, and combined they account for a majority of voting rights. Clearly, such control can result from a shareholders’ agreement obliging the parties to act in the same way. In addition, collective action can occur ‘where strong common interests exist between the minority shareholders to the effect that they would not act against each other’.³ This is much less clear cut. The Notice sets a high standard: a common interest will confer joint control ‘very exceptionally’.⁴ And the greater the number of parent companies, the lower the likelihood of joint control. It is easy to lose sight of these key principles because the Notice describes several situations where commonality of interests may confer control, which are not exceptional – in fact, they are very common. For example, the Notice states ‘there is a higher probability of a commonality of interests if the shareholdings are acquired by means of concerted action’.⁵ This covers a wide range of joint venture investments. Helpfully, the Notice clarifies that, in general, a common interest as financial investors, such as club deals by private equity firms, does not confer joint control. Moreover, a high degree of mutual dependency between the parent companies to reach the strategic objectives of the joint venture – for example, where each parent contributes vital technology to the joint venture – is also indicative of a commonality of interest and joint control. Similarly, dependency of a majority shareholder on a minority shareholder – such as where the joint venture depends on the minority shareholder for financing or know-how – is another example where the parents may have a common interest.

Creation of full-function joint ventures

In addition to acquisition of joint control in existing businesses, the ECMR applies to the creation of ‘full-function’ joint ventures. This covers the situation where two undertakings – the parents – transfer parts of their existing businesses to a newly created undertaking – the joint venture – with the intention for the joint venture to operate autonomously on the open market on a lasting basis. Creation of joint ventures which do not have full-function status does not constitute a concentration and, it follows, does not require ECMR notification, but may be investigated under article 81 of the EC Treaty, as discussed in section 1(c) below.
‘Full functionality’ is a term of art. The ECMR simply states ‘the creation of a joint venture performing on a lasting basis all the functions of an autonomous economic entity shall constitute a concentration’.⁶ The Notice does not add much by stating that full function essentially means that a joint venture must perform the functions normally carried out by other market participants. This only indicates that full functionality should be determined based on a comparison with existing market players, and does not advance our understanding of the full-functionality concept.
To get closer to an answer, it is necessary to address the four questions set out below. An overall assessment taking into account all facts is required to conclude on full functionality; however, if one or more of the questions is answered in the negative, the assumption would be that the joint venture is not full-function.

Will the joint venture have sufficient resources to operate independently of its parents?

The main principle has not changed with the Notice: a full-function joint venture must have a management dedicated to its day-to-day business and access to finance, staff and assets to allow it to operate on the market on a lasting basis. The Notice goes on to clarify that staff do not necessarily have to be employed by the joint venture; it is sufficient if staff are provided by third parties such as employment agencies. It may also be sufficient for the parents to supply staff – certainly for a start-up period, but also for longer periods if parents are treated at arm’s length basis and the joint venture remains free to hire staff on the open market.

Will the joint venture have access to customers and suppliers independently of its parents?

A joint venture will not have full-function status if it only takes over a specific function within the parents’ businesses without having access to the market. R&D and production joint ventures are likely to fall within this category. Access to the market, in itself, however, is not enough to confer full functionality: if the joint venture is limited in its sales to the parents’ products, this likely disqualifies it from full-function status. The Notice explains that real estate joint ventures tend to fall outside the definition of full-function joint ventures, if the activity is limited to holding of certain real estate for the parents and financial resources are provided by the parents. By contrast, a joint venture that actively manages a real estate portfolio and acts on its own behalf typically will have full-function status.

Will the joint venture be dependent on sales to, or purchase from, its parents?

If a joint venture relies almost entirely on sales to or purchases from its parents – it is not an independent market player and is unlikely to have full-function status. However, high dependency on sales to its parents for a start up period, typically not exceeding three years, will not exempt it from full functionality. If, however, the intention is for the joint venture to sell its goods and services to its parents on a permanent basis, it is the split between third parties and its parents that determines whether it is independent. The Notice goes on to clarify that sales to third parties exceeding 50 per cent will typically indicate full functionality. Below this level, third-party sales as low as 20 per cent may be sufficient if the relationship between the joint venture and its parents is ‘truly commercial’ in character and it can be demonstrated that the joint venture will supply its goods to the purchaser who values them most. Although it is reasonably clear what the Notice is trying to achieve here, it is not particularly well articulated and it raises at least two questions: can a joint venture, which sells 50 per cent of output to third parties, sell the remainder of its output to its parents at intra-group pricing, without losing full-function status; and who has to ‘demonstrate’ that the relationship is ‘truly commercial’ – the Commission or the parties? In determining the split between parent and third-party sales, the Commission will take into account business plans, past accounts – although one would think these might be rare for newly created joint ventures – and general market structure. The Notice does not advance the purchase element of the assessment. It simply restates the position of the old guidelines: full-function status is questionable where the joint venture adds little value to products received from parents.

Will the joint venture operate on a lasting basis?

Only if the joint venture is intended to operate on a lasting basis will it have full-function status. There is a presumption that a joint venture will operate on a lasting basis if the parents commit the resources indicated in ‘Will the joint venture have sufficient resources to operate independently of its parents?’ above. ‘Lasting basis’ does not mean indefinitely: an agreement which limits the duration of the joint venture does not rule out full-functionality. The Commission has considered a period of eight years and longer sufficient to bring about a lasting change in market structure, but three years has been too short. Possible continuation of a joint venture beyond the specified period increases its chances of having full-function status.
The Notice addresses the situation where the joint venture is dependent on a favourable third-party decision to commence its business, such as a joint venture which becomes operational only if it is awarded a contract, a licence, or access to property (eg, exploration rights for oil and gas). In those situations, the joint venture will not have full-function status, until a decision has been made in its favour. This is an unfortunate and wholly illogical development because it reduces the ability of business to achieve legal certainty until after a positive third-party decision has been made. It goes against the general trend of allowing companies to notify transactions at an early stage (and certainly before transaction agreements have been signed). Moreover, it is not only possible, it is a legal requirement, to notify public bids before the bidders know if they will be successful: why discriminate against joint ventures bidding for a key asset?
The Commission appears to have recognised this weakness in Ses Astra/Eutelsat/JV (2007), which involved the creation of a joint venture active in the provision of infrastructure for broadcasting content and two-way communication services to mobile devices.⁷ The joint venture had not yet secured its relevant spectrum licence at the time of the Commission’s investigation and three other licence applicants had higher priority than that of the joint venture. Thus it appears far from obvious that it would win the licence. Nonetheless, the Commission found that the new entity was a full-function joint venture, noting that the venture was well placed to win the licence. This case demonstrates a willingness on behalf of the Commission to apply the Notice flexibly.

Expanded scope of the joint venture

Expansion of a joint venture’s activities, the Notice clarifies, which does not entail a transfer of assets, contracts, know-how or rights to the joint venture does not constitute a new concentration. So, a joint venture can expand into a new product market without triggering a new ECMR notification, as long as the parents do not transfer any assets or rights, etc. If the parents, however, transfer significant assets or rights to the joint venture, a new concentration may arise. This is the case if (i) the assets and rights form the basis of an extension of the joint venture’s activities into new geographic or product markets; (ii) the original object of the joint venture did not cover these markets; and (iii) the joint venture will perform these new activities on a full-function basis. This new principle leaves ample scope to argue that expansion of a joint venture’s activities does not constitute a new concentration and is unlikely to increase substantially the number of joint venture transactions notified to the Commission. To start, criteria (i) to (iii) are cumulative: if one criterion is not fulfilled no new concentration arises. Moreover, the memorandum of association of joint ventures often defines the joint ventures’ objective in generic terms encompassing a wide range of activities. And in many situations the ‘expanded’ scope will be covered by the original objective of the joint venture. Furthermore, no concentration arises if the new activities are not carried out on a full-function basis. So, it appears, a full-function joint venture active in, say, manufacturing and distribution of toothpaste, which starts distributing toothbrushes manufactured by one of its parents, will not trigger a new concentration, if the joint venture will remain dependent on its parent for supply of toothbrushes. As a separate point, it should be noted that a non-full-function joint venture, which expands its business scope may turn into a full-function joint venture and trigger an ECMR filing. For example, a production joint venture which establishes a sales organisation may confer fullfunctionality.

Joint ventures outside ECMR

Joint ventures that fall outside the ECMR are caught by article 81 of the EC Treaty, if the ventures have an anti-competitive effect and impact on trade between EU member states. Since the introduction of Regulation 1/2003, it is no longer possible to notify joint ventures for article 81(3) exemption and it falls upon the parents to assess whether the joint venture infringes article 81. If it does not – and this would normally be the key risk factor – the joint venture agreement is void and unenforceable before the national courts. There is also a possibility that the Commission could open an investigation under Regulation 1/2003 and fine the parents.
Given the Commission’s limited interest in investigating joint ventures under Regulation 1/2003, joint venture parents may be tempted to structure joint ventures to fall outside the ECMR and avoid Commission investigation. This is what the parties did in the T-Online travel joint venture.⁸ This was initially notified under the ECMR and later restructured, as a result of which the Commission lost jurisdiction and abandoned its investigation. However, the Commission later opened an investigation into the joint venture under article 81. This illustrates a willingness on behalf of the Commission to investigate joint ventures which come to its attention and which it believes may raise concern, even if the joint ventures fall outside the ECMR.⁹

Substantive analysis of joint ventures

Joint ventures notified under the ECMR – both acquisitions of joint control and full-function joint ventures – are assessed against the same substantive test as mergers and other concentrations: would the concentration significantly impede effective competition, in particular as a result of the creation or strengthening of a dominant position (the SIEC test) (referred to as ‘non-coordinated effects’). Additionally, the ECMR states that the legality of notifiable joint ventures with an ‘object or effect the coordination of the competitive behaviour of undertakings that remain independent’, such as its parents, shall be determined based on article 81 of the EC Treaty (referred to as ‘coordinated effects’). While the Commission always assesses the effect of a joint venture against the SIEC standard, it is more unusual for it to apply article 81 in its the ECMR investigations. In the Commission’s view, the ECMR permits it to apply article 81 both to the acquisition of joint control in existing businesses and the creation of full-function joint ventures.¹⁰ It is clear that both tests apply to full-function joint ventures, but it is by no means obvious that the ECMR mandates the Commission to assess acquisitions of joint control under article 81.¹¹
LSG Lufthansa Service Holding/Gate Gourmet Switzerland (2006) gives a good illustration of the Commission’s methodology in the rare instances where it investigates a joint venture’s non-coordinated and coordinated effects.¹² This case concerned the creation of a full-function joint venture between LSG and Gate Gourmet in the field of airline food catering at the Paris Charles de Gaulle airport. The parent companies, LSG and Gate Gourmet, were both active in airline catering across Europe. The Commission first considered non-coordinated effects. For the 2003 to 2005 period, the combined market share of the parents in airline catering at Charles de Gaulle was between 40 and 55 per cent. The only significant competitor, Servair, had a similar market share (between 40 and 55 per cent). The Commission concluded that the joint venture would continue to face competition from Servair and that the transaction would not result in the emergence of an individually dominant firm. Moreover, the Commission noted that the joint venture would result in a near duopoly at Charles de Gaulle and that the transaction might lead to rising prices. The Commission found, however, that competitive pressure exercised by buyers (the airlines), as well as low barriers to entry in the in-flight catering market, acted as sufficient restraints to prevent the joint venture from significantly impeding effective competition, and concluded that the joint venture would not cause anti-competitive effects in the non-coordinated field.
Next, the Commission analysed the possibility of coordination between the joint venture and Servair. The Commission found that the joint venture and Servair would face difficulties coordinating their behaviour because of significant asymmetries in the operations of the joint venture and Servair. The disparity included both production capability, where Servair derived benefits from larger economies of scale, and the proportion of free market competition in the players’ operations. As for the latter consideration, Servair provided exclusive catering service to Air France – its owner – who accounted for 60 per cent of Servair’s activities. Additionally, the Commission found that airline catering is not a homogenous product, noting that each airline has specific quality specifications, making meals not generally interchangeable. Furthermore, the Commission observed that the airlines do not face significant costs in switching caterers. For those reasons, the Commission held the possibility of co-ordination between the joint venture and Servair to be remote. In addition to the overall difficulty in coordination between the joint venture and Servair, the Commission concluded that such coordination would not be sustainable over time. The Commission cited the possibility of new entrants into the catering business and the ineffectiveness of retaliation due to the relatively long period of catering contracts (one to five years) and the staggering of contracts over time as sufficient barriers to the overall sustainability of anti-competitive coordination.
Finally, the Commission analysed the possibility of coordination between the parents of the joint venture under article 81. In so doing, the Commission stated: ‘[i]n order to establish a restriction of competition in the sense of article 81(1) of the EC Treaty, it is necessary that the co-ordination of the parent companies’ competitive behaviour is likely and appreciable and that it results from the creation of the joint venture, be it as its object or effect.’ The Commission found that preventative measures put in place by the parent companies would be sufficient to ensure that the joint venture would not serve as a conduit for information exchange or a forum to discuss operations outside the scope of the joint venture and that the parents would compete in bids involving Paris Charles de Gaulle and one or more other airports. According to the mechanism put in place by the parents, as an integral part of the joint venture agreement ‘each of the Parties shall request a quote from the JV; and in case they negotiate with the airline companies a lower price than the price quoted by the JV, they shall reimburse such price difference directly to the airline company, without the JV or the other company being informed of the price quoted’. The Commission noted that the mechanism together with the joint venture’s separation of board members from sales and marketing functions and the commitment of board members to observe confidentiality constraints would be sufficient to eliminate any threat of coordination through information sharing by the parents. It also noted that the market survey revealed a concern about further collaborative efforts in airline catering between the parents at other airports. The Commission found, however, referring to the small size of the Paris market compared to the parties’ total turnover, that the joint venture would have neither as its object or effect the coordination of the parties’ competitive behaviour. The joint venture was cleared at Phase I without undertakings.¹³
Frequently the Commission also analyses whether the relationship between the parents and the newly created structure will result in market foreclosure. This occurs where actual or potential rivals’ access to supplies or markets is hampered or eliminated as a result of the joint venture. The Commission will use a 25 per cent market share as a benchmark to assess whether a joint venture’s parents have a significant presence on any market vertically related to that in which the joint venture is active. Below that threshold vertical issues are unlikely to arise while a more detailed assessment may be required where the market shares are higher.
Several recent Commission Decisions have addressed the issue of vertical foreclosure effects. For example, in Carlyle/Ineos/JV (2007), the parents overlapped in the production of chemical products and the joint venture was active in a number of downstream markets.¹⁴ Given that the parents’ market share in the upstream market was only 30 to 40 per cent and considering that the entire demand for the upstream product in the relevant downstream market was only 2 per cent the parents and the joint venture would not be in a position to foreclose supply to the joint venture’s competitors.

Notes

1 The other categories are mergers, acquisitions of sole control and creation of full-function joint ventures. Only the latter category falls within the scope of this article, we discuss it below.
2 The ECMR, article 3.2.
3 Notice, at paragraph 76.
4 Notice, at paragraph 76.
5 Notice, at paragraph 79.
6 Article 3(4).
7 Case COMP/M.4477 Ses Astra/Eutelsat/JV, 25 July 2007
8 Case COMP/M.2149 T-Online International; TUI/C&N Touristic, 19 March 2001 and Commission, Internet Joint Ventures and the Quest for Exclusive Content: The T Online Cases, Competition Policy Newsletter (June 2002).
9 Note here that certain national merger regimes have lower ‘control’ thresholds, which may trigger national merger filings. For example, in case COMP/JV.27 Microsoft/Liberty Media/Telewest, 22 March 2000, the parties reduced one joint venture parent’s shareholding interest below the level of control, as a result of which the Commission lost jurisdiction under the ECMR. The new shareholding interest, however, was sufficiently high to give the parent ability to exercise ‘material influence’, which triggered merger investigation in the UK. The UK authority cleared the transaction.
10 See Notice, at footnote 84.
11 Article 2(4) the ECMR states ‘[t]o the extent that the creation of a joint venture constituting a concentration pursuant to article 3’ has as its object or effect the coordination of competitive behaviour it shall be assessed against article 81. Article 3 only refers to the term ‘joint venture’ in the context of full-function joint ventures and not in relation to acquisition of joint control.
12 Case COMP/M.4170 LSG Lufthansa Service Holding/Gate Gourmet Switzerland, 19 July 2006
13 See also Case COMP/M.4760 Amadeus/Sabre/JV, 12 September 2007
14 Case COMP/M.4927 Carlyle/Ineos/JV, 20 December 2007

An extract from The European Antitrust Review 2009