7.01  The emphasis of non-coordinated (or unilateral) effects analysis is not on market definition but on the elimination of the pre-merger competitive constraints that the incumbents imposed on each other. Unilateral effects arising from a merger are the same regardless of whether the merger generates high concentration within a narrow market, or causes the loss of close, direct competition between the merging parties within a broader market.

7.02  Assuming that every percentage point of market share in the market is an equal constraint on every other percentage point of market share in the market, but that every percentage point of market share outside the market is no constraint at all, is erroneous. The ‘more-effects-based approach’ that was introduced with the EUMR aimed to anticipate and gauge better the competitive issues raised by a contemplated transaction without an overly strong reliance on structural parameters. An important part of the rationale of the SIEC test and the Horizontal Merger Guidelines¹ was to provide more accurate guidance and enhanced predictability for merging firms and their advisors. The substantive test of the EUMR addresses all mergers that are likely to significantly impede effective competition.

(p. 222) 7.03  This chapter begins with a brief summary of the main types of mergers and then focuses on the anticompetitive impact of horizontal mergers in the form of non-coordinated effects. Chapter 8 focuses on the anticompetitive impact of horizontal mergers in the form of coordinated effects. Horizontal mergers are mergers between parties that operate in the same relevant market. Such mergers can increase the market power of the merging firms so that they could unilaterally induce harm to competition by imposing, inter alia, a profitable post-merger price increase. Other firms in the market might raise their prices in response, also unilaterally. Thus, rivalry might weaken. Moreover, a horizontal merger may increase the likelihood of (and/or stability and sustainability of) collusion, either tacit or explicit, between the post-merger firms in the market. This will be addressed in Chapter 8.

7.04  Non-horizontal mergers include vertical, conglomerate, and diagonal mergers. Vertical mergers are mergers between parties which operate at different levels of an industry. Such mergers, although often pro-competitive, may in some circumstances reduce competitive constraints faced by the merged firm as a result of increased barriers to entry, raising rivals costs, substantial market foreclosure, or increased likelihood of collusion. This risk is, however, unlikely to arise except in the presence of existing market power or in markets where there is already significant vertical integration as well as vertical restraints. Conglomerate mergers are mergers between firms in apparently unrelated markets which would rarely significantly impede effective competition. However, in some jurisdictions, mergers of this type have been found to create competition problems; for example, through the exercise of ‘portfolio power’.

7.05  Diagonal mergers may entail both horizontal and non-horizontal effects where the merging firms are in a vertical or conglomerate relationship, and also actual or potential competitors of each other.² One such merger was the Google/DoubleClick merger. The Commission argued that the parties were not competitors as DoubleClick was not an online advertising space provider but only a provider of ad technology (ad serving) and Google does not provide ad serving tools on its own.³ There was a concern that advertisers and publishers see text advertising and display advertising as substitutes. Complainants alleged that this created a diagonal relationship between Google and DoubleClick that would make unilateral price increases profitable for a combined entity. Lofaro and Lewis (2008) note that an increase in the price of DoubleClick’s advertiser-side ad serving solution would increase the total cost to an (p. 223) advertiser of purchasing display ad space in the non-integrated channel, to which DoubleClick’s product is an input. Since the non-integrated channel is viewed as substitutable for the integrated channel in which Google sells text advertising, there would be some diversion of demand to this channel, and some diversion of demand to Google.⁴ As Lofaro and Lewis (2008) argue, this diversion would be internalized by the combined entity, leading to an incentive to increase the price of the display ad serving solution. After a careful analysis by the Commission, it concluded that Google and DoubleClick did not exert a significant competitive constraint on one another and therefore that no unilateral price increase would be profitable post-merger.

7.06  A similar diagonal merger was the Friesland Foods/Campina⁵ merger which raised horizontal concerns in basic dairy, cheese, cream, flavoured dairy drinks, emulsions, and pharma lactose. The Commission also argued that the parties had a relatively strong position in the market for procurement of raw milk in the Netherlands (their combined market share would be in the range of 70–80 per cent). In addition, the parties’ downstream competitors needed access to raw milk as a critical input. The Commission cleared the case subject to commitments which took into account the impact of the merger on different dairy product markets (horizontal concerns), but also the need for access to the raw milk market as a necessary condition for the effectiveness and viability of divestments (vertical concerns).⁶ Anticompetitive vertical effects will generally be more likely where the merged firm has significant market power either upstream (for input foreclosure) or downstream (for customer foreclosure). Thus, in diagonal mergers, only a degree of market power significant enough to give cause for concern over horizontal unilateral effects will be sufficient to give cause for concern over anticompetitive vertical effects.

7.07  Before proceeding with the analysis of non-coordinated effects, we should emphasize that such effects may take several forms. The most usual form will be an increase in the price of the relevant products. In addition to price, competitive harm can arise in relation to non-price parameters such as the quantity produced, service quality, product range,⁷ product quality,⁸ geographical location, productive capacity, and (p. 224) innovation.⁹ The ability of firms to adjust these elements, and also the time within which they can do so, will depend upon the market concerned.¹⁰

7.08  In a nutshell, horizontal mergers can produce non-coordinated effects if the elimination of the competitive restraints that the merging firms exercised upon each other increases the combined firm’s market power. Thus, the merged group is able to profitably raise prices, reduce choice or innovation through its own acts, without the need for a coordinated response from competitors. Furthermore, non-merging firms in the same market can also benefit from the reduction of the competitive pressure resulting from the merger, since the merging firms’ price increase may switch some demand to the rival firms, which in turn may find it profitable to increase their prices. In such cases, the firms in the marketplace are not coordinating their competitive behaviour, but merely reacting to changes in each other’s behaviour. The reduction in these competitive constraints could lead to significant competitive harm in the relevant post-merger market.

7.09  By eliminating a particular rival, a horizontal merger may eliminate important competitive constraints on one or more incumbents, and may decrease the level of rivalry between them which in turn enables the merging parties to exercise market power. This type of merger can result in a new entity post-merger that would have an appreciably larger market share than its competitors. Horizontal mergers may also lead to an oligopolistic market in which the elimination of an important competitive constraint leads to substantially less competition in the market.

7.10  The elimination of competitive constraints enables merging parties, inter alia, to profitably raise prices as well as reduce output. Other firms may then find it profitable to raise prices because of some switching of demand towards their products/services. The extent of this switching depends on the closeness of competition between the merging parties themselves, and between them and other firms. This tendency for prices to rise will depend on the degree of heterogeneity between the parties’ products and those of their competitors. The more heterogeneous alternative products are the fewer consumers will tend to switch between them (that is, diversion between them), and thus the more profitable the post-merger price increase by the incumbents will be. Thus, the incumbents in the post-merger market are prone to (p. 225) increase prices unilaterally.¹¹ The stronger the substitutability between the merging products, the greater the diversion between them, and the stronger/more likely unilateral effects become. In this case, pre-merger market shares may not be a good indicator of levels of rivalry between the merging parties. In markets where products are relatively homogeneous, market shares can be a good indicator of substitutability. Where firms’ products are differentiated, pre-merger market shares may not be a good indicator of levels of rivalry between the merging parties.

7.11  Thus, the magnitude of any unilateral effects from a merger depends in part on the extent to which the merger allows the parties to recapture customers that would otherwise have been lost had either firm individually raised its prices. In general, this recapture rate will be measured by the diversion ratio between the two firms. This diversion ratio is calculated by looking at customers’ choices if they switch away from a firm that increases its price. If the products are differentiated, so that products sold by different participants in the market are not perfect substitutes for one another, a merger between firms in a market for differentiated products may diminish competition by enabling the merged firm to profit by unilaterally raising its price above the pre-merger level, without at the same time incurring substantial switching of customers from its products to rivals’ products.¹² If, on the other hand, there are a number of alternative suppliers to whom a significant number of customers are willing to turn, the threat of losing these customers may be adequate to place a constraint on the merging parties. However, product differentiation as well as the inability of competitors to react by either increasing output (if spare capacity is limited), or repositioning in order to place a constraint on the parties post-merger, is conducive to the creation of unilateral effects in the post-merger market.

7.12  As was mentioned above, non-coordinated effects can arise in concentrated markets where firms compete with differentiated products and the products of the merging firms are particularly close substitutes, so that the diversion ratio between the merging parties’ products is significant. Similarly, unilateral effects may arise if the merging parties are close competitors because of their close geographic location. In (p. 226) the case where products are differentiated based on brand image, technical specifications, quality, or level of service, customers often prefer specific suppliers. If firms producing close substitutes merge, the combined entity is more likely to increase price post-merger than if competitors that do not produce close substitutes merge. If other incumbents can alter their product lines and become close substitutes to the merged entity, the induced non-coordinated effects will be mitigated. Furthermore, in case competitors are located within close proximity, even if a relevant geographic market is relatively large, competition can be localized and the geographic location of suppliers will be a significant factor. Finally, concerns of non-coordinated effects in oligopolistic markets can also arise if the suppliers’ capacities are the main driver of competition, rather than product differentiation, and competitors would be unlikely to increase output in response to a price increase (and output reduction) of the combined entity due to capacity constraints they may be facing.¹³

7.13  In 2004, along with the adoption of the new EUMR, the Commission issued guidelines setting out the Commission’s approach to horizontal mergers, clarifying the applicability of the EUMR and explaining the economic rationale the Commission employs in assessing such mergers. The Guidelines add some additional references to the Commission’s practice regarding the EUMR’s substantive test and provide a clear legal basis and framework for non-coordinated effects (and coordinated effects) analysis.

7.14  The ‘Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings’ (‘Guidelines’)¹⁴ describe both the theoretical basis and the analytical methodology for evaluating the potential for anticompetitive effects resulting from a horizontal merger. More importantly, in line with the desire to improve its economic reasoning, the Guidelines focus on the nature of the analysis needed to identify the competitive constraints that each of the merging parties currently poses on the other. The aim is to provide greater predictability with a view to increasing legal certainty for all parties concerned.

7.15  Carlton,¹⁵ in analysing the 2010 US Merger Guidelines,¹⁶ emphasizes the importance of empirical evidence in overcoming enforcement agencies’ concern about (p. 227) the role of competition in constraining post-merger price increases and other such conduct. Carlton argues that the distinction between ‘unilateral’ and ‘coordinated’ effects is artificial and should be de-emphasized. He adds that historical evidence about the impact of entry or large buyers with significant buying power on preserving post-merger competition, and about the firms’ past track record in achieving cost savings should be given significant weight in merger analysis. Third, he argues that greater emphasis should be placed on merger-related savings in fixed costs as efficiencies. Fourth, there should be a well-established basis for any numerical thresholds for safe harbours in guidelines (for example, HHI). Fifth, the difficulty of anticipating future changes in competitive conditions should be stressed and it should be acknowledged that merger enforcement should not be based on speculative assessments of future conditions. Sixth, the approach to geographic market definition should be revised, shifting the focus of the analysis from one using supplier locations as a starting point to one based on the competitive alternatives faced by consumers at different geographic locations. Finally, Carlton adds the use of the term ‘maverick’.¹⁷

7.16  The Guidelines explore the possible anti-competitive effects of horizontal mergers and consider the main ways in which a horizontal merger may significantly impede competition. By eliminating the competitive constraints between the parties, a horizontal merger may allow the merged firm to increase its prices regardless of the response of its remaining competitors, and thus may lead to non-coordinated effects.

7.17  The Guidelines’ discussion of ‘non-coordinated’ effects begins with an introduction that distinguishes non-coordinated from coordinated effects. The Commission will examine whether a proposed transaction will allow the merged entity to raise prices profitably (non-coordinated or unilateral effects). These non-coordinated effects typically arise where the combined entity has an appreciably larger market share than the next competitor post-merger.¹⁸ A significant impediment to effective competition can also result from a reduction of the competitive constraints that the merging parties previously exerted on each other, which could include a situation in which the combined entity does not have the largest share in the relevant market.¹⁹(p. 228) The Commission will also examine whether a proposed merger in a concentrated market will increase the likelihood that companies will coordinate their behaviour (coordinated effects).

7.18  The distinction between non-coordinated and coordinated effects rests on whether the analysis focuses on a static assessment of the conduct of the merged entity and the resulting impact on rivals’ conduct (non-coordinated effects), or a dynamic assessment of the interrelation of the post merger entities (coordinated effects). As Carlton states (referring to the US Guidelines, but the argument also applies to the Commission’s Guidelines), the US Guidelines appear to suggest that unilateral and coordinated effects analysis may be distinguished based on whether the merging firms are each other’s next best substitutes and whether rivals hold prices constant in response to price changes by the merged firm. Interestingly he notes that there is no need for an anticompetitive merger to be limited to firms that are each other’s next-best substitutes and to the reaction of rivals. Carlton adds that the distinction between unilateral and coordinated effects is artificial and provides a misleading view of the economic theory. He notes that both unilateral and coordinated effects analyses should properly be understood as variations of oligopoly theory which provide the general theoretical basis for any concerns about the potential adverse effects of mergers on consumers.²⁰

7.19  The Guidelines present a non-exhaustive list of factors, which taken separately are not necessarily decisive, but may influence the assessment of non-coordinated effects that may result from a merger. These factors, which are analysed in the Guidelines are whether:²¹

• •  large market shares are held by the merging firms;

• •  the merging firms are close competitors;

• •  customers have limited possibilities of switching between suppliers;

• •  competitors are unlikely to increase supply if prices increase;

• •  the merged entity is able to hinder expansion by competitors; and

• •  the merger eliminates an important competitive force.²²

7.20  Market shares and the structure of the market may provide an indication of whether the merger would be likely to lead to non-coordinated effects. In general, the larger the market share, the more likely a firm is to possess market power. And the larger the addition of market share, the more likely it is that a merger will lead to a significant increase in market power. As the Commission states in its Guidelines, market shares and concentration levels provide useful first indications of the market structure and of the competitive importance of both the merging parties and their competitors.²³ Estimated competitive shares in a properly defined market are pertinent to assessing the market structure and the competitive importance of both the parties and the competitors.

7.21  The Guidelines institute a safe harbour. Concentrations which by reason of the limited market share of the undertakings concerned are not liable to impede effective competition may be presumed to be compatible with the common market. This presumption applies where the market share of the undertakings concerned does not exceed 25 per cent. In the history of EUMR enforcement, the Commission has applied a filtering threshold at around 50 per cent level (with exceptions), but in recent years gradually expanded its substantive jurisdiction by scrutinizing mergers which would historically have been approved.²⁴ Market shares of the merged entity between 25–40 per cent usually do not lead to any concerns,²⁵ whereas market shares of higher than 40 per cent are more likely to be found to raise competition concerns.²⁶ However, it should be emphasized that there is no presumed safe harbour for levels of market share of the merged entity that are higher than 25 per cent.

7.22  It is also helpful to compare the combined market share with those of other market players. If the competitors’ market shares are markedly lower than the merged entity’s market share, this is an indication of the ability of the merged entity to (p. 230) adopt anticompetitive conducts, while a small gap between the market shares may not prove lack of ability by the merged entity. A merger that gives a combined market share of 55 per cent with an increment of just 5 per cent is not as likely to give cause for concern over unilateral effects as a merger that gives a combined market share of 75 per cent also with a 5 per cent increment, because the latter market is more concentrated.

7.23  The Commission has cleared mergers leading to significant combined market share with a large difference to the second largest firm in the market. This illustrates that merger assessment is a holistic process and no assessment factor can individually lead to a prohibition or clearance of a merger. In Courtaulds/SNIA,²⁷ the Commission found that despite a difference in market shares of 45 per cent between the merged entity and its nearest competitor, the merger was not prohibited as there were other important factors such as barriers to entry that rendered anticompetitive behaviour unlikely. In Enso/Stora²⁸ the Commission found that despite the merged entity having 60 per cent market share, it would not be dominant as it would be faced with a very concentrated buying sector. The Iberia/British Airways merger²⁹ led to substantial overlaps on routes between London and Spain with combined market shares between 70 per cent and 80 per cent. The Commission cleared the merger without remedies on the basis of considerable competitive constraints exerted by Easyjet and Ryanair as well as on the basis of the continuation of the close cooperation between British Airways and Iberia as part of One World alliance in the absence of the merger.

7.24  However, in a recent prohibition, in the merger Olympic Air/Aegean Airlines,³⁰ the two largest Greek airlines, led to 80 per cent of flights and 90 per cent of capacity on the Greek domestic market, and in the absence of viable remedies the Commission prohibited the merger. The European Commission prohibited the TNT Express/United Parcel Service merger, both major players in the small package delivery sector. The Commission’s investigation indicated potential competition concerns in the markets for small parcel delivery services, in particular international express services, in a number of member states (especially South European countries), where the parties would have very high combined market shares. UPS and TNT Express were two out of the only four so-called ‘integrators’ currently operating in Europe.³¹

7.25  The Commission’s initial investigation has shown that small package delivery services form a highly differentiated market which can be divided into several (p. 231) segments, depending notably on the committed delivery time frame associated with the services. It appears that other integrators would be the only significant competitive constraint on the merged entity for most express services, especially for the fastest time-commitment deliveries. As the proposed transaction would reduce the number of integrators competing in the EEA from four to three, the competitive constraint on the merged entity would be significantly reduced. This would lead in many member states to a highly concentrated market for domestic and, even more so, for international express delivery services.³²

7.26  The Commission calculates market shares on the basis of sales revenue,³³ volume,³⁴ or capacity.³⁵ The market shares of the merging parties’ licensees or franchisees should be attributed to the parties if the licensees or franchisees are unable to exert any competitive pressure on the merging parties or to enter the market as independent competitors, (for example, due to the existence of a long-term exclusive license/contract).³⁶ Production of vertically integrated firms that is aimed for internal use is not taken into account for the calculation of market share unless the ‘captive production’ can be easily redirected for sale to third parties.³⁷ In addition, the Commission has accepted merging parties’ arguments that their market shares had recently declined and would most likely continue to decline.³⁸ In mergers where the markets operate on the basis of firms competing for long-term supply contracts, or where the market is characterized by intense innovation,³⁹ the Commission, rather than take a static approach and focus on the combined market shares at a particular point in time, will assess average market shares over a period of time as well as the capacity, or success in the bids of the incumbents if the contracts are awarded through tenders.⁴⁰ Thus, in markets characterized by large and infrequent orders (p. 232) made by a small number of customers, it might be useful to analyse market shares over a period longer than one year using historic data, and investigate any variance in market shares over time.

7.27  The concentration level in a market may also provide useful information about the competitive situation. In order to assess concentration the Commission often applies the Herfindahl-Hirschman Index (HHI). The HHI gives proportionately greater weight to the market shares of the larger firms. Although it is best to include all firms in the calculation, lack of information about very small firms may not be important because such firms do not affect the HHI significantly. The HHI is calculated by summing the squares of the individual market shares of all the participants. Unlike the four-firm concentration ratio, the HHI reflects both the distribution of the market shares of the top four firms and the composition of the market beyond that of the top four firms. However, it gives proportionately greater weight to the market shares of the larger firms, in accordance with their relative importance in competitive interactions. In the US guidelines, markets can be broadly characterized as unconcentrated (HHI below 1,500), moderately concentrated (HHI between 1,500 and 2,500), and highly concentrated (HHI above 2,500).⁴¹ However, a DOJ/FTC study on HHIs in merger challenges between 1999 and 2003 shows that from nearly 1,300 mergers challenged by US agencies over these five years no merger with post-merger HHI below 1,400, and change in the HHI of approximately 85 was challenged.⁴²

7.28  While the absolute level of the HHI can give an initial indication of the competitive pressure in the market post-merger, the change in the HHI (known as the ‘delta’ or ‘dHHI’) is a useful proxy for the change in concentration directly brought about by the merger. The Commission is unlikely to identify horizontal competition concerns in a market with a post-merger HHI below 1,000. The Commission is also unlikely to identify horizontal competition concerns in a merger with a post-merger HHI between 1,000 and 2,000 and a delta below 250, or a merger with a post-merger HHI above 2,000 and a delta below 150, except in exceptional circumstances. The Guidelines present such circumstances: where a merger involves a potential entrant or a recent entrant with a small market share, one or more merging parties are important innovators in ways not reflected in their market shares, there are significant cross-shareholdings among the market participants, one of the merging firms is a maverick firm, there are indications of past or ongoing coordination, or facilitating practices, and where one of the merging parties has a pre-merger market share of 50 per cent of more.⁴³

(p. 233) 7.29  Notwithstanding complexities of market definition where products are differentiated, this de minimis market-share guideline acknowledges that market shares can still provide a reasonable initial basis for assessing whether a merger raises serious competition concerns. In the presence of highly differentiated products, if, even within a very narrowly defined relevant market, the combined share of the two merging parties is relatively low, it is unlikely that the merger will give rise to serious competition concerns. But in these markets the discretion for defining narrow markets is also at its greatest, so this part of the Guidelines again falls short of providing useful guidance.

7.30  In an empirical study of 16 Phase I conditional clearances, and four Phase II decisions which were cleared, Martinez et al. assessed the probability of prohibition in relation to the market share. If a concentration does not raise barriers to entry, a small acquisition, with 10 per cent increase in market share, will lead to the conclusion that the probability of non-clearance is 14 per cent. A 27 per cent increase in market share will lead to a very high probability of non-clearance (75 per cent). When the increase in market share is around 38 per cent, the merger will almost certainly be prohibited (the probability of non-clearance reaches 95 per cent).⁴⁴

7.31  In the assessment of unilateral effects in differentiated products mergers the link between the market shares and competitive effects may easily be weak.⁴⁵ As the Guidelines note,⁴⁶ products may be differentiated in various ways, in terms of geographic location, based on branch or stores location, in terms of brand image, technical specifications, quality, or level of service. In such markets the higher the degree of substitutability between the merging firms’ products, the more likely it is that the merging firms will raise prices significantly.⁴⁷ Indications of the degree of substitutability include customer preference surveys, estimation of the cross-price elasticities of the products involved, diversion ratios,⁴⁸ and analysis of purchasing patterns.

7.32  Closeness of competition denotes the trend of customers of one of the merging parties to switch to products supplied by the other merging party. In markets where products are differentiated by brand or geographic proximity (that is, horizontal (p. 234) differentiation) or where firms are differentiated by quality (that is,vertical differentiation), unilateral effects depend on the degree to which sales lost as a result of a price increase of one merging party’s product divert to the other merging party’s product.

7.33  The magnitude of any unilateral effects from a merger depends in part on the extent to which the merger allows the parties to recapture customers that would otherwise have been lost. The greater this diversion between the two merging parties, the closer competitors the two firms are said to be and the more harmful the merger is likely to be for competition. It is widely recognized that diversion ratios can provide a useful tool when analysing the unilateral effects of a merger. Thus, the diversion ratio between the merging parties is the proportion of switching customers in response to an increased price/deteriorated non-price offer (such as service quality) that would divert customers from one merging party to the other. It also suggests that the merged entity, having internalized this rivalry, would have the incentive to raise prices or deteriorate its non-price offer. When firm A loses sales as a result of any given price increase, the diversion ratio from A to B measures the proportion of those lost sales that are captured by B. Following a merger between A and B, the merged firm takes account of the fact that B recaptures part of the sales lost by A, and this relaxes the constraints on the pricing decisions of the firm. The higher the rate of diversion between A and B, the stronger is the pre-merger constraint that the firms exert on one another and, thereby, the greater the risk of a significant impediment to effective competition.⁴⁹ Especially in differentiated product markets, the merged firm can profit by unilaterally raising the price of one or both products above the pre-merger level, without at the same time incurring substantial switching of customers from its products to rivals’ products.⁵⁰

7.34  In bidding markets the Commission will assess whether historically the submitted bids by one of the merging parties have been constrained by the presence of the other merging party.⁵¹ The Commission will also assess the possibility of product repositioning by competitors or the merging parties and the impact this repositioning may have on the merged entity.⁵² Such analysis (also called win/loss analysis) to assess closeness of competition can be undertaken in such markets. If the two merging firms represent the closest competitors for one another, then one would expect the data to show that in those bids won by one of the merging parties, the (p. 235) other is usually the runner-up to a much greater extent than would be suggested by considering market shares alone. In most tenders, it is the second-placed bidder that effectively determines the price that is the result of a bidding contest. There is no incentive for the winning firm to set the absolute lowest price if its marginal cost levels allow for whether a higher price would also suffice to win the deal.⁵³

7.35  In GE/Instrumentarium,⁵⁴ the parties presented a win/loss analysis for each of the affected products and for each European country. In the relevant markets for critical care monitors, C-arms, and mammography devices, the Commission accepted the results of these analyses as providing evidence that the two parties were not the closest competitors. In Cadbury/Kraft⁵⁵ the product markets were defined as chocolate confectionary, which included separate market segments such as tablets and pralines. The Commission required divestments of Cadbury’s chocolate business in Poland and Romania where the combined market share of Kraft and Cadbury would have reached 60–70 per cent, and Cadbury and Kraft were close competitors. Even though there were high market shares in the United Kingdom and Ireland, the Commission did not require divestitures as Cadbury and Kraft were not very close competitors.⁵⁶ In the Unilever/Sara Lee Body Care⁵⁷ merger the Commission considered that the merger might give Unilever the ability to raise prices for deodorants in some of its existing brands as post-merger Sara Lee’s Sanex would no longer exert any competitive pressure on Unilever’s close substitutes, Dove and Rexona. The merger therefore raised serious competition concerns in a number of member states and was authorized only subject to the divestiture of Sanex.

7.36  The Guidelines also explain that in differentiated product markets, mergers between firms producing non-close substitute products are unlikely to restrict competition. The primary concern should be whether the merging parties’ products are close substitutes, and whether the loss of competition between the merging parties will enable the merged entity, as well as rival firms, to raise prices. The stronger the substitutability between the merging products, the greater the diversion ratio between them, and the stronger/more likely unilateral effects become. In this case pre-merger market shares may not be a good indicator of levels of rivalry between the merging parties. In addition, the competitor’s ability to raise prices will depend on the substitutability of their products with the ones of the merged entity and on the barriers to expansion that they may face. In VNU/WPP⁵⁸ the Commission commented, obiter, a reduction in the number of credible competitors from three to two (p. 236) would be likely to raise strong competition concerns, because it would leave only one alternative bidder in cases where the customer has a preference for changing to a new supplier.

7.37  The Guidelines state that customers of the merging parties may have difficulties switching to other suppliers because there are few alternative suppliers or because they face substantial switching costs.⁵⁹ Evidence of past customer switching patterns and reactions to price changes may provide important information in this respect. The more limited switching is, the more harmful the merger, ceteris paribus, is likely to be.

7.38  When market conditions are such that the competitors of the merging parties are unlikely to increase their supply substantially if prices increase, the merging parties may have an incentive to reduce output below the combined pre-merger levels, thereby raising market prices.⁶⁰ Output expansion is unlikely when competitors face binding capacity constraints and the expansion of capacity is costly or if existing excess capacity is significantly more costly to operate than capacity currently in use. There may be mergers that may lead to anticompetitive effects in post-merger markets when such markets are characterized by homogeneous products and by capacity constraints faced by the incumbents. In such cases, in the post-merger market the merged entity as well as the rivals may raise prices and no firm will be able to increase capacity in order to absorb the switching of demand due to some firms’ increase in prices. Although capacity constraints are more likely to be important when goods are relatively homogeneous, they may also be important where firms offer differentiated products.

7.39  Inability of competitors to increase supply—for example, because they produce at full capacity—indicates that the merged entity’s customers who may wish to switch to rivals’ products as a result of deterioration in the price or non-price parameters in the post-merger market will not be in a position to achieve such switching. Thus, these customers are ‘captive’ within the merged entity, which can induce the latter to increase prices, lower quality, etc.

7.40  In some cases, the merged firm is in a position where it would have the ability and incentive to make the expansion of smaller firms and potential competitors more difficult or otherwise restrict the ability of rival firms to compete. Competitors may not be in a position to constrain the merged entity to such a degree that it would not (p. 237) increase prices or take other actions detrimental to competition. The Commission in the Guidelines identifies a number of factors that can grant the merged entity the ability to hinder expansion by competitors such as a degree of control, or influence over, the supply of inputs or distribution possibilities that expansion or entry by rival firms may be more costly.⁶¹

7.41  The term ‘maverick’ denotes firms that are very competitive in a market and can induce a higher degree of competition dynamics in the markets they operate. Such firms may have more of an influence on the competitive process than their market shares would suggest. A merger involving a maverick may change the competitive dynamics in a significant, anticompetitive way, in particular when the market is already concentrated.⁶² The removal of an important competitive force can slow the market dynamics and can lead to post-merger markets that will not exhibit the same degree of competition as was the case with the presence of the maverick firm.

7.42  In markets where innovation is an important competitive force, a merger between two important innovators may induce anticompetitive effects in the market. However, in such markets a merger increases the firms’ ability and incentive to bring new innovations to the market. Thus, merger assessment in such markets needs to take place on a case-by-case basis. In GE/Instrumentarium⁶³ and Siemens/Drägerwerk⁶⁴ the Commission concluded that GE and Siemens were key players that exercised important competitive constraints pre-merger despite the fact that their market shares were below 10 per cent, which the Commission usually considered to be a non-appreciable market share increment.

7.43  According to the network effect theory, a good or service is valuable to a customer depending on the number of customers already owning that good or using that service. Each new user of the product derives private benefits, but also confers external benefits on existing users. The latter impact amounts to an externality arising from the operation of networks (network externalities).⁶⁵ Furthermore, the larger the (p. 238) number of customers using a product, the more valuable the product for the next potential customer. Networks may not reach optimal size because users fail to take account of external benefits. Markets in which incompatible standards compete may ‘tip’ in the direction of a standard that achieves an early advantage.⁶⁶

7.44  Direct network effects have been defined as those generated through a direct physical effect of the number of purchasers on the value of a product; for example, computer printers. Indirect network effects are ‘market mediated effects’ such as cases where complementary goods (for example, toner cartridges) are more readily available or lower in price as the number of users of a good (computer printer) increases.⁶⁷

7.45  Examples of markets where network effects are important include, inter alia, stock exchanges, software markets, and car markets. Network externalities are characteristic for the trading industry. Externalities concern both investors and market institutions. In general, networks exhibit positive consumption and production externalities. A positive network externality implies that the value of a unit of the good increases with the number of units sold. There are many products for which users’ utility increases with the number other users consuming the good.⁶⁸ The essential relationship between the components of a network is complementary.

7.46  Networks exhibit positive size externalities, since a product’s value to the user increases as the number of users of the product grows. As a direct consequence of their self-reinforcing nature, networks frequently exhibit positive critical mass. Network effects become significant after a certain subscription percentage has been achieved, called critical mass. At this subscription percentage, the value obtained from the good or service is greater than or equal to the price paid for the good or service. Thus, ‘no network of size smaller than this positive size, called critical mass, is ever observed, at any price’.⁶⁹ However, certain networks become congested beyond critical mass, and thus future customers may seek alternatives.

7.47  In particular markets, firms may be very reluctant to change their way of operation, especially if they have to pay the costs of transition. Thus, the self-reinforcing nature (p. 239) of networks creates switching costs for the existing customers. The existence of positive critical mass often means that in the presence of one network, a differently organized one may not even exist. The first company to attain sufficient critical mass in a market sector characterized by network effects can expect to benefit from a ‘snowball effect’ that will reinforce its position in the market. This is ‘tipping’. This means that the factors considered for unilateral effects assessment may not be fully informative. By acquiring market share in a sector characterized by strong network externalities, a leading player may be able to obtain a dominant position even when the static market share following the acquisition would not ordinarily be sufficient to indicate possible harm. Thus, in markets with large positive network externalities, competition can be for the market but not in the market.⁷⁰

7.48  Such network markets are the markets in which stock exchanges are operating. The European Commission prohibited the merger between Deutsche Börse and NYSE Euronext,⁷¹ as it would have resulted in a quasi-monopoly in the area of European financial derivatives traded globally on exchanges. Together, the two exchanges would control more than 90 per cent of global trade in these products. The Commission’s investigation showed that new competitors would be unlikely to enter the market successfully enough to pose a credible competitive threat to the merged company. The two companies offered in particular to sell Liffe’s European single stock equity derivatives products where these compete with Eurex. However, the divested assets would be too small and not diversified enough to be viable on a stand-alone basis. In the commercially more significant area of European interest rate derivatives, the companies did not offer to sell overlapping derivatives products, but only offered to provide access to the merged company’s clearing for some categories of ‘new’ contracts. This was considered as insufficient, in particular because it did not extend to existing competing products. There were also fundamental concerns about the workability and the effectiveness of such an access remedy.⁷²

7.49  The Commission considered that the listing decision of issuing companies depends on external factors such as regulation, investor base, business strategies as well as on elements influenced by the exchanges themselves (such as listing fees, trading services), and the liquidity of the trading venue. It adds that a platform provider would not start offering a trading service for a new equity type, however easy it may be, if it does not make economic sense, for instance, because it would be difficult to attract (p. 240) sufficient liquidity.⁷³ The Commission argued that Eurex, operated by Deutsche Börse, and Liffe, operated by NYSE Euronext, are the two largest exchanges in the world for financial derivatives based on European underlyings and are each other’s closest competitors. The proposed merger would have eliminated this global competition and created a quasi-monopoly in a number of asset classes, leading to significant harm to derivatives users and the European economy as a whole. The Commission also assessed non-price factors and argued that there would also be less innovation in an area where a competitive market is vital for both SMEs and larger firms.

7.50  This analysis illustrates that there are a number of factors that are taken into consideration in the assessment of non-coordinated effects from a merger. As far as the relevant importance and weight of the assessment criteria are concerned, Lindsay, Lecchi and Williams (2003)⁷⁴ examined the factors which the Commission took into consideration under the EUMR in assessing whether a merger would lead to the creation or strengthening of dominance. The decisions they used in their sample were adopted between 1 January 2000 and 30 June 2002. The authors concluded that market shares are very significant in determining whether a merger will be cleared by the Commission, although the Commission has argued that the market shares are only initial indicators of the likely adverse effects on competition of a merger. Low barriers to entry contribute to the clearance of concentrations with high market share, while buyer power does not seem to be too significant. The geographic coverage influences the assessment if the Commission. Finally, the HHI was not significant but the authors argued that this might be attributed to inaccuracies in the data.

7.51  Bergman et al., using a sample of 96 mergers notified to the Commission, analysed the Commission’s decision process. They found that the probability of a Phase II investigation and of a prohibition of the merger increases with the parties’ market shares. The probability also increases where the Commission finds either high entry barriers or that post-merger collusion is easy. The nationality of the merging firms or the identity of the commissioner did not seem to affect the Commission’s decisions.⁷⁵ Lindsay et al. (2003) argue that the probability of prohibition is significantly explained by market share which is contrary to the Commission’s rhetoric on the importance of market shares in merger assessment.⁷⁶

(p. 241) 7.52  Martinez Fernandez et al. (2008) analysed a sample of 16 Phase I conditional clearances, and four Phase II decisions which were cleared.⁷⁷ They find that only an increase in market share is significant and that the Commission, ceteris paribus, is biased to prohibit mergers involving market leaders. They add that the Commission clears mergers when the analysis shows other competition-reinforcing elements that mitigate the harmful effects of significant market shares. High barriers to entry increase the probability of prohibition and they add that there is no prejudice against American or Canadian mergers. Their analysis also indicates that mergers including vertical effects are not significantly discriminated against. This is in accordance with economic theory which suggests that vertical mergers are likely to be more pro-competitive than horizontal mergers. Interestingly the authors argue that there is an incentive for companies to cheat as they are going to be penalized only 1 per cent of the aggregate turnover while the potential benefits of such malicious behaviour (clearance of the merger) might outweigh the costs of being fined.

7.53  Meier-Rigaud and Parplies (2009) analyse merger enforcement in the first five years of the EUMR and derive some interesting conclusions.⁷⁸ They find a reduced number of prohibition outcomes which could indicate that improved predictability has enabled firms to improve their prediction of the outcome of merger assessment and pursue deals that would overcome this hurdle. A potentially worrying sign for the authors is that clearance decisions have been adopted in heavily concentrated markets in which the Commission subsequently launched ex post interventions under Article 102, sector enquiries (for example, in the energy and financial services sectors) or ad hoc sectoral interventions (as in the mobile telephony market). The authors add that the Commission’s enforcement activity measured by interventions including prohibition decisions, clearances with remedies in Phase II and Phase II withdrawals has fallen by more than half between 1994–2003 (Regulation 4064/89, ECMR) and 2004–2009 (Regulation 139/2004, EUMR), from 4.6 per cent of all notified cases to 2.2 per cent. The authors also note that the majority of appeals against Commission interventions in merger cases have been rejected but the appeals statistic highlights the asymmetric litigation risk faced by the Commission in merger procedures. The authors find that an appeal arising by the merging parties and their competitors is likely, while at the same time there is virtually no appeal risk from adopting a clearance decision in cases where only final consumer interests are harmed.

7.54  We should emphasize here that there is no one single factor that would be determinative in the clearance or prohibition of a merger. The Commission assesses all the factors and concludes on the competitive harm of a merger. In quite a few merger cases the factors may be pointing to contradictory conclusions (some may indicate (p. 242) absence of competition harm while some may indicate competition harm) but the Commission will use its discretion in concluding whether a merger needs to be cleared, cleared with remedies, or prohibited.

7.55  Horizontal mergers leading to non-coordinated effects in oligopolistic markets are, as the name suggests, horizontal mergers that induce non-coordinated effects in markets that are oligopolistic, thus they constitute a subset of the mergers that induce non-coordinated effects. The Guidelines do not differentiate between these two theories of harm, and the assessment of the Commission does not differ either. We present here this theory of harm for two reasons, one justifying the reform in the EUMR in 2004 and the other, and more relevant for this chapter, presenting the assessment of the Commission in a number of seminal non-coordinated effects cases.

7.56  The main substantive reason for the reform of the merger regulation in EU was the gap in the application of the dominance test. The Recast EUMR explicitly recognized the concept of non-coordinated effects in oligopolistic markets (or non-collusive oligopolies) as a result of the prevailing perception that some mergers could lead to a harmful effect on competition that could not be addressed using the concepts of single firm and/or collective dominance pursuant to Regulation 4064/89. The ‘gap’ corresponds to the situation where the post-merger entity’s market share falls below the level required for dominance, and where the merger may nonetheless still lead to unilateral effects. Until the adoption of the EUMR in May 2004, there was no published decision under the ECMR that unequivocally challenged the phenomenon of non-coordinated effects in oligopolistic markets.⁷⁹ This inability of the dominance test of Regulation 4064/89 to address such mergers was deemed the ‘gap’ in the dominance test and these mergers were deemed to be ‘gap’ mergers.⁸⁰

7.57  The legal substantive test in the EUMR, the SIEC test, is intended to fill the perceived gap in the application of the dominance test which was illustrated by cases (p. 243) such as Airtours/First Choice⁸¹ and Heinz.⁸² The term ‘non-coordinated effects in oligopolistic markets’ was introduced in Recital 25 of the EUMR. The EUMR explicitly recognized the concept of non-collusive oligopolies as a result of the prevailing perception that some mergers could lead to a harmful effect on competition that could not be addressed using the concepts of single firm and/or collective dominance that were employed in Regulation 4064/89.

7.58  The ‘gap’ corresponds to the situation where the post-merger entity’s market share falls below the level required for dominance and where the merger may nonetheless still lead to unilateral effects. The two terms ‘mergers leading to non-coordinated effects in oligopolistic markets’ and ‘non-collusive oligopolies’ refer to situations where the remaining firms in the post-merger market have the incentive and ability to adopt conduct inducing an adverse impact on competition, and thus profit from exerting their market power in the post-merger market, without being dependent upon a coordinated response on the part of the other members of the oligopolistic market structure.⁸³

7.59  Gap cases are mergers that lead to a significant impediment to effective competition by reducing the competitive constraints not only between the merging parties but also the competitive constraints exerted on the rival firms in the market.⁸⁴ In the case of such a merger, in the post-merger market the oligopolistic interdependence is altered and thus the incumbents may adopt a pricing or production structure which may have an adverse impact on competition. The incumbents will find it individually profitable to increase prices taking into account the higher prices charged by the merged entity. The most direct impact on competition will be the elimination of the competitive constraints that the merging firms exerted on each other prior to the merger. In addition, non-merging firms can also benefit from the reduction of competitive pressure that results from the merger since the merging firms’ price increase or output reduction may induce the switching of some demand to the rival firms, which, in turn, may find it optimal to increase prices. This might (p. 244) happen in particular in differentiated product markets⁸⁵ where a merger can lead to incentives for conduct having an adverse impact on competition, without creating a single leading player, and without significantly increasing the feasibility of tacit collusion. The latter situation, which could not be dealt either as single-firm dominance or as collective dominance, pursuant to the Regulation 4064/89, is known as the ‘gap’ in the application of the dominance test.⁸⁶

7.60  Mergers that are deemed ‘gap’ mergers lead to non-coordinated effects if they result in the elimination of important competitive constraints on one or more firms. The post-merger market structure will be oligopolistic and comprise a small number of firms, whose products are not close substitutes.⁸⁷ In addition, the competitors of the merged entity are not likely to be in a position to expand their capacity or switch their production, and new firms have no incentives to enter the market in the presence of an increase in the price or a reduction in the quantity produced by the merged entity.

7.61  As regards market shares, for non-coordinated effects in oligopolistic markets, what one would expect to find is market shares in the post-merger market that are similar to a market which is prone to coordinated effects. Thus, the post-merger firm may not be the one with the highest market share. In addition, the post-merger firm may have the highest market share, but that market share may not be adequate for single-firm dominance to be established, as would have been the case under Regulation 4064/89.

7.63  However, although there were no explicit decisions dealing with gap mergers, the Commission’s decision in the Airtours/First Choice⁹¹ case was controversial and ambiguous. It was not entirely clear what the Commission was arguing in this decision, but one possibility was that it was a so-called ‘gap’ case: the Commission argued that the merger would lead to collective dominance, but, depending on how one interprets the decision, it was possible that it considered that the merger would lead to non-coordinated effects in an oligopolistic market.⁹²

7.65  There are several scenarios and examples regarding the features of market structures voiced in the academic literature where, in the post-merger market, a merger may lead to non-coordinated effects without the post-merger firm enjoying a situation either of dominance⁹³ or coordinated effects.

7.66  Sir Derek Morris⁹⁴ suggested that where a new entrant appears with advanced technology and innovative ideas and one of the non-dominant companies buys this new entrant, the acquisition may not lead to the creation or strengthening of dominance. However, the merger may not only have a significant adverse impact on potential competition but may lead to a significant impediment to competition by creating the firm with the second-highest market share. Whish (2002)⁹⁵ refers to a market structure where a three-to-two merger in a market with a small number of players may not lead to single-firm dominance, and the post-merger market structure does not seem to be prone to collective dominance. However, there is a possibility that the remaining firms in the market may be able unilaterally to increase prices or decrease the quantity of their products; thus, their conduct may have an (p. 247) adverse impact on competition. Such mergers can produce non-coordinated effects in oligopolistic markets if non-merging firms in the same market can also benefit from the reduction of the competitive pressure resulting from the merger, since the merging firms’ price increase may switch some demand to the rival firms, which, in turn, may find it profitable to increase their prices. This can result in them increasing prices in the post-merger market. In such cases, the firms in the marketplace are not coordinating their competitive behaviour, but merely reacting to changes in each other’s behaviour. Although the dominance test will not prove useful in blocking this type of merger, the SIEC test will.

7.67  Additional examples of non-collusive mergers include a merger between two small players in a market where a third firm is dominant,⁹⁶ a merger in a stable oligopoly where one firm buys a smaller rival or entrant, as well as a merger between small players that make the market more symmetric and in which the merged firm does not join the ‘big’ players.⁹⁷

7.68  In order to address the problem that there are few, if any decisions, adopted explicitly on the basis of non-coordinated effects in oligopolistic markets, this chapter will include a comparative approach of examining mergers assessed under the EUMR. When the decisions in these merger cases were taken, the concept of non-collusive oligopolies was not recognised in the EUMR.⁹⁸ However, the market features that contribute to non-collusive oligopolies might have been in existence and prevalent in these market structures. Hence, this chapter will provide evidence of non-collusive oligopolies and thus confirm the need to reform the substantive test of EUMR. In addition, the analysis of these cases, especially the ones after the adoption of the SIEC, illustrate the approach of the Commission to assessing non-coordinated effects. The cases that will be analysed include Oracle/PeopleSoft,⁹⁹ Syngenta CP/Advanta,¹⁰⁰ Johnson and Johnson/Guidant,¹⁰¹ and T-Mobile Austria/Tele.ring.¹⁰²

7.69  In Oracle/PeopleSoft the Commission found that the relevant product markets for the assessment of the transaction were the markets for high-function Financial Market Software (FMS) and human resources (HR) software applications. As HR and FMS applications were not substitutable for buyers, the Commission considered these as two distinct markets. In addition, the market investigation showed that characteristics of high-function HR and FMS software were different from those of mid-market products. The market investigation clearly indicated that the geographic scope of the markets for high-function HR and FMS solutions for large and complex enterprises was worldwide.

7.70  In order to assess whether or not the notified concentration would lead to non-coordinated effects, the Commission conducted a bidding study, and investigated to what extent the competitive situation of a particular bid (measured by the number of final round bidders) had an impact on the discounting offered by the seller in question (that is, PeopleSoft in PeopleSoft’s dataset and Oracle in Oracle’s datasets). The Commission found that there was a very strong relationship between the size of the deal and the discount offered. However, once the size of the deal was taken into account in the analysis, the number of final bidders no longer provided any additional explanatory element over the discount offered and no general pattern emerged regarding the presence of a particular competitor prompting high discounts.¹⁰⁴

7.71  A finding, as in this case, that the number and identity of competitors in a given bid appeared not to have an effect on a firm’s behaviour did not in itself prove the lack of harmful effects of the merger on customers. There might have been a variety of reasons why such an effect was absent from the bidding data. One reason for the absence of any effects in the data could be that the quality of the data might have been low or that it might have been characterized from a bias in selection. Another could be that Oracle (and its competitors), when deciding what to bid, did not consider the information about actual competitors sufficiently reliable to want to base its behaviour on it. A third reason could be that the identity of bidders in the final round was an incomplete picture of the actual competitive process.¹⁰⁵ The absence of an appreciable effect of competition on Oracle’s behaviour made the bidding data unsuitable to rely on as determinative proof of an adverse effect of the merger on competition.

(p. 249) 7.72  Based on an amended market definition (after documents of the US proceedings became available to the Commission), the Commission could not conclude that the merger would lead to a collective dominant position of a combined Oracle/PeopleSoft and SAP. The other vendors, Lawson, Intentia, IFS, QAD, and Microsoft, appeared to be suitable alternatives as the Commission’s data and the dataset submitted by Oracle after the oral hearing showed that those vendors had won bids for software in the relevant markets. In addition, the highly differentiated nature of the HR and FMS high-function software,¹⁰⁶ the lack of structural links, the asymmetric market shares, the large number of incumbents, the limited transparency and inadequate retaliatory mechanisms made the achievement of a tacitly collusive outcome very unlikely. The Commission declared the concentration compatible with the common market.

7.73  Baxter and Dethmers (2005)¹⁰⁷ argue that the Commission relied upon unilateral effects and the analysis was devoid of any single dominance considerations. The DOJ reviewed this transaction under a unilateral effects standard and dismissed coordinated effects. In addition, as Ehlermann, Völcker, and Gutermuth (2005) mention,¹⁰⁸ in this case, the Commission appears to have viewed non-coordinated effects as a category of collective dominance. The Commission actually tried to close the enforcement gap in the application of the dominance test in this case, by stretching the concept of collective dominance. The market structure, as mentioned, was not conducive to collective dominance and the fact that SAP and Oracle/PeopleSoft had similar market shares in an innovative market with differentiated products would render a credible allegation of single-firm dominance difficult.

7.74  As Werden (2005)¹⁰⁹ mentions in relation to this merger in the US, significant unilateral effects can occur despite the presence of several rivals even if the merging products are not particularly close substitutes. Thus, the existence of a number of firms in the post-merger market (Microsoft, Lawson, etc.) offering products in the high-function HR and FMS markets, which arguably are not as efficient as those of the three majors, does not preclude the existence of unilateral adverse effects on competition from the merger in the form of non-coordinated effects in the post-merger oligopolistic market.

7.75  The concentration would result in the disappearance of PeopleSoft as an independent competitor, and would reduce the number of major firms that compete in the market (p. 250) for high-function HR and FMS software from three to two. This would decrease the intensity of competition in the market, leading to significant adverse effects on customers in terms of price, product variety, product quality, and innovation. Even under a wide definition of the market, when the market exhibits some degree of differentiation and it is difficult for undertakings to offer slightly different products from the ones they currently offer, the undertakings may devise strategies to apply localized price increases, leading thus to an adverse impact on competition in the market.

7.76  Competition in the market for high-function FMS and HR was characterized by vendors’ attempts to offer a software solution that best met the specific demands of specific large customers, while at the same time decreased the total cost of ownership.¹¹⁰ The products of software vendors exhibited a degree of product differentiation, characterized by high degree of innovation. In addition, customers of high-function HR and FMS solutions often had specific needs requiring customization of the basic product offerings and were highly sophisticated buyers.

7.77  These three factors (product differentiation, the ability to provide customisation, and superior knowledge of the customer) might have provided each EAS vendor with a certain margin in its price setting vis-à-vis the other players which did not have the same product offering available, did not have the same capability to customise their product to the needs of the customer involved, or did not have the same degree of insight into the needs of that customer.¹¹¹ Similarly, it provided for a certain margin in the extent to which each EAS vendor wanted to optimize the product it offered to a specific customer.

7.78  The post-merger market structure was thus likely to lead to significant adverse effects in terms of price and product offerings for consumers. The group with particularly limited choice after the merger consisted, inter alia, of those customers who did not find one of the two products (of Oracle and PeopleSoft) suitable for their needs; those who preferred to run their software on a non-Oracle database;¹¹² those who, in order to avoid being dependent on one supplier, might prefer to purchase their FMS and HR software from two different suppliers; and those customers who might prefer not to have software and database provided by the same supplier.¹¹³ Thus, as a result of the merger most customers could be faced with the prospect of significantly reduced competition between the two major remaining vendors.

(p. 251) 7.79  As already mentioned, the Commission initially based its concerns on the finding of non-coordinated effects in the post-merger oligopolistic market. The Commission was not able to request remedies for clearance or prohibit this merger due to the gap in the application of the dominance test on mergers leading to non-coordinated effects in oligopolistic markets.

7.80  This merger affected several markets, for sugarbeet seeds, maize seeds, sunflower seeds, oilseed rape seeds, spring barley seeds, pea seeds, and onion seeds in the EU. As regards the relevant geographic market for the seed markets, prices and supply conditions to final consumers differed in various member states, and commercial seed was, to a substantial degree, customized (for example, disease resistance, chemical treatments, size of seeds, etc.) to suit the conditions of each country or regional area. In addition, official trials were conducted in the respective member states to assure quality control.¹¹⁵ Thus, the Commission concluded that the relevant geographic markets should be national in scope. As regards the market for seed treatment, the Commission’s market investigation indicated that seed treatment constituted a separate national market although the Commission left the geographic definition open since the merger would not lead to any competition concerns in these markets. In the market for sugarbeet seeds the Commission concluded that the parties’ high combined market shares and significant overlaps, and the reduction in the number of major sugarbeet producers creating a strong market leader in a R&D-based industry, might significantly impede effective competition by the creation of a dominant position of the merged entity in the market for sugar beet seeds in Finland, the Netherlands, Portugal, Spain, Austria, Ireland, and Italy, and by the creation of non-coordinated effects in an oligopolistic market for sugarbeet seeds in Belgium and France.¹¹⁶

7.81  The Commission’s analysis indicated that the merger might significantly impede effective competition in the market for maize seeds in Denmark, the Netherlands, and the United Kingdom as well as in the market for sunflower seeds in Hungary and Spain. As far as the markets for spring barley seeds and pea seeds were concerned, the Commission concluded that the merger might significantly impede effective competition in France for spring barley seeds and in the market for vining-pea seeds in the UK. The parties offered commitments to alleviate the Commission’s concerns which consisted of the divestment of Advanta’s European business. The divested business consisted of all companies belonging to the Advanta (p. 252) group incorporated in Europe (including Russia and Turkey), with the exception of Advanta Technology Ltd. UK, a company that owned or had the right to certain intellectual property rights related primarily to Advanta’s North American maize and soybean operations.¹¹⁷

7.82  The Commission argued that the merger might lead to the creation of non-coordinated effects in an oligopolistic market for sugarbeet seeds in Belgium and France.¹¹⁸ The merger would lead to high combined market shares in Belgium (50–60 per cent), and France (40–50 per cent). The main competitor was KWS with 40–50 per cent in Belgium, and 30–40 per cent in France. In Belgium and France the combined parties would have been the market leader but the remaining market shares would almost entirely have been held by KWS. The merger might not have been blocked by the Commission had it been assessed under the dominance test. It should be noted at this point that although the merged entity would have been the market leader in Belgium and France, the remaining market shares were attributed only to one competitor (KWS). Thus, in the post-merger market there were two main firms with roughly equal market shares.¹¹⁹

7.83  As the Commission mentioned, the seeds industry was a research-based industry (R&D plays a crucial role in the seed industry) and competition could be described more as ‘competition for the market’ than ‘competition in the market’. The aim of companies in this market was to compete in order to develop a variety that would be the sole choice of the farmers.¹²⁰ Furthermore, the market shares showed relative stability over time, countervailing buyer power was deemed insufficient to pose a significant constraint to the merged entity, and entry barriers were high due to the costly and time-consuming process of entry in such an innovative market. These factors, which would contribute to a significant impediment to competition, would have influenced the Commission’s assessment of the merger under the dominance test as well.

7.84  In previous merger cases involving similar products and similar market conditions, the Commission alleged that the merger would have led to single-firm dominance with the largest competitor having a much wider difference in the market share compared to the merged entity.¹²¹ Contrary to the Commission’s argument (p. 253) regarding the strong competitors of the merged entity the Commission in Syngenta CP/Advanta alleged that the merger would lead to adverse competition effects by the creation of non-coordinated effects in an oligopolistic market, even though the merged entity and its largest competitor had roughly equal market shares. The Commission, based on the same argument (that is, strong competitors), reached a different conclusion as regards the effects of the merger under the dominance and the SIEC test.

7.85  The Johnson &Johnson/Guidant merger involved four main areas within the cardiovascular medical products business: i) interventional cardiology devices; ii) endovascular devices; iii) cardiac surgery devices; and iv) cardiac rhythm management devices. All markets were deemed to be national in scope.

7.86  As far as steerable guidewires are concerned, the merger would result in a quasi-monopoly situation in some member states. Guidewires in general are only moderately differentiated products. The merger was likely to result in a significant impediment to effective competition in the common market and the EEA for steerable guidewires as a result of the strengthening of Guidant’s dominant position.¹²³ In the market for balloon-expandable (BX) endovascular stents, Johnson and Johnson and Guidant were two of the strongest players, and there were high barriers to entry and insufficient countervailing buyer power. The merger would combine the leader and the number two in the BX stent markets. The merger would therefore significantly impede effective competition in the markets for BX stents in a number of countries, in particular as a result of the creation of a dominant position.¹²⁴ In the carotid endovascular stents of the endovascular stents, the three main players were Johnson and Johnson (J&J), Guidant, and Boston Scientific, together accounting for between 83 and 96 per cent of the market. The concentration would (p. 254) either reinforce the leadership of J&J or Guidant (in Austria, Finland, Netherlands, Portugal, and Spain), or combine the second and third player to create a new market leader (in Belgium, Germany, and Italy).

7.87  As far as the non-carotid endovascular stents of the endovascular stents were concerned, both J&J and Guidant market non-carotid self expandable (SX) stents in the EEA. The Commission’s market investigation indicated that J&J and Guidant products were considered to be the closest substitutes by the majority of respondents who procured non-carotid SX products. J&J and Guidant non-carotid SX stents were high-quality products, and close substitutes due to their superior performance compared to competing stents. The Commission concluded that given the characteristics of the markets of non-carotid SX stents in Austria, Belgium, Germany, and Netherlands in terms of concentration, barriers to entry, customer loyalty, closeness of substitution, and as a result of the elimination of a major competitive constraint, the concentration would give rise to non-coordinated effects in these national markets and therefore impede effective competition.

7.88  In conclusion, in the market for endovascular stents the concentration would either consolidate an existing leadership position of one of the merging parties or create a new market leader. The relevant product markets were characterised by differentiated products with J&J’s products being closer substitutes to Guidant’s products. The Commission further stated that there were considerable barriers to entry in the form of intellectual property (IP) rights, know-how, access to customers, and reputation of the firms, as well as large sunk costs. The parties offered commitments to alleviate the Commission’s concerns.¹²⁵ The Commission considered that the commitments were suitable for remedying the significant impediment to effective competition and cleared the merger.¹²⁶

7.89  In some of the markets analysed, the Commission argued that the merger would significantly impede effective competition; however, it might not have concluded that the merger would lead to a creation or strengthening of dominant position if the merger had been assessed under the dominance test of the ECMR.

(p. 255) 7.90  Regarding the market for non-carotid endovascular stents of endovascular stents, the Commission argued that the merger would reduce the number of competitors from four to three in the non-carotid stents market. At EEA level, the combined entity had a market share of 30–40 per cent (J&J: 20–30 per cent, Guidant: 10–20 per cent). J&J’s market share had been relatively stable for the past three years. Conversely, Guidant entered the market in 2000 and since then its market position constantly grew to reach 10–20 per cent. After the transaction the HHI would be 2,691, with an increment of 600. In Belgium and the Netherlands the combined market share would be 45–55 per cent; in Germany it would be 40–50 per cent. In these markets, J&J was market leader, while Guidant was the third player in Belgium (after Bard) and the fourth in Germany and the Netherlands (after Boston Scientific and Bard). Together, J&J, Guidant, Boston Scientific, and Bard accounted for 85–95 per cent in Belgium, 80–90 per cent in Germany, and 80–90 per cent in the Netherlands,¹²⁷ while the concentration ratio of the three largest firms was 70–80 per cent in Belgium, Germany, and the Netherlands.

7.91  The features of this market included high degree of product differentiation, significant barriers to entry, substantial countervailing buyer power and low degree of innovation, and the existence of IP rights.¹²⁸ It is unlikely that the Commission would have been able to argue adverse impact on competition on the basis of single-firm dominance had it assessed this merger under the dominance test. With such low market shares as well as countervailing buyer power and the presence of at least one more significant competitor in the post-merger market, the Commission would not, in my opinion, be able to allege that the merger would lead to single-firm dominance.

7.92  As regards coordinated effects between the incumbents in the post-merger markets, the significant degree of differentiation of the products, the substantial countervailing buyer power, the likely asymmetric cost structures due to the fact that the industry was characterized by innovation and R&D investments, as well as the existence of IP rights, render the sustainability of a collusive equilibrium unlikely.

7.93  The merger between Johnson &Johnson and Guidant would lead to a significant impediment to effective competition and was cleared after the submission of commitments. If the Commission had assessed the merger under the dominance test, it might not have been able to convincingly argue that the merger would have induced an adverse impact on competition in the markets analysed, mainly due to the differentiation of the products, asymmetric cost structure, the countervailing buyer power, and the low market shares of the merged entity. In these markets the (p. 256) merger would give rise to neither single-firm dominance nor collective dominance. However, the incumbents in the post-merger market would be able to unilaterally increase their prices and thus the merger would have non-coordinated effects in these oligopolistic markets.

7.94  This merger was assessed under the SIEC test. T-Mobile and Tele.ring operate mobile networks in Austria and were also active on related end-customer and wholesale markets. They also both provided fixed network services, but the Commission argued that the merger had no effect on these markets.

7.95  The Commission concluded that a single market existed for the provision of mobile telephony services to end customers, in so far as they could be provided on both a 2G and a 3G basis.¹³⁰ The issue whether there was a separate market for specific applications available only on the basis of 3G technology was left open since, inter alia, multimedia services had recently become available on the market. The geographic scope of the market was defined as national. As regards the wholesale market for call termination,¹³¹ as established in previous Commission decisions,¹³² the Commission argued that each individual network constituted a separate market for termination and national in scope.¹³³ Finally, concerning the wholesale market for international roaming,¹³⁴ demand for wholesale international roaming services came from foreign mobile operators who wished to provide their own customers with mobile services outside their own network and, downstream, from subscribers wishing to use their mobile telephones outside their own countries.

7.96  Turning to the competitive assessment, there were four main companies on the Austrian market operating mobile telephone networks based on GSM technology. They were Mobilkom (a subsidiary of Telekom Austria), T-Mobile, ONE, and Tele.ring. The market share of the merged entity was 30–40 per cent, while Mobilkom (p. 257) had 35–45 per cent, ONE had 15–25 per cent, and H3G had under 5 per cent.¹³⁵ These market shares were calculated on the basis of turnover, but did not change if the market shares were calculated according to the number of customers.¹³⁶

7.97  The Commission concluded that the elimination of Tele.ring as an independent network operator, the emergence of a market structure with two large network operators of similar size (Mobilkom and T-Mobile), a far smaller operator (ONE), and a very small operator (H3G),¹³⁷ would give rise to non-coordinated effects, even though T-Mobile would not have the largest market share after the merger. The Commission analysed nine factors in order to assess the adverse impact of the merger on competition. It analysed the market shares, the HHI, customer switching, price development, incentive structures, importance of national network, network capacity, the role of other competitors, as well as the future development of Tele.ring.¹³⁸

7.98  Regarding customer switching the market-share data in itself suggested that a large proportion of customers who had left T-Mobile and Mobilkom had become customers of Tele.ring. The data collected by the Austrian regulator on the basis of number portability further supported this interpretation. In 2005 more than half of all customers who switched provider and made use of number portability went to Tele.ring, and between 57 per cent and 61 per cent of those who left T-Mobile and Mobilkom with their telephone numbers switched to Tele.ring. In second place behind Tele.ring in 2005 was H3G, which picked up around some 20 per cent of all customers switching provider and using number portability.¹³⁹

7.99  The Commission’s price analysis illustrated that, overall, prices constantly fell in the reference period and that Tele.ring had offered its services since the third quarter of (p. 258) 2002 at significantly lower prices per minute than the other three network operators and that since the first quarter of 2002 at lower prices per minute than the market average. Tele.ring’s prices were well below the per-minute prices charged by the three leading operators. The Commission concluded that during the period under investigation (from 2002 to 2005) Tele.ring was the most active player in the market, and that it exerted considerable competitive pressure on T-Mobile and Mobilkom in particular and played a crucial role in restricting their freedom on pricing. The price analysis therefore suggested that Tele.ring’s role in the market was that of a maverick.

7.100  The Commission concluded that Tele.ring’s incentive to charge very competitive prices was a consequence of the number of its existing customers. T-Mobile had not pursued such a strategy and the combination of T-Mobile and Tele.ring would have even less incentive to do so in future. Tele.ring’s was regarded by customers as particularly inexpensive, but was not highly rated on other factors such as quality, innovation, or service.

7.101  After completion of the proposed merger, not only would the Tele.ring network be eliminated, but, presumably, the T-Mobile network would be used to full capacity to a far greater extent than was the case at that point in time. The proposed merger would therefore lead to a situation where instead of there being three operators the considerable reduction in spare capacity would also reduce the incentives for network operators to attract new customers by offering low prices in order to use up significant spare capacity. Thus, the merger would lead to a significant overall reduction in capacity in the market. According to the Commission, this reduction in available capacity would suggest that the merger would have a considerable impact on competition.¹⁴⁰

7.102  The Commission found no signs that a new network operator might be intending to enter the Austrian market. It concluded that it was unlikely that H3G or ONE/YESSS! would occupy a place in the market comparable with Tele.ring once the transaction was completed, or that they would have been able to discipline the competitive behaviour of T-Mobile and Mobilkom in particular. Similarly, service providers would also not be able to assume such a role. The Commission concluded that Tele.ring would continue to operate in future as a price-aggressive service provider on the Austrian mobile telephone market. The Commission concluded that, with the elimination of the maverick in the market and the simultaneous creation of a market structure with two leading, symmetrical network operators, it was likely that the merger would produce non-coordinated effects and significantly impede (p. 259) effective competition in a substantial part of the common market. After commitments submitted by the parties, the Commission cleared the merger.

7.103  This is a case that clearly indicates the existence of the gap in the dominance test of the ECMR. The Commission seems to analyse most of the factors that are essential in order for a merger to lead to non-coordinated effects in oligopolistic markets. The merged entity would have the second place in the post-merger market with 30–40 per cent, while Mobilkom would have 35–45 per cent. In addition, HHIs indicated a significant degree of concentration in the post-merger market. In the post-merger market there would be limited customer switching between the merged entity and Mobilkom, since once Tele.ring disappeared from the market, H3G would be the major destination of customers who would like to switch as is indicated by eliciting the 20 per cent of all customers switching provider and using number portability.

7.104  Limited switching between the merged entity and Mobilkom, the two largest firms in the post-merger market, indicated that both firms were likely to increase prices without having any significant risk of customers switching to the other. Although customers of both firms could have switched to the other competitors, as the Commission argued, it was unlikely that H3G or ONE/YESSS! would occupy a place in the market comparable with that of Tele.ring once the transaction was completed, or that they would be able to discipline the competitive behaviour of the merged entity and Mobilkom. Thus, both these firms could unilaterally increase prices in the post-merger market.

7.105  This case shows that the finding of non-coordinated effects is not limited to a situation where the merging parties are the closest competitors to each other. In addition, the merged entity had the second largest market share in the market, with Mobilkom being the largest firm. It is inconceivable that the Commission could allege that the merger would lead to the creation or strengthening of a dominant position, had the merger been assessed under the dominance test.¹⁴¹

7.106  As the above analysis illustrates, in most of the above cases, the post-merger market structure, in some markets, consisted of a small number of incumbents, with similar market shares, producing differentiated products. In these markets the criteria for collective dominance were not satisfied (for example, transparency, retaliation, monitoring, countervailing buyer power). In addition, the post-merger market was (p. 260) not conducive to the creation of unilateral effects either. However, the reduction in the competitive constraints in the market might have led to an adverse impact on competition, since the incumbents in the post-merger market would have the incentive to unilaterally increase prices, reduce quality, and adopt conduct having an adverse impact on competition. The Commission was unable to address the competition concerns due to the inapplicability of the dominance test of Regulation 4064/89 to mergers leading to non-coordinated effects in oligopolistic markets, which are addressed under the SIEC test of the EUMR.

7.107  Even though the legal substantive test has been changed from the ‘dominance test’ to the SIEC in the EUMR,¹⁴² and thus would appear to rectify the ‘gap’ in the European Community merger regime, the occurrences of such ‘gap’ cases may not cease under national laws that still adhere to the traditional dominance test. Such regimes are still likely to experience cases where they will be facing mergers, which will have the features of a non-collusive oligopoly but will be unable to apply the current dominance test. They may thus resort to other methods of trying to deal with mergers having an adverse impact on competition.