Previous business/academic article Next business/academic article
Academic Articles Awards > Mergers

Abandon All Hope, ye Who Enter Here? Efficiencies in European Merger Control: A Few Lessons from Recent Decisional Practice

François-Charles Laprévote, Concurrences Journal N° 2, May 2014

See François-Charles Laprévote's resume

This article considers four recent European Commission decisions on key merger cases in which efficiencies played a significant role in the parties’ arguments—and were either partly accepted (Deutsche Börse/NYSE Euronext and UPS/TNT Express) or rejected (Western Digital/Viviti and Hutchison 3G Austria / Orange Austria) by the Commission. By requiring a high standard of proof as regards efficiencies and rejecting the position that direct efficiencies would necessarily benefit consumers, the Commission has shown some reluctance to accept efficiencies as a counter-weight to the potentially anti-competitive effects of a merger. However, the recent decisions have also confirmed important principles and methodologies that may open the door for efficiencies to tip the balance in a complex merger case. This article advocates a number of improvements in the assessment of efficiencies, including a definitive repudiation of the “efficiency offense,” a better accounting of direct efficiencies (and where applicable, of inefficencies) and a more symmetrical standard of proof for the benefits of merger efficiencies and the predicted anti-competitive effects of a merger.

1. Efficiencies are often, if not always, at the core of a merger’s economic rationale. Companies contemplating a merger or an acquisition routinely invoke cost synergies, distribution savings, complementary products or services, or technological improvements arising from the consolidation of R&D efforts—all of which may be considered as efficiencies—as compelling reasons for the transaction. Although most of these proclaimed benefits will typically accrue to the combined entity and its shareholders, they may also reach consumers through cheaper, better, or more innovative products and services.

2. Against this background, one might expect that efficiencies should play a crucial role in the assessment by antitrust authorities of the desirability of a merger for the consumers and for the economy as a whole. This would resonate with economic theory and with the words of Judge Posner, according to whom “the only goal of antitrust should be to promote efficiency in the economic sense.” [1] On both sides of the Atlantic—and particularly in Europe—antitrust regimes and authorities have typically been reluctant to accept efficiencies as a way to counter-balance the potentially anti-competitive effects of a merger. As of today, while the European Commission has recognized efficiencies in a number of cases as an additional, subsidiary reason for authorizing a merger, there is to our knowledge not a single Commission decision where efficiencies have on their own been considered sufficient to remedy potential competitive concerns and turn what would otherwise have been a prohibition decision into an authorization. [2] The paradox is that if the number of prohibition decisions is used as a benchmark, this reluctance does not appear to have resulted in a more prohibitive merger control system over the past decade. [3]

3. This article will try to explore some of the reasons for this paradox in the light of four recent decisions by the Commission on major merger cases where efficiencies played a significant part in the parties’ arguments—and were either rejected or partly accepted by the Commission. Part I of this article provides a reminder on the legal and economic assessment of efficiencies in merger control with a particular focus on the EU. Part II briefly summarizes the four most recent decisions by the Commission on merger cases involving efficiencies (Western Digital/Viviti, Deutsche Börse/NYSE Euronext, Hutchison 3G Austria/Orange Austria and UPS/TNT). Part III explores selected issues raised by these decisions in respect of the verifiability of efficiencies, their passing-on to consumers and their balancing against other anti-competitive effects of the merger. Part IV concludes and provides some suggestions for the treatment of efficiencies in European merger control policy.

I. Why efficiencies matter and how they are taken into account in European competition policy

1. Why efficiencies matter: A reminder on the economics of efficiencies

4. The economic case for taking efficiencies into account in merger control is well-documented. The generally accepted framework of analysis is based on Williamson’s seminal article (1968), [4] which identifies the respective role of efficiencies and consumer harm generated by market power in the assessment of a merger’s overall impact on the economy and on consumers. Assume on Figure 1 that pre-merger price and quantity is Ppre and Qpre and that the firm (pre- and post-merger) faces a downward sloping demand curve D(p).

5. If the merger leads to a moderate decrease in marginal costs (MCpost1), the new company will likely set a higher price than before the merger (Ppost1), thereby creating allocative inefficiency in the form of dead weight loss (DWLpost1), without the productive efficiencies (Epost1) being passed on to consumers. If, however, the merger creates a larger decrease in marginal costs (MCpost2), even a monopolist, setting its output where the marginal revenue curve meets the marginal cost curve, may end up setting a lower price than before the merger (Ppost2).

Figure 1

Figures are available in the PDF version of the article

6. Even in the first scenario, where the efficiencies are not sufficient to prevent a price increase, the merger may increase total welfare if the dead-weight loss generated by the price increase is lower than the additional profits realized by the combined entity through the cost savings. But if antitrust regimes are exclusively focussed on beenfits to consumers and are not ready to allow such redistributive effects between consumers and producers—and a fortiori if efficiencies are not sufficient to prevent a decrease in total welfare—the transaction should not be allowed without remedies cancelling the overall anti-competitive effect of the transaction.

7. This standard trade-off model emphasizes the potentially crucial role of efficiencies in certain merger cases:

– First, in some cases efficiencies may be sufficient to entirely remedy on their own any anti-competitive effects resulting from the transaction and “flip” a case in favor of an authorization decision;

– Second, in other cases efficiencies may not be sufficient on their own, but may partly address anti-competitive effects and should therefore reduce the scope of remedies required to clear the transaction.

8. The limitations of the model are equally clear:

– First, like any predictive model, this one rests on assumptions that may not turn out to be correct and can therefore result in so-called “Type 1” mistakes. [5] Although this observation is also valid for the predictions of anti-competitive effects, [6] the issue may be compounded by the fact that information regarding efficiencies generally rests with the merging parties, which will have a vested interest in presenting only optimistic forecasts to antitrust authorities.

– A second potential shortcoming relates to timing: while efficiency gains may take a long time to materialize, the anti-competitive effects of an increase in market power would arguably be felt much more quickly. This may justify discounting in full or in part efficiencies that do not materialize shortly after the merger.

– Finally, efficiencies produced by a merger are usually limited to the combining firms. By contrast, potential anti-competitive effects may materialize across the entire industry, for instance in case of coordinated effects. This could also justify introducing a “discount rate” on efficiencies invoked by parties when weighing them against possible price increases.

2. Efficiencies in European merger control policy: A checkered history

2.1. The Commission’s decisional practice: Progress in words, status quo in deeds?

9. European merger control policy has initially handled efficiency arguments with great reluctance. In the 1991 Aérospatiale-Alenia/Havilland case—one of the first cases where efficiency considerations were brought up to mitigate the possible anti-competitive effects of a transaction—the Commission rejected the parties’ arguments on the ground that the potential cost savings arising from the concentration, as quantified by the parties, “would have a negligible impact on the overall operation of ATR/de Havilland, amounting to around 0.5% of the combined turnover,” and that the other cost savings identified by the parties were not merger-specific. [7] The Commission maintained this reluctance throughout the ’90s with a series of similar rejections. [8]

10. In several cases, the Commission even characterized efficiencies as offences or aggravating circumstances rather than defenses, by stating that they could reinforce the dominant position of the merged entity and make it more difficult for consumers to bypass this entity on the market. [9] This position—which is very debatable from an economic standpoint [10] —also resonated with U.S. case law of the ’60s (in particular Brown Shoes and Procter & Gamble). [11]

11. Since the 2000s, however, the Commission’s public statements on efficiencies changed markedly. The notion of “efficiency offence” was publicly repudiated. [12] The 2004 Merger Regulation expressly provided that the Commission should take into account “the development of technical and economic progress provided that it is to consumers’ advantage and does not form an obstacle to competition” [13] and envisaged that efficiencies might counter-balance anti-competitive effects arising from a concentration. [14] This evolution was confirmed shortly thereafter in the Commission’s Horizontal Merger Guidelines, which outline three cumulative conditions under which efficiencies may be invoked: [15]

(i) Consumer benefit: “Efficiencies should be substantial and timely, and should, in principle, benefit consumers in those relevant markets where it is otherwise likely that competition concerns would occur;” the Commission generally considers that only efficiencies generating a decrease in variable costs (as opposed to fixed costs) may result in price decreases and benefit consumers; [16]

(ii) Merger specificity: “Efficiencies are relevant to the competitive assessment when they are a direct consequence of the notified merger and cannot be achieved to a similar extent by less anticompetitive alternatives;” [17]

(iii) Verifiability: “Efficiencies have to be verifiable such that the Commission can be reasonably certain that the efficiencies are likely to materialise, and be substantial enough to counteract a merger’s potential harm to consumers.” [18]

12. The Commission has applied these conditions in a restrictive way in its decisional practice. Although it did recognize efficiencies in certain cases (in particular in high-tech industries), those were accepted only as secondary arguments in favor of a concentration which, as the Commission had already determined, raised few important competition concerns. [19]

2.2. The Courts’ case law: Some useful clarifications

13. In its 2010 Ryanair judgment (another case where efficiencies were invoked by the parties and rejected by the Commission), the General Court shed more light on the way efficiencies should be treated under the 2004 Merger Regulation regime.

14. In particular, on the merger specificity condition the General Court ruled that the Commission should not take into account “alternatives that [were not] reasonably practical in the business situation faced by the parties (…) having regard to established business practices in the industry concerned.” [20]

15. Regarding the verifiability of the alleged gains, the General Court rejected the Commission’s demands that the parties produce data that may be verified by independent third parties or documents pre-dating the concentration. The Court also recalled that it was sufficient to prove that the efficiencies are “likely” to materialize, and that “when the necessary data are not available to allow for a precise quantitative analysis, it must be possible to foresee a ‘clearly identifiable’ positive impact on consumers, not a marginal one.’” [21]

II. A summary of four recent efficiency cases

16. Over the past two years, four major cases examined by the Commission involved significant efficiency arguments. Interestingly, in two of these cases (Deutsche Börse/NYSE Euronext and UPS/TNT Express) the Commission adopted a prohibition decision despite acknowledging the existence of efficiencies, while in the other two cases (Western Digital/Viviti and Hutchison 3G Austria/Orange Austria) the Commission rejected efficiency arguments and nevertheless cleared the transaction subject to remedies.

17. Out of these four decisions, only two, Deutsche Börse/NYSE Euronext [22] and UPS/TNT Express, [23] were appealed before the General Court. [24]

1. Prohibition decisions despite efficiencies

1.1. Deutsche Börse/NYSE Euronext (“DB/NYX”)

18. The merger contemplated between Deutsche Börse and NYSE Euronext would have led to the creation of the world’s largest stock exchange combining the activities of both parties, including for cash listing, cash and derivative trading and the entirety of their vertically integrated services, including clearing. Nevertheless, the transaction resulted in one of the few prohibition decisions adopted recently by the European Commission, [25] essentially on the ground that it would have created a “near-monopoly” in European exchange-traded derivatives, with a market share of around 90%. [26]

19. This decision is particularly noteworthy given the parties’ emphasis on very substantial efficiency gains, the combined total of which (at least $3 billion) [27] surpassed the merged entity’s combined total revenue on the market where potential competition issues arose. Such efficiencies included:

– A reduction of users’ technology costs (i.e. platform access and digitization of data); [28]

– Approximately €3.1 billion of savings in collateral requirements by “pool[ing] their clearing operations into a single [central counterparty]:” [29] Collateral deposits are required by clearing houses (or central counterparties, “CCPs”) from the parties to a transaction (the stock exchanges’ customers) as a counter-guarantee for the execution of the transaction; they play a particularly important role in the trading of derivatives. The required amount of collateral is calculated with an algorithm, which allows offsetting positions held by the same customer on negatively correlated derivatives in order to reduce the collateral requirement (“cross-margining”). Cross-margining opportunities (and therefore collateral savings) obviously grow with the number of transactions cleared by the CCP;

– An increase in liquidity on the spot and derivatives markets stemming from “the increase of the number of participants [and] improved distribution;” the parties argued that “increased product combinations and innovations w[ould] multiply trading opportunities,” and that this increase of liquidity would reduce user access costs and lower the implicit transaction costs resulting from the price volatility and bid-ask spread. The parties based their forecast on Euronext’s previous merger experience to quantify the expected liquidity gains; [30]

– Cost reductions derived from the synergies achieved by the proposed concentration, which were estimated at €400 million; [31] and

– Wider positive effects of greater liquidity on the economy. [32]

20. The parties supported their positions with five economic studies. A specificity of this case was that most of the efficiencies invoked by the parties would have accrued directly to consumers by a reduction of their general transaction costs, rather than through price decreases from the combined entity.

21. In its final prohibition decision, the Commission nevertheless rejected most of the parties’ arguments in what is probably one of the most detailed analysis realized to date on efficiencies in a merger transaction in the EU: [33]

– Regarding reduced users costs, the Commission questioned the verifiability of the efficiencies cited by the parties and did not consider their estimates to be sufficiently reliable to meet the required burden of proof; [34]

– Regarding the reduction of collateral generated by additional cross-margining opportunities, the Commission found that “it is not the collateral savings but the opportunity cost of holding cash or securities posted as collateral which is the relevant measure of actual cost savings from lower collateral requirements.” [35] As a result, “the savings would most likely be in the range of approximately Euro [40-60] million to Euro [100-120] million.” [36] The Commission also considered that only a part of these savings would have been merger-specific and passed on to customers; [37]

– The Commission considered that the forecasted liquidity increase was not sufficiently verifiable. The parties supported their conclusions with econometric studies on the integration of the French, Belgian, Dutch and Portuguese stock markets with Euronext, as well as the integration of Euronext’s derivatives platform from 2000-2004. The Commission, however, only found the studies to be “informative” and not “probative,” as the market conditions, in its view, had changed since then; [38]

– The Commission considered that the wider economic benefits claimed by the parties were unverifiable; [39]

– Finally, the Commission considered that the parties’ expected cost reductions essentially related to fixed costs and were therefore unlikely to be passed on to customers through price decreases by the combined entity. [40]

22. Ultimately, the Commission rejected all the efficiencies raised by the parties, except for those resulting from collateral savings. The Commission stated that these residual gains were insufficient to compensate the harm caused to clients by the contemplated merger, as they could easily be recovered by the combined entity through a modest price increase. [41]

1.2. UPS/TNT

23. On January 30, 2013, the European Commission prohibited the acquisition of TNT Express by UPS on the ground that this transaction would have resulted in a significant impediment to effective competition in the 15 Member States where the operation would have reduced the number of competitors from three to two express delivery operators. [42]

24. UPS and TNT also argued that the combination of their respective networks would have brought about €400-550 million in cost savings, notably in (i) ground transportation costs; (ii) air network; and (iii) management and administrative overheads. These arguments were only very partly accepted by the Commission:

– The Commission reported that the parties were unable to adduce evidence that overheads costs were allocated to individual packages and therefore had an impact on the firm’s customer pricing policy. Instead, it viewed these efficiencies as fixed costs reduction. The Commission therefore considered that such cost savings were unlikely to be passed on to consumers. Potential ground transport efficiencies ascribed to express delivery were also deemed “not clear enough.” Although the Commission apparently accepted that they may benefit customers, it also considered that such efficiencies would mainly have positive effects “on other services” than express parcel delivery—hence not alleviating the Commission’s concerns on that market.

– However, the Commission agreed with the parties’ claim that air transport efficiencies would result from the merger and would “normally benefit consumers.” The Commission expected such efficiencies to materialize “sufficiently quickly” after the implementation of the merger, and it also deemed unlikely that such efficiencies would have been achievable by other means than the merger.

– Nevertheless, according to the Commission’s calculations (which partly relied on calculations provided by the parties themselves), the mere air transport efficiencies would not outweigh the price increase that would result from the combined entity’s reinforced market power [43] .

2. Two clearance decisions with no efficiencies

2.1. Western Digital/Viviti Technologies

25. In Western Digital/Viviti Technologies, the Commission authorized the acquisition by Western Digital of Viviti Technologies, a subsidiary of Hitachi in the hard disk drive sector, after a phase two investigation and subject to remedies. [44] The Commission had identified potential competition concerns on the 3.5-inch hard disk drives for “Business Critical” computers and hard disk drives for consumer electronic products, to the extent that the merged entity’s sole competitors would have been Seagate/Samsung.

26. Western Digital had claimed, among others, that the transaction would create important efficiencies, mostly to the benefit of consumers. In this respect, the company had submitted a number of detailed economic studies that were largely based upon the experience acquired in previous acquisitions. This included in particular a regressive study, according to which, based on prior transactions, the merged entity would be able to reduce its fixed and variable costs by around 10% across all relevant market segments, which would lead to a price reduction of 8.3% for consumers.

27. The Commission rejected this argument in its entirety:

– With respect to verifiability, the Commission considered that the efficiencies invoked were too general, and even speculative, in nature. The fact that the party expected the swift fruition of these efficiencies (within a year) was, according to the Commission, insufficient to support its pro-competitive claims;

– Regarding merger specificity of R&D efficiencies, the Commission considered that the parties had failed to sufficiently prove that R&D agreements between them would not be able to generate similar efficiencies; the Commission also quickly dismissed the merger specificity of the other claimed efficiencies (increasing yield, improving quality and reducing inventories) by considering that they would also materialize absent the merger.

– Finally, the Commission admitted that in dynamic markets with rapid innovation, the reduction of fixed costs “could benefit consumers” through lower prices, but decided that the parties did not sufficiently substantiate this point. The Commission also did not accept the parties’ arguments regarding the expected reduction of the merged entity’s variable costs and its consequences on final prices. According to the Commission, the competitive pressure exerted by the post-merger market structure (with essentially a duopoly) would be insufficient to affect the merged entity’s pricing. Last but not least, the Commission rejected the regressive analysis submitted by the parties on the grounds that (i) the data had been collected in a more competitive context than that which would exist in a post-merger scenario, and (ii) the proposed regression method allowed a gross overestimation of consumer benefits which would be accrued from the concentration.

2.2. Hutchison 3G Austria/Orange Austria

28. In Hutchison 3G Austria/Orange Austria, [45] the Commission authorized the acquisition by Hutchison 3G Austria of Orange Austria after a phase 2 investigation and subject to remedies. [46] The parties invoked four efficiencies: network capacity increase, which would benefit consumers by reducing bottlenecks, improving the speed of the network and allowing for more aggressive pricing; faster 4G Long Term Evolution rollout, enabling H3G to modernize Orange’s 2G network; improved network coverage; and reduction of alleged scale disadvantages, generating cost savings.

29. The Commission determined that the three-pronged test set in the Horizontal Merger Guidelines had not been met for any of the four efficiencies invoked by the parties. Regarding the verifiability of the efficiencies, the Commission considered that, in view of contradicting evidence as to H3G’s network’s top quality, the evidence put forward by the parties failed to establish that network-related efficiencies would be attained, and network congestion avoided, through a network capacity increase. [47] However, the Commission acknowledged that the Notifying Party had verified, by means of internal documents, the level of network-related savings to be generated by the reduction of alleged scale disadvantages. [48]

30. Regarding the merger specificity of the efficiencies, the Commission stated that the alleged efficiencies could be achieved by less anti-competitive alternatives, such as a network-sharing agreement. [49] Regarding the benefit to consumers, the Commission stated that cost savings to be generated by the reduction of alleged scale disadvantages may not lead to a price decrease. [50] As to the first three efficiencies invoked by the parties, even though they were “demand-side efficiencies,” the Commission refused to conclude that they would necessarily amount to a net benefit for consumers. [51]

III. Selected issues raised by the Commission’s recent decisional practice

31. The recent Commission decisions raise—or leave open—several questions on the application of the verifiability criterion (1.), the requirement to demonstrate a benefit to consumers (2.), the methodology for balancing efficiencies with the anti-competitive effects of a merger (3.), the possible difference of treatment of efficiencies under Article 101 (3) TFEU (4.), and the role of the Courts in the assessment of efficiencies (5.).

1. The verifiability of efficiencies: Are the rules of evidence over-restrictive?

32. In the four recent decisions cited above, the Commission seems to have applied demanding rules of evidence regarding the verifiability of the efficiencies put forward by the notifying parties. This position is all the more noteworthy that in several of these cases the parties had relied on detailed econometric studies to substantiate and quantify the magnitude of the efficiencies. [52]

1.1. Burden of proof: Is a strict interpretation of the Horizontal Guidelines always a good thing?

33. The Horizontal Merger Guidelines clearly bestow the burden of proof for efficiencies on the notifying parties. It is for the parties to “provide in due time all the relevant information necessary to demonstrate that the claimed efficiencies are merger-specific and likely to be realized.” [53] Notifying parties must also “show to what extent the efficiencies are likely to counteract any adverse effects” the proposed merger may have on competition. [54] This rule is not considered controversial as information regarding the expected efficiencies arising out of a concentration—particularly the cost synergies—is generally “solely in the possession of the merging parties.” [55]

34. On this point, the NYSE/DB case raised a novel issue: in addition to “classic” efficiencies resulting from cost synergies, the notifying parties invoked several “direct” efficiencies, the effects of which would not materialize in the combined entity’s costs (and later in prices charged to customers), but directly in their clients’ cost base. Such efficiencies included in particular the reduction of clients’ access and maintenance costs arising from the harmonization of the exchanges’ information systems and connections. In this specific case, however, and in contrast with classical, “indirect” efficiencies, the parties did not have a particularly privileged access to the information allowing to quantify the expected benefits and could therefore only produce a rough estimate of the savings generated by the deal. Nevertheless, the Commission followed a close reading of its Guidelines and rejected these estimates because of their alleged lack of accuracy; “given (…) the lack of reliability based on the significant deficiencies submitted by the Notifying Parties,” the Commission therefore concluded “that the claimed cost savings ha[d] not been substantiated to the required standard.” [56]

35. This strict application of the burden of proof principles laid out in the Guidelines seems questionable. It is not controversial that the burden of proof should be on the notifying parties to establish efficiencies that relate to their own cost base. But when a concentration is likely to create “direct” efficiencies for customers—which are also the most meaningful efficiencies, since they do not require to make any assumptions as to their “passing-on” to customers through price decreases by the combined entity—and when the parties are able to provide only a rough estimate of such efficiencies, would it not be possible for the Commission to use its own investigative powers and attempt to verify their existence (and quantification) during its market tests? This would only reflect the economy reality about the availability of information: while customers might understandably be reluctant to disclose details about their cost base to their suppliers, they might be more willing to do so in their contacts with the Commission. Such solution would be in line with the obligation set by the Merger Regulation for the Commission to “take into account (…) the development of technical and economic progress,” [57] and more generally with its “duty (…) to examine carefully and impartially all the relevant aspects in the individual case.” [58]

1.2. Standard of proof

36. The rejection by the Commission of detailed economic studies provided by the parties to quantify the expected efficiencies in the NYSE/DB and Western Digital/Viviti cases raises the issue of the standard of proof that should be applied.

37. In NYSE/DB, the parties produced econometric studies analyzing the 2000-2004 mergers that led to the creation of Euronext, which if extrapolated to the NYSE/DB transaction demonstrated that the merger would create significant efficiencies in the form of increased market liquidity. Based on counter-evidence produced by DG Competition’s Chief Economist, the Commission rejected these studies principally on the ground that they were based on past events and may have ignored other factors. The Commission considered that “given the multitude of changes which ha[d] taken place over the last ten years, it [was] impossible to impute any efficiency gain from a 2002 merger into a transaction that to be completed in 2012.” [59] The Commission also inferred from several academic studies and from a presentation prepared by a competitor that, in the notifying parties’ studies, the effect of the 2007 Markets in Financial Instruments Directive (“MiFID”) was “underestimated”, and so was the effect of new entries on the markets. [60] The Commission therefore concluded that the notifying parties’ analysis suffered “from a severe identification problem” because it allegedly could not identify the specific impact of the Euronext mergers in the noted rise of liquidity. [61]

38. The Commission appears to have shown a similar distrust in an econometric study produced by a notifying party in Western Digital. The study showed, based on past party behavior, that the merged entity’s variable cost savings would be transferred to clients via price reductions. Again, the Commission appears to have discarded the study on the grounds that it relied on a period where the market was more competitive than it would be post-transaction—in addition to having been persuaded by economic counter-evidence that questioned the causal link between variable costs and pricing.

39. These two examples illustrate the high standard of proof required in practice by the Commission to demonstrate efficiencies and raise several questions:

– First, the coherence of this stringent standard of proof applied in these cases with that proposed in the Horizontal Merger Guidelines and case law may be questioned. Although the wording of the Guidelines is somewhat convoluted, it does ultimately rely on a concept of probability as opposed to absolute certainty. [62] Furthermore, the quantification of expected gains is not necessary if they can be clearly identified. [63] Finally, the Guidelines clearly state that “historical examples of efficiencies and consumer benefit” are among the probative elements that may be used to support efficiencies. [64] In this context, is it possible for the Commission to discard a quantified study produced by the parties because it is based on historical examples or to reject it because it is based on general academic studies that do not deal with the specific case at hand? Even if the study’s calculations were incomplete or overestimated the alleged efficiencies, would that be sufficient to deny that such efficiencies are “clearly identifiable” or to reject them entirely rather than take them partly into account?

– On the other hand, if the Commission rejects an econometric study with counter-evidence and with its own economic study, this may raise issues relating to the parties’ rights of defense. This may be the case if the Commission’s own study (and its underlying data) are not communicated to the parties in a manner that allow them to provide meaningful comments during the proceedings. Furthermore, if the Commission’s decision is appealed, the Courts may be tempted to consider that such debates between the economic experts of the Commission and the parties fall within the Commission’s margin of appreciation on complex economic assessments. Deutsche Börse seems to have launched the debate on this front in its appeal against the NYSE/DB decision by alleging a violation of a party’s right to be heard and the introduction of post-hearing arguments with respect to efficiencies. [65]

2. Consumer benefits: From pass-through to recovery theory

2.1. The classical pass-through test

40. Under the Horizontal Merger Guidelines parties must demonstrate that efficiencies will materialize within a short period and will benefit consumers. In the (most frequent) case of cost synergies, parties have to show that they will not simply internalize the lower costs achieved by the merged entity and that those price reductions will be passed on to consumers through lower prices or improved products. [66]

41. This can for example be determined using the upward pricing pressure (UPP) test, which shows the net effect of two countervailing forces: on the one hand, the efficiencies created by the merger (downward pricing pressure); on the other hand, the elimination of competition between the merging firms (upward pricing pressure). If Firm A produces Product 1 and Firm B produces Product 2, the merger will likely create upward pricing pressure on Product 1 if:

D12 (P2pre – MC2pre) > E1MC1pre

42. Where: D12 is the diversion ratio from Product 1 to Product 2 before the merger, i.e. “the proportion of customers of A which would switch to B in the event of a price increase;” [67] P2pre the price of Product 2 before the merger; MC1pre and MC2pre the marginal costs of production of Product 1 and Product 2 before the merger; and E1 the efficiencies created by the merger, i.e. the percentage reduction in the marginal cost of Product 1.

43. The purpose of the test is to check whether, where a pre-merger price increase would have made A lose profits to B, a post-merger price increase will be profitable for the merged entity. Indeed, by retaining some of the switchers, the merged entity will “internalise part of the losses which a price increase would otherwise bring about.” [68] Such analysis is necessary to determine whether productive efficiencies will actually be “passed on” to consumers.

2.2. The issues raised by “direct” efficiencies

44. The DB/NYSE case raised a novel question in this respect: “direct” efficiencies for the merged entity’s clients made unnecessary any “pass-through” between the merged entity’s costs and the prices billed to its clients. [69] Was it not sufficient that such efficiencies directly reduced the clients’ cost base to meet the test set in the Horizontal Guidelines? The same question was raised in Hutchison: for demand-side efficiencies such as improved network quality and coverage, or faster LTE rollout, would a quantification of the benefit to consumers not be superfluous? [70]

45. In both cases, the Commission responded in the negative and rejected the allegation that direct efficiencies would necessarily benefit consumers. In its opinion, it was necessary to verify that the merged entity would not “claw back” such efficiencies through higher prices. Thus, when considering the benefit to consumers from the alleged network capacity increase and faster LTE rollout in Hutchison, the Commission held that “the fact that higher network quality would be experienced directly by subscribers does not mean that they would have a net benefit from the merger. In particular, if not constrained by enough competitive pressure, the merged entity could in principle increase prices so as to partially or wholly claw back any benefits at the customer level.” [71] This new “claw back” test raises several questions:

– First, the “claw back” theory, which was elaborated by the Commission for the first time in DB/NYSE, is not mentioned in the Horizontal Merger Guidelines, which only provide that efficiencies must benefit consumers. To some extent, the claw back theory even seems to contradict the wording of the Guidelines. For instance, in Hutchison the Commission considered that higher network quality or faster LTE technology experienced directly by subscribers would only constitute efficiencies if not followed by a price increase. However, the Horizontal Merger Guidelines state that consumers may benefit from “improved products or services,” just as they would from lower prices. [72] Why could an increase in quality, even followed by a price increase, not result in a net benefit for consumers?

– It might be argued that the theory on the “claw back” of “direct” efficiencies does not differ from the theory on the “pass-through” of “indirect” transfers: in both cases, it is necessary to determine the ways in which the merged entity could capture wealth created by the increased efficiency to the consumers’ detriment. Under this argument the means used for such capture (price increases or refraining from price decreases) is irrelevant. But is this reasoning economically valid? Is refraining from price decreases following a cost reduction by the merged entity necessarily equivalent to a price increase where the merger entity’s costs remain unchanged? “Clawing back” efficiencies might in practice be a more difficult exercise than maintaining prices unchanged. Such claw back indeed would require (i) perfect knowledge by the merged entity of each of its clients’ costs; (ii) perfect ability to price-discriminate among various clients; and (iii) no potential negative externalities or costs (for instance “menu costs”) resulting from price increases. These are all strong assumptions, which would have deserved to be justified in more detail by the Commission.

– Third, as regards the standard of proof, the “claw back” and “pass-through” theories are not equal: in the case of a “pass-through” the parties must establish a positive effect (a decrease in prices) while in the case of direct efficiencies they are required to prove a negative (the absence of price increases) [73] —a more difficult achievement in practice.

– Crucially, the “claw back” theory may also lead to double accounting of price increases to the extent it conflates the assessment of efficiencies with their weighing and balancing with the anti-competitive effects of the merger, which in principle only takes place once the efficiencies have been established. This risk is apparent in the DB/NYSE decision, where the “claw back” theory is very expressly linked to the Commission’s theory of harm: “the merger entity’s incentive to leave efficiency gains from such cost savings with customers depends on the existence of competitive pressure from the remaining firms in the market and from potential entry. If no such or only limited pressure exists, the merged entity can in principle increase any explicit fee so as to partially or wholly claw back any cost savings at customer level” (§1179, emphasis added). In other terms, according to this paragraph (i) the claw back price increases are a direct consequence of the transaction’s anti-competitive effects, and (ii) such anti-competitive effects are presumed to be potentially infinite since they can recover “any” cost savings. This is a strong assumption even for a merger-to-monopoly if one considers that the alleged liquidity efficiencies in this case were several times higher than the merged entity’s turnover and their claw back would therefore have required the entity to more than double (or possibly treble) its prices. In DB/NYSE, the Commission denied any confusion between the competitive assessment, which “assess[es] the merger’s unilateral effects considering the coming together of the two firms without any changes in costs or technology,” and the efficiencies, which “assesses efficiencies and how the price adjusts following cost savings.” [74] But this reasoning, which seems to assume two price increases (one taking place before the efficiencies materialize and the second one taking place afterwards) does not solve the initial question: where does the post-efficiency “price adjustment following cost savings” come from if not from an assumption on the competitive effects of the merger?

– These assumptions in practice shift the burden of proof regarding the anti-competitive effects of the merger, which are simply assumed by the Commission (and must be disproved by the parties) in the claw back theory. This is apparent in the Hutchison case, where the Commission simply assumes that “if not constrained by enough competitive pressure, the merged entity could in principle increase prices so as to partially or wholly claw back any benefits at the customer level” and states that the notifying party “has not brought forward any justification” of why it should not be so (§424). In sum, while the Commission bears the burden of establishing any anti-competitive effects of the merger, the Hutchison decision seems to imply that it can just presume such effects when it assesses efficiencies, and that it is for the parties to rebut these assumptions. This does not appear to provide a consistent analytical framework, whether from an economic or a legal perspective.

3. Balancing efficiencies: The return of the efficiency offence?

46. Having found that the efficiencies stated by the notifying parties met the Guidelines’ criteria, the Commission must, pursuant to these Guidelines, consider whether these efficiencies outweigh the competition problems identified elsewhere. The application of this principle by the Commission in DB/NYSE calls for two observations here:

– First, the Commission followed the parties’ suggestion to conduct a quantified balancing test by comparing the efficiencies involved (which were reduced to a few dozens of millions, while the parties initially claimed efficiencies worth billions of euros) with the combined entity’s expected revenues. In this case, the Commission considered that the admissible efficiency gains—which only reached a few percentage points of the merged entity’s combined derivatives revenue—would not offset the potential increase in prices on the market that was allowed by a merger to monopoly. [75] While the exact amount of this increase was not precisely calculated for the purpose of the efficiencies test, this comparison appears in principle correct (notwithstanding the fact that some efficiencies in this specific case were rejected on the ground of the “claw back” theory, as mentioned above). In this case, one might wonder what would have happened if the Commission had accepted the efficiencies associated with increased liquidity, which represented more than 100% of the combined entity’s derivatives revenue. Could the Commission have comfortably concluded that the entity, even with a quasi-monopoly position on the market, would have been able to double its prices because of the concentration and, as a result, cancel out all the efficiencies? Such an outcome seems unlikely. This implies that in this case and others to come, efficiency gains, if verifiable, could in theory invert a decision.

– On the other hand, one may wonder whether the Commission’s decision was entirely consistent and did not indirectly reintroduce the notion of an “efficiency offence.” In its assessment of the merged entity’s market power, the Commission seems to have placed some weight on a competitor’s claims, according to which considerable collateral savings and the expansion of the entity’s liquidity pool (which are in essence efficiencies benefitting the consumers) would have allowed the party to foreclose its competitors from the market. [76] This assertion is surprising. If the efficiencies were neither significant nor verifiable as the Commission claimed, how could they allow the merged entity to foreclose its competitors? This seems to re-introduce a form of efficiencies offence and result in an asymmetrical standard of proof for the benefits of merger efficiencies and the predicted anti-competitive effects of the merger. [77]

4. A comparison with the treatment of efficiencies under Article 101(3) TFEU

47. The European Commission must also consider efficiencies when applying Article 101(3) TFEU to agreements otherwise prohibited under Article 101(1) when such agreements “contribute[s] to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit.” The rules applying to these efficiencies have been clarified in the Commission’s Guidelines on the application of Article 81(3) (now 101[3]) of the Treaty. [78] They are, unsurprisingly, very similar to those applied for merger control:

– The burden of proving and substantiating the efficiencies claimed under Article 101(3) rests on the parties. The Commission in its assessment takes several factors into account, including the nature of the efficiency gains. [79] Regarding cost efficiencies, the Commissions requires that parties calculate or estimate their amount as accurately and reasonably as possible—with “verifiable data, describe in detail how the amount has been computed, and describe the methods by which the efficiencies have been or will be achieved.” [80]

– The restrictions contemplated in the agreement must be indispensable to attain the predicted benefits. [81] This condition resonates with the “merger-specificity” condition in merger cases.

– Under Article 101(3), a “fair share” of the claimed benefits must be passed-on to consumers. This is similar to the consumer benefit test in merger control. Like in merger cases, the Commission will deem a decrease in variable costs more likely to be passed on to consumers than a decrease in fixed costs. [82]

– Finally, the efficiencies and the pass-on to consumers generated by the restrictive agreement within a relevant market “outweigh” its anti-competitive effects, i.e. the allocative inefficiency generated by the agreement within that same market. [83] Any reduction in costs will be balanced against the parties’ incentive to raise price due to their increased market power. Like the Horizontal Merger Guidelines, the Article 101(3) guidelines provide that in case of a “substantial reduction” of competition, “extraordinarily large cost efficiencies are normally required for sufficient pass-on to occur.” [84]

48. While the rules applying to both efficiencies are similar, the Commission has already accepted in Article 101(3) that such efficiencies could outweigh possible anti-competitive effects. For instance, in the Eurocheques decision (which predated the ECMR) the Commission considered that a system establishing a pre-determined fee payable by the issuer bank to the foreign payee bank for any cheque payable to a foreign EU bank had an anticompetitive effect, but could be allowed because it contributed to “improving payment facilities within the common market” and benefitted customers. [85] More recently, the Court confirmed in the Irish Beef case that national courts could exempt even agreements with an anticompetitive object by reason of the efficiencies they may produce. [86] Similarly, a number of agreements between national banks to harmonize procedures and to set the remuneration for inter-bank services—a restriction by object—were exempted because of the improved standardization and rationalization brought about by the new procedures. Here the “fair share” received by consumers came directly from the improvement in the services, which only indirectly led to welfare gains.

49. Several recent payment cases [87] might suggest that the Commission is now much more reluctant to accept similar efficiency claims. This could however be a somewhat distorted view. First, while considering them insufficient to outweigh the likely anticompetitive effects, the Commission did accept to credit efficiencies under Article 101(3) in some recent cases. In Continental/United/Lufthansa/Air Canada (“Star Alliance”), the Commission even seemed to broaden the way it usually assesses the benefits of cooperation agreements, by considering efficiencies that did not arise directly from the market affected. [88] Second, following the “modernization” of EU competition rules in the early 2000s and the decentralization of antitrust enforcement, most of the cases—and therefore most of the efficiencies analysis—were transferred to national competition authorities. At the same time, the new Block Exemption Regulations, [89] the Guidelines on the application of Article 101(3) [90] and the Guidelines on horizontal cooperation agreements [91] provided for automatic exemptions of agreements between firms of limited market power. As a result, only the most noxious transnational agreements between firms with high market power are still assessed on a case-by-case basis by the Commission, which naturally reduces the chances of efficiencies being able to counterweigh possible anti-competitive effects. [92]

50. There may be reasonable explanations for the apparently greater acceptability of efficiencies in Article 101 cases. For instance, agreements by nature entail less structural anti-competitive effects than mergers. The measurement of efficiencies brought about by agreements may also be easier if such agreements have already been implemented, while the determination of efficiencies in merger cases is necessarily a predicative exercise.

5. The role of the Courts in the assessment of efficiencies

51. With the exception of the Ryanair case mentioned above, the EU Courts have not until now had an opportunity to explore the legal conditions for the assessment of efficiencies. Such control would in any event be limited: in Ryanair, the General Court applied the established case law of the Court of Justice recognizing the Commission’s “margin of discretion with regard to economic matters,” [93] although such margin “does not mean that the EU courts must refrain from reviewing the Commission’s interpretation of information of an economic nature.” [94]

52. In this respect, the appeal raised by Deutsche Börse in the DB/NYSE case may provide the General Court an opportunity to review a number of potential issues that may impact the Commission’s future practice with respect to efficiencies. [95]

53. The General Court will be able to review whether (in the words of Deutsche Börse’s second plea in law) “[i]n relation to its evaluation of both collateral savings and liquidity benefits, the Commission violated the parties’ right to be heard by relying on evidence and arguments introduced after the oral hearing on which the parties were not given opportunity to comment.” [96] Indeed, the EU Courts’ review includes “verifying whether the rules on procedure…have been complied with.” [97]

54. On the substance, the General Court will also have to review whether the Commission’s claw back theory is acceptable or whether, as claimed by Deutsche Börse, this theory and the Commission’s assessment of the merger specificity of efficiencies “were based on new theories and requirements that are not supported by the Commission’s Horizontal Merger Guidelines.” [98]

55. Finally, the General Court will also be able to review the economic evidence relied on by the Commission in its DB/NYSE decision. In particular, Deutsche Börse claims that the Commission’s “assessment of the parties’ efficiencies claims was vitiated by manifest errors and not supported by cogent and consistent evidence.” [99] One might wonder how the General Court will take account of the starkly differing views between the economists of the parties and those of the Commission (for instance, on liquidity efficiencies). Will the Court simply defer to the Commission’s margin of discretion? Or would this kind of case, where Courts may not be able to adjudicate between two experts on highly technical matters, call for a neutral expert’s report as allowed under the Court’s Rules of Procedure? [100] To this date, the Court appointed experts in only two antitrust cases, [101] both related to the proof of concerted practices in the presence of parallel behavior.

IV. Conclusion and recommendations

56. At first sight, the recent Commission decisional practice may not appear very encouraging for companies wishing to emphasize efficiencies in their merger defense before the Commission. If confirmed, the return of a form of efficiencies offence in some of these cases would also be a disturbing development, especially if this resulted in an asymmetrical standard of proof where the negative effects of efficiencies can simply be presumed while their positive effects must be established in painstaking detail. It has even been suggested that some of these cases may betray an overreliance by companies and their economic advisers on efficiencies while other potentially more fruitful defenses (such as counterweighing buyer power or a refinement of the excessively simplistic theoretical models of unilateral effects that invariably generate predictions of post-merger price increases) could have been envisaged in some individual cases. [102] This, however, might not be the time to abandon all hope on efficiencies defenses in EU merger cases. The Commission’s recent decisions have at least confirmed some important principles and methodologies, such as the “balancing” of efficiencies with the merged entity’s revenues on the problematic markets and with the potential price increases resulting from the transaction. Such principles (and the fact that efficiencies are sometimes recognized by the Commission) may open the door in the future for efficiencies to tip the balance in a complex merger case.

57. In this respect, the following recommendations could be made:

– First, the efficiency offence theory should, once again, be repudiated by the Commission, in order to avoid deterring notifying parties from bringing forward perfectly valid efficiency claims.

– Second, when faced with indications of possible efficiencies that directly benefit customers, the Commission should investigate such efficiencies directly with the parties that have the most relevant information, which are typically the customers rather than the notifying parties. The “claw back” theory should be abandoned and replaced with a robust balancing of the direct efficiencies with the price increases brought about by the transactions. The Commission should bear the burden of proof for the demonstration and quantification of these price increases (which may include the claw back), in line with constant case law. For other (non-direct) efficiencies, the parties would still bear the burden of proof of demonstrating their alleged impact on reduced prices.

– Third, the standard of proof for efficiencies, as recently applied in the decisional practice, should be somewhat lowered —for instance, the Commission should acknowledge that assessing efficiencies is necessarily a predicative exercise and that there is no reason in principle why the standard of proof for efficiencies should be higher than the standard of proof for anti-competitive effects. Historical studies should be more readily accepted. This may not necessarily result in a more permissive merger control policy. First, the Commission could apply a discount for efficiencies that only materialize in one entity rather than an entire industry, or which are likely to materialize after the anti-competitive effects of the transaction. Second, nothing should prevent the Commission from taking into account the possible “inefficiencies” created by a transaction. There is ample economic literature showing that mergers can sometimes be driven by empire-building impulses of eager managers rather than by the objective of increasing shareholder value. [103] If such inefficiencies are properly established (for instance though parties’ internal documents), they should be incorporated in the overall efficiencies analysis.

– In cases where efficiencies are insufficient to counterweigh possible anti-competitive effects, the Commission should also clarify that such efficiencies would nevertheless be taken into account to reduce the amount of remedies needed.

– Finally, an ex-post analysis of efficiencies in merger cases, similar to the study conducted by the Commission on the impact of remedies in 2005, could be contemplated. Are the claimed efficiencies typically overstated? Or, on the contrary, did some efficiencies that were initially neglected turn out to be higher than expected? This kind of assessment could allow the Commission and possibly other antitrust authorities to fine-tune their policy on efficiencies in merger control in a way that is as realistic and reflective of public interest as possible.


[1See, Richard A. Posner, Antitrust Law (2001).

[2See, Nicholas Levy, European Merger Control Law: A Guide to the Merger Regulation (2013), §15.02[5]. With the potential exception of Commission Decision of 12 May 2006 in Case COMP/M.4057 Korsnäs/Assidomän Cartonboard, §36 and 57-64, efficiencies “have not played a decisive role to date.” See also Damien Gérard, L’interdiction de la concentration Deutsche Börse/NYSE Euronext : quels enseignements pour la prise en compte des gains d’efficience ? RLC, April-June 2012, No. 31, 8.

[3There were 17 prohibitions decisions between 1994 and 2003 (out of 2,157 notifications, i.e. in 0.78% of cases) and only 6 between 2004 and 2013 (out of 2,990 notifications, i.e. in 0.20% of cases).

[4Oliver E. Williamson, Economies as an Antitrust Defense: The Welfare Tradeoffs’, The American Economic Review, Vol. 58, No. 1 (March 1968), 18-36.

[5Type 1 errors are false positives. It may even be claimed that efficiencies claims result in so-called “Type 3” errors, which encompass the increased costs to businesses, enforcers, and decision-makers from the increased complexity and uncertainty of an efficiencies defense. See Alan A. Fisher and Robert H. Lande, Efficiency Considerations in Merger Enforcement, 71 Cal. L. Rev. 1580, 1684-91 (1983).

[6See e.g. Daniel Crane, Rethinking Merger Efficiencies, Michigan Law Review, Vol. 110, 2011.

[7Commission Decision of 2 October 1991 in Case IV/M.53 Aérospatiale-Alenia/Havilland, 1991 O.J. L334/42, §65.

[8See, Commission Decision of 2 October 1991 in Case IV/M.53 Aérospatiale-Alenia/Havilland, 1991 O.J. L334/42, and, for instance, Commission Decision of 28 April 1992 in Case IV/M.126 Accor/Wagons-Lits, 1992 O.J. L204/1; Commission Decision of 9 November 1994 in Case IV/M.469 MSG/Media Service, 1994 O.J. L364/1; Commission Decision of 9 March 1999 in Case IV/M.1313 Danish Crown/Vertjyske Slagterier, 2000 O.J. L20/1.

[9See, for instance, Commission Decision of 14 May 1997 in Case IV/M.856 British Telecom/MCI, 1997 O.J. L336/1, and Commission Decision of 30 July 1997 in Case IV/M.877 Boeing/Mc Donnel Douglas, 1997 O.J. L336/16.

[10See, Phillip E. Areeda and Herbert Hovenkamp, Antitrust Law, Third Edition (2009), §970b.

[11Brown Shoe Co. v. U.S., 370 U.S. 294, 344 (1962), and FTC v. Procter & Gamble Co., 386 U.S. 568, 580 (1967). For a brief analysis, see Phillip E. Areeda and Herbert Hovenkamp, Antitrust Law, Third Edition (2009), §970c.

[12See, for instance, M. Monti, speech of June 4, 2002 (SPEECH 02/252): “There is no such thing as a so-called ‘efficiency offence’ in EU merger control law and practice.”

[13Council Regulation (EC) No. 139/2004 of 20 January 2004 on the control of concentrations between undertakings (2004 O.J. L24/1), article 2(1)(b).

[14See, European Council Regulation 139/2004, §29 of the preamble: “It is possible that the efficiencies brought about by the concentration counteract the effects on competition, and in particular the potential harm to consumers, that it might otherwise have and that, as a consequence, the concentration would not significantly impede effective competition, in the common market or in a substantial part of it, in particular as a result of the creation or strengthening of a dominant position.”

[15See, European Council Regulation 139/2004, §29 of the preamble: “It is possible that the efficiencies brought about by the concentration counteract the effects on competition, and in particular the potential harm to consumers, that it might otherwise have and that, as a consequence, the concentration would not significantly impede effective competition, in the common market or in a substantial part of it, in particular as a result of the creation or strengthening of a dominant position.”

[16See, Horizontal Guidelines, §79.

[17Horizontal Guidelines, §85.

[18Horizontal Guidelines, §86.

[19See, for instance, Commission Decision of 14 May 2008 in Case COMP/M.4854 TomTom/TeleAtlas, §246-250; Commission Decision of 2 July 2008 in Case COMP/M.4942 Nokia/NAVTEQ, §365-376; Commission Decision of 21 April 2009 in Case COMP/M.5152 Posten A/B/ Post Danmark A/S; and Commission Decision of 21 January 2010 in Case COMP/M.5529 Microsoft/Yahoo!, §957.

[20See, Case T-342/07, Ryanair Holdings plc v Commission [2011] ECR II-03457, §427.

[21Ibid., §406.

[22Action brought on 12 April 2012 – Deutsche Börse v Commission (Case T-175/12, O.J. C 174 from 16.06.2012, p. 25).

[23Action brought on 5 April 2013 – United Parcel Service v Commission (Case T-194/13, O.J. C 147 from 25.05.2013, p. 30).

[24On August 8, 2011, Western Digital appealed the priority decision included in the Decision (2011/C 165/04) of the European Commission of May 30, 2011, in Case COMP/M.6203 – Western Digital Ireland/Viviti Technologies, to open a second phase investigation with regard to the proposed concentration (Case T-452/11, Digital and Western Digital Ireland v Commission. On September 20, 2012, the proceedings were discontinued at the applicants’ request).

[25See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440.

[26See, European Commission Press Release No IP/12/94, February 1, 2012.

[27See, European Commission extends review of planned business combination of Deutsche Börse and NYSE Euronext, Press Release of August 4, 2011,, and the parties first announced $3 million efficiencies, and later raised their estimates.

[28See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1161-1165.

[29See Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1188.

[30See Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1244-1249.

[31See, Deutsche Boerse AG And NYSE Euronext Agree To Combine To Create The Premier Global Exchange Group, Press Release of February 15, 2011, See also, Commission Decision of 1 February 2012 (Case COMP/M.6166), §1331-1334.

[32See Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1329-1330.

[33Out of 350 pages of the decision, 60 pages are dedicated to efficiencies.

[34See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1166-1176.

[35Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1190.

[36Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1202.

[37Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1204-1243.

[38Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1252-1297.

[39Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1329-1330.

[40Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1331-1334.

[41Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1331-1334.

[42See, Case M. 6570, UPS/TNT Express, Commission Decision of January 30, 2013 (not yet published). See also Mergers: Commission prohibits proposed acquisition of TNT Express by UPS – frequently asked questions, Commission Memo, Memo/13/48 of January 30, 2013, available at

[43RBB Economics, Do Efficiencies Ever Deliver? Lessons From The UPS/TNT Case, March 2013, available at

[44See, Commission Decision of 3 October 2011 in Case COMP/M.6203 Western Digital Ireland/Viviti Technologies (not yet reported). Western Digital had to divest essential production assets for hard drives (notably a production plant), together with the divested business’ IP rights, personnel and supply of hard drive components. Western Digital also committed not to close its acquisition of HGST before concluding a binding agreement for the divestment, to a suitable purchaser approved by the Commission.

[45See, Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198.

[46Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §518-534. H3G committed to divest radio spectrum and additional rights to a new entrant in the Austrian market, with the right for the entrant to also acquire additional spectrum at an auction planned in 2013 by the Austrian telecom regulator. H3G also committed to provide on agreed terms wholesale access to its network for up to 30% of its capacity to up to 16 MVNOs in the coming 10 years. Finally, H3G committed not to complete the acquisition of Orange before it has entered into a wholesale access agreement with one MVNO.

[47Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §413.

[48Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §442.

[49Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria / Orange Austria, C(2012) 9198, §418, 438 and 442.

[50Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria / Orange Austria, C(2012) 9198, §444.

[51Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria / Orange Austria, C(2012) 9198, §423-424, 434 and 439.

[52See, Financial News, NYSE-Börse ramp up deal defence, October 20, 2011.

[53See, Horizontal Guidelines, §87.



[56See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1175.

[57See, Merger Regulation No 139/2004, Article 2(1).

[58See, Case T-151/05, NVV v. Commission [2009] II-1219, §163-165.

[59See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1268.

[60Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1285. See also §1264, where, after having quoted a presentation prepared from the London Stock Exchange on the impact on MiFID, the Commission wrote: “The impact on costs from increased competition resulting from MiFID is such that, if there were to be any beneficial effect resulting from the notified transaction, the marginal effect of it is likely to be very small. All in all, and as shown by academic literature, these reductions in transaction costs have been associated with an increase in volumes and reductions in bid-ask spreads.”

[61See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1286.

[62See, Horizontal Guidelines, §86: the Commission must be “reasonably certain that the efficiencies are likely to materialize” (one might inquire what is a “reasonably certain likelihood”, but this does suggest that absolute certainty is not required), and (more explicitly) “the longer the start of the efficiencies is projected into the future, the less probability the Commission may be able to assign to the efficiencies.”

[63See, Case T-342/07, Ryanair Holdings v. Commission [2010] II-03457, §412.

[64See, Horizontal Guidelines, §88.

[65See, Action brought on 12 April 2012 – Deutsche Börse v Commission (Case T-175/12, O.J. C 174 from 16.06.2012, p.25).

[66See, Horizontal Guidelines, §79-84.

[67See, Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §6.

[68Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §316.

[69Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1178 seq.

[70Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §421.

[71Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §424 and 434.

[72See, Horizontal Guidelines, §80-81.

[73See, Commission Decision of 12 December 2012 in Case COMP/M.6497 Hutchison 3G Austria/Orange Austria, C(2012) 9198, §424.

[74See Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1181 and 1237.

[75See, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §1336-1342.

[76See, for instance, Commission Decision of 1 February 2012 in Case COMP/M.6166 Deutsche Börse/NYSE Euronext, C(2012) 440, §827, 839, 843, 887, 959.

[77On the asymmetrical standard of proof for merger efficiencies and predicted anticompetitive effects in the United States and the European Union, see Daniel A. Crane, Rethinking Merger Efficiencies, March 2011, Michigan Law Review, Vol. 110, 2011: “the claim that efficiencies will likely occur and that, in the long run, they will destabilize competition to the detriment of consumers, is necessarily more speculative on average than a merger efficiencies defense since it relies on proof not only that efficiencies will occur but that they will subsequently destabilize competition. It makes no sense to hold defendants to a high standard of proof that pro-competitive efficiencies will result and simultaneously hold the government to a low burden of proof that anti-competitive efficiencies will result. What’s sauce for the goose is sauce for the gander” (p.40).

[78See, Guidelines on the application of Article 81(3) of the Treaty [now Article 101(3) TFEU], O.J. C 101 , 27/04/2004, pp.97-118, (“Article 81(3) Guidelines”).

[79Article 81(3) Guidelines, §96.

[80Article 81(3) Guidelines, §56.

[81Article 81(3) Guidelines, §53.

[82Article 81(3) Guidelines, §96.

[83Article 81(3) Guidelines, § 43 and 90-91. On the “balancing exercise,” see Case T-168/01, GlaxoSmithKline Services Unlimited v Commission [2006] ECR II-2969, §304-307.

[84Article 81(3) Guidelines, §101.

[85See, Commission Decision of 10 December 1984 in Case IV/30.717 Uniform Eurocheques, O.J. L 35 , 07/02/1985, pp.43-53, §33.

[86See, Case C-209/07 Competition Authority v Beef Industry Development Society Ltd, Judgment of the Court of 20 November 2008, § 39. The Commission confirmed that it followed the same approach in its Amicus Curiae submitted before the Irish High Court in Case No. 7764P The Competition Authority v. The Beef Industry Development Society Limited and Barry Brothers (Carrigmore) Meats Limited, §13.

[87See, Commission Decision of 17 October 2007 in Case COMP/D1/38606 Groupement des Cartes Bancaires “CB,” summary decision published in O.J. C 183, 05/8/2009, p.12, and Decision of 19 December 2007 in Cases COMP/34.579 MasterCard, COMP/36.518 EuroCommerce, and COMP/38.580 Commercial Cards.

[88See Commission Decision of 23 May 2013 in Case COMP/AT.39595 Continental/United/Lufthansa/Air Canada (“Star Alliance”), §55-79. See also Commission Decision of 8 December 2010 in Case COMP/39.398 Visa MIF, §62-68. The Commission accepted commitments in both cases.

[89See, Commission Regulation 1217/2010 of 14 December 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of research and development agreements, O.J. L 335, 18/12/2010, p.36, and Commission Regulation No 1218/2010 of 14 December 2010 on the application of Article 101(3) of the Treaty to categories of specialization agreements, O.J. L 335, 18/12/2010, p.43.

[90See, Guidelines on the application of Article 81(3) of the Treaty [now Article 101(3) TFEU], O.J. C 101 , 27/04/2004, pp.97-118, (“Article 81(3) Guidelines”).

[91See, Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements, O.J. C 11, 14/01/2011, p.1.

[92Ekaterina Rousseva, Reflections on the Relevance and Proof of Efficiency Defences in Modern EE Antitrust Law, in J. Bourgeois and D. Waelbroeck, Ten years of effects–based approach in EU competition law, Bruylant, 2012 p.290.

[93Case T-342-07 Ryanair Holdings plc v. Commission, 2010 E.C.R. II-3457, §29-30.

[94Case C-12/03 P Tetra Laval [2005] E.C.R. I-987, §39.

[95Case T-175/12 Deutsche Börse v Commission, action brought on 12 April 2012.


[97Joined Cases C-204/00 P, C-205/00 P, C-211/00 P, C-213/00 P, C-217/00 P, and C-219/00 P Aalborg Portland and Others v. Commission [2004] E.C.R. I-123, §279. See also Case C-42/84 Remia and Others v. Commission [1985] E.C.R. 2545, §34; and Joined Cases C-142 and 156/84 BAT and Reynolds v. Commission [1987] E.C.R. 4487, §62.

[98Case T-175/12 Deutsche Börse v Commission, Action brought on 12 April 2012.


[100Article 65 of the General Court’s Rules of Procedure and Article 64 of the CJUE’s Rules of Procedure.

[101Case C-48/69 Imperial Chemical Industries Ltd. v Commission of the European Communities, Judgment of the Court of 14 July 1972, and Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85, and C-125/85 to C-129/85 A. Ahlström Osakeyhtiö and others v Commission of the European Communities, Judgment of the Court of 31 March 1993.

[102See, RBB Economics, Do Efficiencies Ever Deliver? Lessons from the UPS/TNT Case, March 2013, available at

[103See, e.g. , F.M. Scherer, A New Retrospective on Mergers, 28 Rev. Indus. Org. 327, 329 (2006).

© 2017 - Institute of Competition Law Download our brochure