ICC Policy Statement

Comments on the ICN Merger Remedies Review Project

Prepared by the Commission on Competition
submitted to the ICN Analytical Framework Subgroup, Mergers Working Group

Introduction
The ICC Commission on Competition greatly appreciates the opportunity to review and comment upon the Draft Report of the Merger Remedies Review Project ("the Report"). We believe that a comprehensive review of remedy options and methodologies can provide a useful resource for ICN members.

Overview Comments
The chief benefit to be achieved from the Report, we believe, is the sharing of experiences of various competition authorities on the challenges of developing effective remedies. The most effective passages of the Report are those that examine the past efforts to implement various remedies and identify the key benefits and challenges that have been encountered. To this end, we would encourage the drafters to focus further attention on providing specific information derived from the work papers solicited from various ICN participants and incorporate those examples into the overall analysis. Such illustrations have been successfully used in other ICN fora as well as the past work of the Analytical Framework subgroup. The examples are often provided on an anonymous basis. Notably, the Report would benefit from commentary by both large and small competition authorities on the relative costs and benefits of certain types of remedies as the experiences of agencies as to remedies may differ in this regard.

We also acknowledge and support the development of the "Remedy Options" envisaged as Appendix A to the Report. An organisation and cataloguing of remedy options will engender broader knowledge and understanding of remedies, their use and their potential value in resolving competitive concerns.

We believe that the Report has one significant negative aspect in that, in far too many cases, it takes a prescriptive approach toward remedy measures. While identifying the challenges associated with a potential remedy is a beneficial device, suggestions as to what should or should not be done, or how it should or should not be done, in particular cases are undesirable in this report. In essence, we do not believe that the Report should take on the character of "Recommended Practices" for several reasons. First, we believe that such an approach would be more in keeping with the stated purpose of the Report. Second, the process by which the Report was prepared is not conducive to the development of recommended practices or proposed guidelines in that it did not include from its inception a broad cross- section of the ICN members or NGAs in the drafting or preparation of the proposed language. Third, the ICN Subgroup on Merger Notification and Procedures is preparing a recommended practice on Remedies. By taking a prescriptive approach, the Report threatens both to undermine the Recommended Practices being prepared by the N&P Subgroup and also threatens to create confusion for ICN members in terms of remedy procedures.

There is ample scope for both the Recommended Practice on Remedies and the Merger Remedies Review Project and that, indeed, these two projects can complement one another and, in tandem, create synergies. To do so, however, each should be careful to remain in its own province. To this end, we would recommend a review of the language of the Report to ensure that observations on challenges are not issued as recommended practices. To assist in that effort, we have provided a separate draft highlighting those passages that should be reconsidered in this regard.

One additional concept that deserves more attention in the Report is the interaction between remedies and efficiencies. Importantly, the draft should recognize that most mergers are presumptively efficiency-enhancing - i.e., that the parties seeking to invest in new businesses are investing in the proposition that synergies can be derived. Additionally, free market economies operate on the assumption that there should also be a free market for corporate governance and control. Presumably, parties willing to go forward with a transaction where a remedy is being imposed retain the belief that the transaction will generate such synergies. The Report, however, seems consistently to suggest that the parties should have an affirmative burden of demonstrating that a transaction is efficiency-enhancing in order to warrant agency approval. We believe that a transaction subject to a remedy that addresses competitive concerns should be allowed to proceed without any affirmative demonstration of efficiencies by the parties. Efficiencies are properly taken into effect, we would agree, in those instances in which remedies are not imposed or when remedies do not wholly address likely anticompetitive effects. This concern arises, for example, in Sections 1.4, 2.1, 2.2 and 2.6.

In large part, decisions on remedies involve the application of predictive judgment on the costs and benefits not just of the remedies themselves, but of the overall transaction in light of the remedies sought to be imposed. For instance, a lesser remedy - or no remedy at all - may be appropriate in instances reflecting some degree of competitive detriment but promising significant net efficiencies. This is in keeping with promoting, as the ultimate goal of competition enforcement, consumer welfare (however that term may be defined by a particular agency). While the Report makes reference to this balancing act in a few instances, it may benefit from further analysis of the importance of considering net efficiencies in light of remedies and of affirmative statements that affirmative proof of efficiencies normally is not required in transactions where the competitive concerns have been adequately addressed.

Comments on Particular Provisions of the Report

We offer the following comments on the specific provisions of the Report for your consideration. These are formatted both as substantive editorial suggestions (when in bold) and as comments.

Sections 1.4 and 1.5:
These two sections should contemplate that in considering remedies an agency should evaluate the net competitive detriment attributable to a transaction. This is important, particularly in the context of multi-national transactions, because agencies should allow for the possibility that a transaction with small or uncertain competitive detriment might still be allowed to proceed if: (a) the transaction on balance has significant competition-enhancing economic efficiencies; and (b) the remedy that would be required to eliminate the competitive detriment would either undermine the economic efficiencies to be gained or otherwise be difficult or impossible to administer. This point is made indirectly in Section 2.4, but should be part of the overall balancing act performed by the agency in the first instance, not an afterthought. Suggested changes to the draft language are below.

1.4 Remedies should only be applied to address the identified competitive detriment expected to arise from a merger transaction. In the absence of remedies, competitive detriments resulting from a merger could only be addressed by prohibiting the transaction in its entirety or by accepting the detriment as part of the transaction, if on balance the transaction is pro-competitive. The key contribution of remedies is to enable a modified outcome to merger transactions which restores or preserves effective competition while permitting the realisation of relevant merger benefits, thus achieving a better outcome than straightforward prohibit or permit decisions. An anti-competitive merger with no obvious benefits would be more likely to be prohibited outright - although remedies might be appropriate if the detriment to competition was in a distinct and easily severable part of the merging business. For the purpose of this review, prohibition will not be considered to be a "remedy" but will be regarded as an alternative outcome to a merger decision. .

1.5 In order to put in place an effective remedy, it is necessary first to have a clear idea of the net competitive impact that would result from the merger, i.e., the extent to which competitive harm resulting from the transaction exceeds competitive benefit. Any decision on remedies must therefore follow a decision on net competitive detriment. Merger remedies are not tools of industrial planning and are generally ill suited to achieve aims wider than addressing the competitive detriment. Early discussion of remedies is desirable but should not distort the process of identifying competitive detriments.

Section 1.6
The draft may wish to note that public interest standards, to the extent they are required to be taken into consideration, should be recognized potentially to distort the optimal competitive outcome. In any event, authorities should endeavour to accommodate these non-competition public interest considerations, to the extent required by law, in a way that minimizes the distorting impact of the remedy on competition.

Section 2.3
In addition to noting the nee d to consider the net competitive impact, this section should also note that agencies may need to inform parties of their perspective on the nature and scope of the detriment, in order to enable the parties to propose effective solutions. This is especially true if Section 2.3 is going to suggest that the burden is on the parties to propose remedies in the first instance.

2.3 If a merger were to result in net competitive detriment to which there were no effective remedies, the merger would normally be prohibited. The merging parties will therefore normally have strong incentives to propose acceptable remedies and in many instances, the burden for proposing effective remedies should thus fall mainly on these parties. Agencies may wish to ensure that their views on the nature and scope of the competitive detriment are clearly communicated to merging parties so that effective remedies may be proposed by the parties in a timely manner.

Section 2.4
It would be helpful to add the following language to the end of the draft text: "or where disproportionate efficiencies or other pro-competitive effects of the merger would be prevented by the remedy."

Section 2.5
It would be useful in this section to point out that, as the assessment of competitive detriment is a predictive judgment, some flexibility should be allowed in the development of effective remedy packages. Also, we would recommend modification of the second bullet to eliminate the double negative, as follows:

-- Acceptable Risk. All remedies are, to some extent uncertain as to their eventual impact. Effective remedies should provide reasonable assurance of resolving the competitive detriment. This is particularly important where a competition authority is restricted in its ability to modify a remedy in the event of it failing to perform as anticipated.

Section 2.6
We believe the language in this section needs to be revised in one significant respect. In discussing expected efficiencies, the Section states that efficiencies may only be considered beneficial "to the extent that they are likely to create greater competitive rivalry with resulting benefits to customers." We believe both that this hurdle is too high and that it is not followed by the majority of international competition agencies. Rather, we believe the appropriate standard is that efficiencies should be considered so long as they "are reasonably likely to result in benefits to consumers."

2.6 The potential burden or cost of using remedies is another element which should be taken into account. Burdens may arise in a variety of areas:

- Remedy impact costs. Remedies may result in distortions or inefficiencies in market outcomes. This is more likely to be the case where a remedy involves direct intervention in market outcomes, especially over a long period. For example, price caps may discourage market entry by creating doubt concerning the ability to recoup investment or to maintain profitability
- Remedy operating costs. These comprise the directly attributable costs of implementing and, if necessary, monitoring and enforcing remedies e.g. employing trustees, collecting monitoring information etc.
- Merger efficiencies or other benefits foregone. A frequent advantage of remedies is that they enable the realisation of at least some efficiencies or other benefits expected from a merger that would otherwise be lost through prohibition. Particular benefits expected from a merger may include lower prices, higher quality, a greater choice of products or a greater rate of innovation. Jurisdictions may differ significantly in how merger efficiencies and other benefits are defined and assessed. However, in general, these benefits are only relevant to the extent that they arise from the merger and would not have occurred otherwise. In addition, expected efficiencies to be gained by the merging parties may only be considered beneficial to the extent that they and result in benefits to customers. Depending on the value it places on these efficiencies and other benefits, a competition authority might wish to modify the choice or design of a remedy to minimise the impact on these efficiencies or other benefits. But the competition authority should still ensure that the remedy is effective in addressing the competitive detriments.

Section 3.1
This section should make clear that agencies may want to evaluate remedies only after reaching a determination that a competitive detriment is likely to result from a transaction, not simply because a remedy is offered by the parties. It is important for parties and agencies to avoid the use of remedies as a means of procedural leverage or as a default measure.

Section 3.3
In discussing the desirability of co-operation between competition authorities, this section should suggest that, after completing their assessment of the likely competitive impact of a proposed merger, authorities in jurisdictions where the merger does not have its centre of gravity may wish to wait until authorities in the other jurisdiction(s) which may be better situated in terms of nexus to the transaction and enforceability of remedies, have settled upon a proposed remedy. Where the remedy proposed in the latter jurisdiction would also substantially address the identified competition concerns in their jurisdiction, the former jurisdictions may wish to consider the desirability of refraining from requiring additional remedies in respect of comity considerations.

Section 3.5
It would be helpful to relocate this section to follow the current Section 3.7. It is somewhat out of place to put this discussion before the explanation of structural and behavioural remedies. Additional proposals for modification of the language, for purposes of clarification, are below.
3.5 (new 3.7) In many jurisdictions there is an initial presumption, at least for horizontal mergers, that a structural remedy is preferable to a behavioural remedy. These jurisdictions reason that a structural remedy is likely to be more effective, as it addresses the cause of the competitive detriment directly, and will incur lower ongoing costs of monitoring or possible market distortion. However, as noted later in this section, there may be significant constraints on a divestiture which may significantly affect the design and suitability of this remedy.

Section 3.8
Similar to Section 2.6, the language of this section suggests a higher burden than exists in most jurisdictions, i.e., it suggests that the merger must strengthen competition rather than preserve competition. The proposal below is designed to alleviate this issue.

3.8 Divestitures are generally the most common form of structural remedy. In essence, a divestiture seeks to remedy the competitive detriments of a merger by either creating a new source of competition through disposal of a business or set of assets to a new market participant or permitting an existing source of competition to step into the competitive role of one of the merging parties through disposal of assets to an existing market participant independent of the merging parties. To be effective, a divestiture should involve the sale of an appropriate divestiture package to a suitable purchaser through an effective divestiture process. These three key elements may be subject to significant constraints in individual merger cases. The effect of these on the suitability and design of divestitures is explored later in this section.

Section 3.10
The initial sentence should establish the fundamental premise that the scope of relief should normally be the "smallest divestiture package sufficient to address the expected competitive detriments." The final sentence of this section, which reads, "[i]t may be necessary to add to this package in order to secure a suitable purchaser," undermines the principal that precedes it. We believe the statement should be removed. At a minimum, if this statement is made, it should be noted that adding to the scope of a divestiture package beyond the smallest package necessary to relieve the competitive detriment can distort the optimal outcome: it removes the matter from the realm of competitive analysis and permits a form of procedural "extortion" on the part of the prospective buyer that can result in undesirable outcomes, including a chilling effect on the willingness of merging parties to propose or accept remedies.

Section 3.11
This Section should recognize that a fringe player, with adequate current involvement in the market, or a non-committed entrant, with existing production assets, may prove exceptions to the general rule that a structural divestiture normally will require the divestiture of an ongoing business. A fringe player may require only additional production assets, without requirement of the additional trappings of an ongoing business (e.g., sales and distribution assets), and may be a more efficient competitor for not having these additional assets. Likewise, a non-committed entrant may require only sufficient distribution resources or customer contracts in order to become an effective competitive counterweight to the merged firm.

Section 3.12
A slight modification should be made to the language of this section to allow for a connection, which occurs frequently, between the purchaser of the divested assets and the seller. This happens, for instance, where Buyer is acquiring a division of Seller, and the remedy results in Seller maintaining some portion of the business it had proposed to sell to Seller. It also happens, in some cases, where the divestiture is made to top executives of Seller. The proposed language is designed to eliminate this issue.

3.12 A suitable purchaser should have no significant connection to the acquiring party, should have the necessary resources and expertise to develop as an effective competitor and should not itself be subject to significant competitive concerns if the divestiture proceeds. An exception may be appropriate to allow, for example, for the supply of inputs to be maintained for a transitional period of time. A competition authority will also wish to satisfy itself that the purchaser has appropriate business plans and incentives for competing in the relevant markets.

Section 3.13
This section should state the basic premise, which currently is not stated, that a merger normally is permitted to proceed when the parties have committed to undertakings that reasonably can be expected to be carried out by the parties or enforced by the agency if the parties fail to carry it out.

Section 3.15
Clarifying language, as follows, would be helpful.

3.15 In order to protect a divestiture against likely asset risk, it may be necessary to require the divestiture package to be held and managed separately from the retained business, following the closing of the transaction. Appointment of an independent monitoring trustee is generally desirable to ensure that these "hold separate" conditions are complied with and that the divestiture package is not allowed to deteriorate. The use of trustees is discussed in more detail in part 4 below.

Section 3.19
This Section should note that the view of what constitutes a "behavioural remedies" varies according to jurisdiction. The statement of preference for structural remedies undermines to a significant degree the utility and frequency of behavioural remedies as they are used in many jurisdictions. Perhaps a distinction should be made between behavioural remedies that are of a curative nature or which interfere directly with the operation of the market - such as price controls - and behavioural remedies that are of a precautionary nature - such as firewalls to guard against improper sharing of information. The latter behavioural remedies are frequently used both in conjunction with structural remedies and independently and should be highlighted as a beneficial form of remedy.

Section 3.20
This section should also note that many jurisdictions do not deem it necessary to include conditions in a remedy that simply restate existing laws that can be enforced by the agency under its existing statutory authority.

Section 3.21
The parenthetical example in the third sub-bullet, "(as for example in some vertical mergers)," should be removed. The potential for significant economic benefits apply at least equally to horizontal mergers and other transactions. A stated, this parenthetical causes confusion. If particular case studies demonstrate instances in which behavioural remedies have worked well in vertical mergers, such cases bear mention and further discussion, but not to the exclusion of other types of mergers. It would also be useful to mention that behavioural remedies may have particular utility in those situations in which the problem sought to be prevented is itself behavioural rather than structur al. This might occur, for instance, where the concern arising from a merger relates to future product tying or, in some jurisdictions, product bundling.

Section 4.1
The language in this section, particularly the language in the first sub-bullet reading "[i]t should be borne in mind at the design stage that an element of complexity in the design of a remedy may rapidly escalate into very high levels of complexity when seeking to implement the remedy in a legally enforceable form," is confusing and would benefit from clarification. Also, it is not clear that the language is particularly helpful or instructive without concrete examples.

Section 4.2
The last sentence should be deleted. Trustees should be independent, subject to a certain degree of oversight by competition authorities. That concept is already covered in section 4.3.

Section 4.4
It would be helpful to add a statement to indicate that the trustee generally should not impose undue costs or burdens on merging parties.

Section 4.6
It is not clear whether this section is speaking to administrative burden or administrative effectiveness. In any event, we believe the statement that "[i]t is easiest to involve market participants in monitoring when they are relatively large, well informed and well resourced" is unhelpful. The purpose of remedies should be to protect all consumers, regardless of size, and large consumers are not always representative. Indeed, large consumers can often protect themselves against the exercise of market power when small consumers cannot. Well-informed viewpoints should be solicited from consumers irrespective of size and based primarily on the nature of the competitive detriment and of the proposed remedy.

Section 4.7
The suggestion of regular monitoring points is desirable, but the interval chosen "(e.g., every six months)" would be more useful if it reflected common practice. In our experience, it is far more common for compliance to be reviewed annually.

Section 4.8
Without further explanation or analysis based on the experience of agencies, this section seems speculative and should be removed.

Document n° 225/617
12 April 2005