This is the fourth part in the Commerce Amendment (Promoting Competition and Other Matters) Bill (Bill) series. This article looks at proposed amendments to the merger control regime.
Problem definition(s)
The relevant Cabinet Paper details several perceived issues with the merger control regime which, overall, it considers could be improved to reduce regulatory burden, costs to business, predictability and transparency:
- a perceived issue with the Commission’s ability to assess so-called ‘killer’ and ‘creeping’ acquisitions. The second edition of this series examined killer acquisitions;
- uncertainty around terms such as “assets of a business” and “substantial degree of influence”;
- a ‘gap’ in the Commerce Commission’s (Commission’s) toolkit, meaning that it is only able to accept divestments of assets or shares to remedy competition issues, but not ‘behavioural’ undertakings such providing access to a key input or interoperability;
- a perceived limit on the Commission’s ability to intervene in mergers that are not formally notified to it, which gives rise to a risk that “anti-competitive mergers are completed before scrutiny is possible;
- a lack of clear statutory timeframes, which can cause uncertainty for businesses and consumers; and
- no time limits for the Commission to publish determinations, which delays guidance to the market, limits opportunities for challenge and may deter similar transactions.
Assessing the proposed amendments
In previous editions in this series the discussion followed a ‘problem definition – proposed amendments – discussion of the proposed amendments’ format. For merger control there are just too many amendments for that to work, and so the proposed amendments are set out and discussed in turn immediately below.
‘Killer’ acquisitions
The second edition in this series examined the proposal to expand the meaning of the substantial lessening of competition (SLC) test, and why the case for change does not appear to stack up (in particular for changing it throughout the Commerce Act (Act)).
‘Creeping’ acquisitions
Creeping acquisitions are commonly described as a person making multiple acquisitions over a period of time, where the effect of each individual acquisition does not SLC, but their combined impact does.
The Bill would introduce a new section 3(8) into the Act providing that the “effect” of an acquisition would include, where relevant, the combined effects of:
- the current acquisition; and
- in the preceding three years, any other acquisition by any party to the current acquisition (including interconnected or associated persons).
The provision is similar to the one that has been introduced into Australia’s Competition and Consumer Act 2010 (CCA), but applies more broadly. Whereas Australia’s provision limits the aggregation to those targets involved in the supply of substitutable or competitive goods or services, the proposed amendment to the Act would not.
Carparking (specifically, Wilson Parking) has been singled out as an industry where creeping acquisitions are a potential concern. However the Commission has assessed several acquisitions by Wilson Parking over the years, and has succeeded in obtaining divestments where it identified competition concerns. See here, for example.
The Regulatory Impact Statement (RIS) says that cumulative acquisitions have been observed in other industries such as veterinary care, health and funeral services, but does not identify any competitive harm or, more importantly, any inability for the Commission to address it under the current test.
Given the above, the case for change does not appear to have been made out. The SLC test as it currently operates is well-understood, flexible and adaptive. There is no evidence it is not capable of dealing with creeping acquisitions. Indeed, as discussed in the second part of this series, the Commission’s stance is that “as a matter of law” the SLC test applies to all acquisitions, no matter the market share increment. It has been able to take action against Wilson Parking.
As for the amendment that is proposed, aggregating the effect of multiple transactions that occur over several years raises significant analytical issues. The SLC test, as applied in practice, involves comparing the likely state of competition ‘with’ and ‘without’ an acquisition. In this way, the competition effects of individual acquisitions can be isolated from other changes occurring in markets.
Aggregating the effect of multiple acquisitions over time does not (easily) lend itself to this sort of analysis. The fact that the Australian Competition and Consumer Commission’s (ACCC’s) merger assessment guidelines do not provide much, if any, insight into how it plans to approach this assessment may be indicative of the complexities involved. This is likely to increase uncertainty for stakeholders.
Finally, the provision appears to be drafted more broadly than is required to address the concern about creeping acquisitions, which are the aggregation of market share over time. Such aggregations arise through horizontal mergers, and the equivalent provision in the CCA therefore limits consideration of past mergers to those involving substitutable or competitive goods or services. There does not appear to be a clear basis for extending the reach of this provision to all acquisitions in New Zealand, and doing so is likely to increase regulatory burden.
Clarifying key terms
The Bill would amend two key terms relevant to merger control: “assets of a business” and “substantial degree of influence”.
Assets of a business
Acquisitions of “assets of a business” that give rise to an SLC are prohibited under section 47 of the Act. What “assets of a business” means has been the subject of debate over the years. Does it mean any assets that form part of, or are owned by, a business? Or does it mean that only those assets that form the core of a business ie that are essential to the operation of a business, are captured?
The proposed amendments attempt to resolve this debate by deleting the words “of a business” in section 47(1) and amending the definition of “assets” in section 2(1) to provide that it includes property of any kind, whether tangible, intangible, legal or equitable rights, as well as non-property interests.
The upshot of the proposed amendments is that it would be clear that individual assets and property fall within the ambit of merger control. For example, anticompetitive land banking, or the acquisition of IP rights, would clearly be caught.
This appears to be a sensible amendment and will reduce uncertainty.
Substantial degree of influence
The second clarification concerns what constitutes a “substantial degree of influence” for the purpose of determining whether persons are “associated” Whether an acquirer is “associated” with another person is important because, if they are, the Commission will treat them as a single person for the purpose of assessing the effects of a merger.
The proposed amendment would add a list of non-exhaustive factors that the Commission is able to consider when determining whether an acquirer has a substantial degree of influence over another person (and vice versa). These include things such as voting rights, the right to appoint and remove directors and veto powers over strategic decision-making.
While the amendment to the Act would bring clarity into the legislation, the Commission’s Merger Assessment Guidelines (MAGs) already provide guidance, including a list of non-exhaustive factors that cover much the same ground as the proposed amendment (see paragraph [2.8] of the MAGs).
Given the above, this would not appear to make much of a difference to how the Commission currently undertakes this assessment.
The behavioural undertakings ‘gap
Now we come onto one of the more significant amendments in the Bill: the proposal to enable the Commission to accept ‘behavioural undertakings. The Commission would get this power under the Bill, but with some significant limits. The new regime would look like this:
- in addition to divestments, in giving clearance or authorisation to a merger the Commission would be able to accept undertakings “to take or refrain from taking an action specified in the undertaking” and to pay the Commission’s costs of monitoring or enforcing the undertaking;
- however, the Commission would only be able to accept an undertaking to take or refrain from taking action if it is satisfied that:
- it could not give approval in the absence of the undertaking; and
- divestment would be “insufficient” or the undertaking is “reasonably necessary” to give effect to a divestment undertaking; and
- the undertaking is reasonably practicable for the Commission to monitor and enforce, having regard to the costs of doing so and any undertaking to pay those costs; and
- on appeal, the court may not reverse or modify any part of a determination only because the Commission did not accept an undertaking for a person to take or refrain from taking an action.
The upshot of these amendments is that the Commission would be able to accept behavioural undertakings, but in quite limited circumstances, and even then it would have a wide discretion to do so (or not).
A couple of observations on this.
First, in order to accept a purely behavioural undertaking (that is, not one that supports a divestment), the Commission would need to be satisfied that a divestment would be “insufficient”. What this wording means in context is unclear. Presumably it means that a divestment would be insufficient to mitigate or avoid the effects of an SLC. But there will almost always be a divestment that is, at least theoretically, capable of doing that (just not always one that is commercially viable). If this is what is intended, then the current wording may be, well, insufficient.
An alternative would be to shift the focus to whether a behavioural remedy is proportionate to the harm that it is seeking to address, taking into account monitoring costs etc. If a behavioural undertaking is as (or more) effective than a divestment, and the ongoing costs are proportionate to the harm being avoided, then the Commission should be able to (and should) accept such an undertaking.
Second, on appeal the courts would not be able to overturn or modify a determination solely on the basis that the Commission did not accept a behavioural undertaking. This would essentially leave the Commission’s decision not to accept a behavioural remedy as unreviewable by the court. While the court would still be able to refer the matter back to the Commission for reconsideration under section 94 of the Act, this is a blunt and inefficient instrument. In my view all aspects of merger determinations should be subject to review by the courts.
Finally, the amendments are confined to the merger clearance and authorisation processes – a court would still not be able to accept behavioural remedies in merger litigation. The likely intention here is to incentivise merger parties to apply for clearance. However, at the end of the day, New Zealand’s merger control system is still voluntary, and the courts should have at their disposal a toolkit that enables them to remedy competition issues where they arise (indeed, elsewhere in the Bill the Government is proposing to give them performance injunction powers to remedy competition concerns). In cases where divestments are appropriate/possible, the only lever they have to pull is penalties, and this denies them the opportunity to actually deal with competition effects arising from permanent changes to market structure. Not giving them the necessary tools would ultimately harm competition and consumers.
Non-notified mergers
Next up is another one of the key amendments in the Bill: giving the Commission new formal powers to ‘suspend’ acquisitions and require merger parties to apply for clearance or authorisation.
Under new section 47E if a person has, or proposes to, acquire assets or shares and the Commission has reasonable grounds to believe that it is necessary to protect competition while the Commission assesses the potential application of section 47, then the Commission may “impose a suspension” on an acquisition. During the suspension period – which must not be longer than 40 working days – the person would be prohibited from acquiring the assets or shares, or must take steps to ensure they are safeguarded. The Commission may set out what steps are to be taken to safeguard the assets or shares.
Under new section 47F, the Commission could compel a person to apply for merger clearance or authorisation if the Commission has “reasonable grounds” to believe that a proposed acquisition “has the potential to” substantially lessen competition in breach of section 47.
Under section 47G, when exercising these powers the Commission would be required to give reasons, to all merger parties, and to make these reasons public.
In terms of addressing concern that anticompetitive mergers may be slipping through the cracks, the Commission already operates a highly successful informal ‘courtesy letter’ process whereby parties can – and frequently do – advise the Commission of proposed acquisitions notwithstanding that they do not intend to apply for clearance. The Commission receives between 10-20 of these each year (and it is on track to receive far more than that this financial year). Of these, very few are ‘called in’ for clearance.
While non-notified mergers do occasionally lead to court proceedings, there is no indication that these are at problematic levels. The Commission typically brings about one merger case to the courts every 4-5 years.
So, the proposed amendments may be unlikely to significantly change how merger parties engage with the Commission in many cases.
A couple of observations on the terms of the amendments themselves.
First, the Commission would be able to issue a suspension order in relation to a proposed acquisition under section 47E if it has reasonable grounds to believe it is necessary to protect competition while it assesses whether the acquisition “has the potential to” breach section 47 ie substantially lessen competition. In other words, it would need to form a view that there is competition that needs protecting before it has reached a view that it needs protecting. Leaving aside the inherent contradiction in the wording, it would be inappropriate for the Commission to intervene in a transaction when it has not even determined whether the transaction “has the potential” to breach section 47. This would give the Commission more power than is warranted, and too early, to take what is likely to be very disruptive action.
Second, no end date is proposed for ‘call in’ notices issued under section 47F. While section 47G provides for the Commission to withdraw a notice at any time, this would be entirely at the discretion of the Commission. Market conditions can change over time (particularly in digital markets, where they can change quickly) and there should not be the potential for transactions to be barred in perpetuity. This is particularly the case where even acquiring some of the assets or shares covered in a notice would be prohibited, and a notice would therefore effectively prevent a revised deal to proceed that did not raise competition issues.
Finally, there are likely to be circumstances where publishing reasons for suspension and call-in notices will be inappropriate and damaging to the interests of the merger parties. Fact confidentiality is important for many transactions before they are publicly announced, and in some cases parties decide not to proceed with a transaction if clearance or authorisation is required. The Commission’s MAGs recognise the importance of confidentiality in relation to pre-notification discussions, which it treats as fact-confidential (see paragraph [6.7] of the MAGs). It also does not publish (or publicise) informal call-in letters under the current regime. Confidentiality in deal-making should continue to be provided for.
Time limits
Finally, the Bill would introduce new ‘maximum’ time limits for the Commission to make determinations on mergers, and to produce written reasons.
Currently the Commission has 40 working days to make a decision on merger clearance, and 60 working days for merger authorisation. These timings can be extended by agreement with the applicant. There are no restrictions on why they can be extended, and no provision for the statutory timeline to be paused in any way.
Under the proposed amendments, the statutory maxima would be 140 working days for clearances and 160 working days for authorisations. However, extensions and pauses to the statutory timeline would be permissible where certain pre-conditions are met. For extensions these are:
- the complexity of the acquisition;
- the Commission deciding to hold a conference;
- undertakings are offered during the Commission’ assessment; or
- a change in circumstances or “other event” occurs that materially affects the Commission’s consideration of the application.
A pause to the statutory timeline would be permissible where:
- an overseas regulator is considering the acquisition; or
- in the case of non-compliance with a voluntary or compulsory information request by a non-applicant.
In relation to written reasons for merger clearances and authorisations, the Commission must:
- provide a summary of reasons for its decision within 1 working day;
- provide a full statement of reasons within 20 working days; and
- make the reasons for its decision publicly available as soon as practicable, acknowledging that it may omit certain eg confidential information.
The Commission has in the past been criticised for taking too long to publish documents. These include not only final determinations, which are the subject of the proposed amendments, but provisional and procedural documents as well, such as Statements of Issues.
In recent times the Commission has made a concerted effort to reduce the time it takes to publish these documents, and is making excellent progress with publication times reducing significantly. These amendments build on that progress, but some observations.
First, one of the reasons for extending time is the “complexity of the acquisition”. However it is the complexity of the competition analysis that is more likely to be relevant (how many different relevant markets, the economic analysis required, etc.). This is a more appropriate trigger in my view.
Second, an applicant’s failure to respond to an information request would not be grounds for the Commission to pause the statutory timeline. Presumably this is to incentivise applicants to respond quickly, however applicants are generally very highly motivated to respond to Commission information requests in a timely manner. As drafted, applicants could be unfairly disadvantaged in the event that a Commission information request is requires clarification or refinement, or where it requests a large amount of information.
Summing up, the proposed amendments to the mergers regime are a bit of a mixed bag. The creeping acquisitions amendment appears unnecessary, complex to apply and likely to increase uncertainty. Giving the Commission the power to accept behavioural undertakings is a welcome development. It is unclear how new formal suspension and call-in powers will affect many deals in practice.
If you would like to discuss any aspect of what has been covered in this article, please get in touch.
