Select Committee submission on the Cartel Criminalisation Bill

I have previously posted about the fact that the government is proposing to criminalise cartel conduct, and that while I am generally supportive of this type of measure the hard policy work to ensure it works well doesn't seem to have been done in this case.

On Friday, I made a Select Committee submission on the to the Commerce (Criminalisation of Cartels) Amendment Bill setting out my views and recommendations. The text of that submission is set out below (with footnotes omitted). In short, I consider that criminalisation should be restricted to serious and obvious cartel conduct until the hard policy work is done to better understand its likely effects in the New Zealand economy.


My name is Dr Edward Willis. I am an academic at the University of Auckland’s Faculty of Law where my research specialisations include competition law issues. Prior to taking up an academic appointment I was a senior practising solicitor and widely recognised as a leading New Zealand expert on economic regulation.

This document is my submission to the Economic Development, Science and Innovation Select Committee on the Commerce (Criminalisation of Cartels) Amendment Bill (Bill). The submission is based on my research into, and experience working with and advising on, New Zealand competition law issues. I wish to appear before the Committee to speak to my submission.

Effective competition law is essential to New Zealand’s economic success. Meaningful market‑based competition is the engine that drives innovation and productivity, delivering value for consumers (through new products, lower prices and higher quality) and strengthening New Zealand businesses to compete on the world stage. The Bill provides an opportunity to consider an important aspect of competition law, and I thank the Committee for the opportunity to make this submission.


Cartel conduct is recognised internationally as the ‘supreme evil’ of competition law. Cartels allow businesses to shield themselves from competition, deny the benefits of that competition to consumers and the wider New Zealand economy. Because cartels are by their nature covert, they can be extremely difficult to detect and so can cause lasting damage to the national economy for the period in which they operate.

I understand that the policy intent of the Bill is to promote stronger deterrence for cartel conduct. I broadly support that policy intent of the Bill. However, I consider that the scope of criminalisation should be more limited than currently provided for in the Bill. In particular, I consider that criminalisation should not apply to all cartel conduct, but only the more egregious and clear-cut example of such conduct. My reasons for holding these views are:

  • the risk of over‑deterrence in New Zealand’s small, isolated economy, which relies on a high degree of collaboration to produce the benefits of market competition;

  • the likely lack of effectiveness of criminalisation as an additional deterrent in the majority of situations in which cartel conduct occurs;

  • the challenges of successfully prosecuting all but the most egregious conduct under the institutional arrangements of New Zealand’s competition law enforcement regime; and

  • the potential of criminalisation to impact negatively on international co‑operation arrangements to which the Commerce Commission is a party.

For those reasons, I recommend that the Bill be amended so that:

  • big‑rigging is defined as a separate offence under the Commerce Act, which is subject to a criminal sanction of a maximum 7 years’ imprisonment;

  • cartel conduct that involves co‑ordination among competitors in a deliberately deceptive manner is defined as a separate offence under the Commerce Act, which is subject to a criminal sanction of a maximum 7 years’ imprisonment; and

  • all other horizontal conduct is treated as a regulatory offence only, and is not subject to criminal sanctions.

New Zealand economic context

The New Zealand economy is small and isolated. This has important implications for competition law policy.

There is only limited demand in many sectors of the New Zealand economy. Research indicates that where demand is limited it is difficult for businesses to achieve minimum efficient scale before market demand is exhausted. As a result, small market economies tend to be characterised by highly concentrated markets. The available evidence suggests that New Zealand’s economy is no exception, and is perhaps unique in the OECD as to how acutely these effects are felt due to its relative isolation.

There are a number of implications resulting from New Zealand’s high market concentration and small economy. The two that are most relevant to the issue of cartel conduct are that:

  • single businesses are unlikely to operate efficiently or be able to pursue meaningful innovation (the driver of consumer value) in the absence of collaboration; and

  • industry associations and alliance structures will be essential features of many sectors.

In practice, this means is that collaboration and joint activities, including between (potential) competitors, are critical to achieving the efficiencies that are central to New Zealand’s competition policy. This is already recognised in the Commerce Act, which contains a ‘collaborative activities’ exemption for conduct beneficial to the New Zealand economy that might otherwise be classed as cartel conduct.

None of this is intended to suggest that cartels are a feature of competitive or efficient markets. They are not, and can only be harmful to the market mechanism that we rely on to deliver innovation and economic growth. However, the need for collaboration points to the difficulty of developing coherent competition law policy with respect to cartel conduct. Collaboration may be desirable (indeed, necessary) or harmful depending on the context. In a small market economy such as New Zealand that relies on collaboration, there is a serious and very real risk that enforcement of cartel prohibitions will result in over‑deterrence that impacts consumers and businesses negatively for no apparent benefit.

This is the market and economic context in which it is proposed that criminalisation of cartel conduct be introduced. It is important to take account of this market and economic context because empirical studies of the effectiveness of cartel criminalisation are predominantly US based, where the national economy is significantly larger and the risks of over‑deterrence may be small to negligible. We cannot expect those studies to have direct relevance to the policy debate in New Zealand. We must assess the available evidence for ourselves.

Impact of criminalisation

Against this market and economic context, the key policy question is whether the distinction between efficiency‑enhancing conduct and harmful cartel conduct is sufficient clear and appropriately delineated so that criminalisation does not inadvertently or unnecessarily inhibit desirable collaboration.

The apparent policy justification for the introduction of criminalisation is the potential for increased deterrence. There is an intuitive appeal here that can be difficult to displace. As an economic offence, fines and damages may simply be treated as the cost of doing business and not operate to deter undesirable conduct, especially where that conduct is clandestine and the risk of discovery may be low. From that perspective, increasing compliance incentives through the threat of imprisonment would appear to have real benefits.

However, it is important to recognise that those benefits occur only because the sanctions for successful prosecution are deliberately being set out of all proportion to the offending. This can be optimal from an economic standpoint because of the low chance of detection. However, given the risks of over‑deterrence the benefits need to be corresponding large and sufficiently likely to justify the introduction of criminalisation.

Member of the Committee should carefully question whether those benefits are indeed sufficiently large and likely in a New Zealand context. There does not appear to be any robust policy analysis to indicate that this is case, and there is good anecdotal evidence to suggest that they may not be. For example, the competition law and policy community understands that the vast majority of cartel conduct cases are likely to involve:

  • large international cartels that touch on the New Zealand market only incidentally, in which case the deterrent effect is driven by the sanctions imposed in other jurisdictions; or

  • small New Zealand businesses who simply do not understand that their conduct risks being construed as breaching competition law standards, in which case no deterrent effect is likely.

An example may assist with understanding the second category of conduct. Imagine a scenario where a small manufacturer wishes to develop new market opportunities, and contracts with a distributor for that purpose. Both manufacturer and distributor participate in the retail market, and so are direct horizontal competitors. In order to protect its current avenues to market, the manufacturer insists on delineating responsibility for separate geographic markets and/or a retail price floor to ensure the retail value of the manufacturer’s product is not undermined.

Variations on this type of scenario are extremely common in New Zealand, and in a professional capacity I have been asked to advise on them regularly. The scenario is clearly an example of cartel conduct in breach of the Commerce Act because it contains both price fixing and market allocation elements. However, this is not often understood by the participants involved, who would usually not even turn their mind to the issue unless they had not sought competent legal advice. The deterrent effect in this scenario is limited if it exists at all.

These scenarios suggest that in New Zealand’s small market context the benefits of criminalisation are unclear outside of a narrow set of circumstances. Those circumstances are deliberate, exploitative market manipulation by ostensible competitors in the full awareness that their actions cheat consumers and harm economic markets.

The Commerce Act as currently drafted does not adequately distinguish in the abstract between beneficial and harmful horizontal conduct. As a result, the proposal to introduce criminalisation is not well targeted to an appropriate set of circumstances. Under the Commerce Act, the work for differentiating cartel conduct from beneficial collaboration is done by the ‘collaborative activities’ exemption, mentioned above. However, the collaborative activities exemption is extreme opaque and difficult even for competition law experts to apply unambiguously. The relevant legislative provisions are new, having been enacted in 2017, and they change the law significantly. The expectation appears to be that the courts will develop precedent on the proper application of the exemptions over time, creating a greater level of certainty for businesses and investors.

It is difficult to see case law precedent being developed in this way in a timely or effective manner. The small size of the New Zealand economy is reflected in both the enforcement and judicial dimensions of competition law enforcement. Developing reliable precedent through case law takes considerable time, and can be extreme controversial. For example, the provisions of the Commerce Act relating to single firm abuse of market power went through a 15‑year period of development by appellate courts and have remained so controversial that the Commerce Commission’s current policy is simply to avoid using the provisions. There is a risk that marginal cases, where the costs and benefits are significant but the balance between them is unclear, will result in a similar outcome in respect of cartel conduct and the exemptions that apply.

Criminalising conduct where the potential defences are unclear and likely to continue to be so is contrary to basic principles of justice and fairness. Criminal law sanctions categorically should not be applied in such an ambiguous context.

As a result, the best approach for introducing criminal sanctions for cartel conduct may be to target the most egregious conduct with specific offences dealing exclusively with situations where the benefits of potential collaboration are negligible and the need for specific defences eliminated. This type of approach has been successfully pursued in other jurisdictions. For example, Germany has opted for a highly successful variant of criminalisation focussed solely on bid-rigging, while other types of horizontal conduct are subject to regulatory prosecutions (such as fines) only. A similar approach could be adopted in New Zealand to secure the benefits of criminalisation (high deterrence for most serious and obvious cases, publicity for successful prosecutions) without the undue risk of deterring efficiency enhancing conduct.

If that type of approach was adopted, my recommendation is that:

  • big‑rigging be defined as a separate, specific offence under the Commerce Act, in line with the German approach; and

  • cartel conduct that involves co‑ordination among competitors in a deliberately deceptive manner is also separately defined as a specific offence.

The focus on deliberate deception is key to the second specific offence. It incorporates a mens rea element that is essential for the credibility of the offence from a criminal law perspective, and it delineates conduct for which there is no clear economic justification. The New Zealand economy does not benefit in any way from deception.

I consider that this model better balances the risks and benefits of cartel criminalisation than the model currently proposed by the Bill. For clarity, under the model I am proposing all other horizontal conduct would treated as a potential regulatory offence only, and would not be subject to criminal sanctions. If the more limited model I propose proves successful over time, then incremental extension of criminalisation to other horizontal conduct can be considered on the merits of that success.

Institutional issues

The move to criminal prosecution of cartel offences raises issues of institutional capacity and capability that need to be considered carefully. The two key considerations are:

  • the institutional separation of investigatory and prosecutorial functions; and

  • the ability and willingness of the courts to impose imprisonment on the basis of economic harm established by economic evidence.

Narrowing the scope of criminalisation away from all horizontal conduct to bid‑rigging and conduct involving deliberate deception is the best approach to address these institutional issues.

Criminalisation will result in separation between the investigatory and prosecutorial functions of cartel enforcement, with the Commerce Commission having responsibility for the former and the Crown prosecutor having responsibility for the latter. Once the Commission refers a cartel matter to the Crown prosecutor for prosecution, the Commission is no longer in control of how that prosecution will succeed. While the Commission will have a role in any prosecution, that role will be as a witness for the prosecution rather than as an instructing client. The Commission will not be able to directly influence enforcement proceedings in a way that accords with its own institutional interests or so as to promote maximum deterrence in respect of cartel conduct over time.

This reduction in autonomy for the Commission risks impacting on the ability of New Zealand competition law to secure greater compliance through stronger enforcement action in the following ways:

  • ambiguous lines of accountability, with greater scope for conflict and disagreement between the Commission (whose concern ought to be the efficiency and effectiveness of the regime overall) and the Crown prosecutor (who is likely to be more focussed on the result of an individual prosecution);

  • the credibility of leniency programme, which is managed by the Commission as an investigatory measure but which relies on the exercise of prosecutorial discretion for its operation;

  • fewer settlements as a result of (1) the all‑or‑nothing nature of prosecution, which leaves defendants little incentive to consider settlement; (2) the lack of incentives for the Crown prosecutor to secure settlements, because they are not required to consider the efficiency and effectiveness of the regime as a whole; and (3) the lack of clear sentencing guidelines and experience in the practice of plea bargaining; and

  • risk of investigatory error on the part of the Commission given the higher evidentiary standards of criminal prosecutions, and a resulting reluctance on the part of the Crown prosecutor to take actions that they view as unlikely to be successful.

These issues will impact directly on the deterrent effect of the criminalisation regime, undoing the very incentives that the threat of imprisonment is intended to impose.

Even in cases where there is a high degree of operational and philosophical alignment between the Commerce Commission and the Crown prosecutor, the courts may be reluctant or feel unable to impose a criminal conviction in all but the most egregious circumstances. The New Zealand courts have a history of apparent discomfort with economic regulation, often ‘reading down’ the scope of legislative prohibitions to avoid engaging with economic evidence and argument. This discomfort may be the result of a perception that the courts cannot or should not engage with such evidence for institutional or philosophical reasons, but whatever the reason it places New Zealand in a different position to other jurisdictions where cartel criminalisation has been implemented and the courts have been willing to engage in the active development of competition law.

In the cartel context, we already see a clear divergence between the Commission and Crown prosecutor on the one hand, and the Courts on the other, where defendants elect to contest enforcement proceedings. Take, for example, the two most recent contested hearings dealing with cartel conduct:

  • In Commerce Commission v Siemens AG the High Court found (contrary to the argument of the Commission and the Crown prosecutor) that there was no cartel conduct in respect of an international cartel that had attempted but failed to enter the New Zealand market.

  • In Lodge Real Estate Limited v Commerce Commission the High Court found (contrary to the argument of the Commission and the Crown prosecutor) that an agreement among competitors did not impact on pricing decisions, and so fell outside the definition of cartel conduct in the Commerce Act.

Neither of these cases were consider controversial when enforcement action was taken. I do not suggest that either of these cases was correctly (or for that matter incorrectly) decided. I raise them here simply to demonstrate that it can be difficult to secure an enforcement‑favourable judgment even in circumstances where the enforcement agencies consider that the case is clear cut.

This potential for divergence between enforcement agencies and the courts is only likely to grow if cartel conduct is criminalised. The risk is that a judiciary reluctant to convict economic crimes because of discomfort with economic analysis may hesitate before convicting. Relatively harsh criminal penalties such as the maximum 7 years’ imprisonment may cause judges to baulk, undermining the deterrence‑based rationale of the Bill.

These institutional issues will be far less acute where criminalisation is only imposed in the most obvious and egregious circumstances. This reduces the scope for philosophical and operational disagreement, and places greater reliance on legal evidence rather than economic analysis (because the economic harm is patent). For these reasons, focusing criminalisation on bid‑rigging and deliberately deceptive market co‑ordination is likely to improve the deterrent effect of the competition law regime.

International co‑operation

One final point that the Members of the Committee ought to consider is the impact of criminalisation on incentives for overseas enforcement bodies to share information with the Commerce Commission. In jurisdictions where the enforcement culture does not favour criminal sanctions, there may be a new reluctance to share information with the Commission if that could result in a criminal prosecution, or even simply if that information could be passed to a separate entity with prosecutorial discretion (the Crown prosecutor).

This is a real concern given that many of the large cartels that affect New Zealand markets are likely to be international in scope, and only become known in New Zealand because of international networks among enforcement agencies. Again, a narrow focus on criminalising only the most egregious conduct reduces the risk appreciably.


On the basis of the analysis set out in this submission, I recommend that the Commerce (Criminalisation of Cartels) Amendment Bill be amended so that:

  • big‑rigging is defined as a separate offence under the Commerce Act, which is subject to a criminal sanction of a maximum 7 years’ imprisonment;

  • cartel conduct that involves co‑ordination among competitors in a deliberately deceptive manner is defined as a separate offence under the Commerce Act, which is subject to a criminal sanction of a maximum 7 years’ imprisonment; and

  • all other horizontal conduct is treated as a regulatory offence only, and is not subject to criminal sanctions.

#CommerceAct #competitionlaw #cartels #LawReform

Recent posts
Tag search
No tags yet.