“Characterisation is the biggest problem of our time” according to the Competition Commission’s (Commission) Cartels Division Manager’s statement at the 16th Annual Competition Law, Economic & Policy Conference (Conference).  A statement that was followed by a definitive “we need to get rid of it”.

These statements, reflecting strong conviction, are not surprising given the apparent divide between how the Commission (tasked with investigating and prosecuting contraventions of the Competition Act), on the one hand, and the Competition Tribunal (Tribunal) and Competition Appeal Court (CAC) (the adjudicative bodies), on the other hand – approach “characterisation”, specifically when it comes to written commercial agreements between competitors. The Commission’s Chief Legal Counsel at the Conference made reference to “old generation” cartel cases which show an “accommodative stance” to the Commission, being the cartel cases the Commission was successful in prosecuting.  These were distinguished from the “new generation” cartel cases, which are “characterised by a clear shift in the pendulum”, being the cartel cases the Commission has been unsuccessful in prosecuting. 

The key difference between these cases being that the Tribunal and the CAC are increasingly dismissing the Commission’s argument that a consideration of an agreement on its face suffices to determine whether there has been a contravention of the Competition Act.  Using this approach, by way of example, the Commission would argue that an agreement which incorporates a provision prohibiting a competitor from engaging in certain activities divide the markets in contravention of the Competition Act, based on a reading of the plain text with no further evidence being required.  This approach has not met the favour of the adjudicative bodies who favour a broader approach to interpretation of written agreements, taking into account the context and purpose.

It is these divergent approaches which lead to the Commission’s Chief Legal Counsel, stating that “cartel law is now at an inflection point”.  But is it?  A closer look at characterisation and its function would indicate that characterisation is simply a tool of interpretation useful for understanding written agreements in alignment with the interpretation tools used by South African courts.  In fact, its application is warranted as there is no reason why agreements assessed under the Competition Act must be interpreted (or characterised) in any manner different to the approach of the superior courts, and as set finally by the Constitutional Court.

Characterisation at a glance

Taking a step back, it is important to look at what characterisation is, in context, and what it entails.

First, some context.  In determining whether an agreement between competitors has contravened section 4(1)(b) of the Competition Act, the so-called “hard-core cartel” conduct (price fixing, market division or collusive tendering / bid rigging) the competition authorities must establish whether the agreement falls squarely within the definition of the prohibited conduct.  It is particularly important to get it right as agreements contravening section 4(1)(b) can never be justified with reference to pro-competitive outcomes or any other positive impact, including increased efficiencies. (This is distinguishable from other anti-competitive agreements prohibited by section 4(1)(a), whose anti-competitive would be weighed against technological, efficiency or pro-competitive gains resulting from that agreement).

Since the proof of a section 4(1)(b) offence carries significant consequences (including potential criminal prosecution for individual directors), the exercise of establishing whether an agreement contravenes section 4(1)(b) is vital to ensure that there are no miscarriages of justice.  Where contravening conduct is concerned, such conduct is always established with reference to the evidence on context and purpose.  However, there appears to be resistance on the part of the Commission in adopting the same approach when it comes to written agreements between competitors that could be seen as hard-core collusion.

Turning to what characterisation is – it is often said that characterisation was “introduced” into South African law by the Supreme Court of Appeal (SCA) in the ANSAC case.  In the ANSAC case, the SCA, while referring to an American doctrine, held that: the first stage of characterisation is a matter of interpreting the statute, which is then followed by a factual inquiry.  This has since been crystallised in subsequent decisions, with the CAC most recently in the I&J decision stating that “characterising” an agreement simply means looking at its context and character / nature to determine whether the agreement is prohibited.  At its core, therefore, characterisation is simply the exercise of interpreting an agreement to identify whether it is prohibited.  On this basis, it is not entirely accurate to say the SCA introduced a new interpretive tool for agreements into South African law, especially since the tools of interpretation used are based on the common law developed over time.  Characterisation is, simply put, a different term for the process of interpretation of agreements.  When one looks at what it is in context, it is simply interpretation of the language of an agreement with reference to the context and purpose.

The “new generation” cartel cases

Using the Commission’s parlance, the “new generation” cases are properly represented by the Tribunal and the CAC’s decisions in the case involving a manufacturing agreement between Irvin & Johnson Limited (I&J) and Karan Beef Pty Limited (Karan).  The Commission alleged that the manufacturing agreement contravened s 4(1)(b)(ii), the prohibition on market division / allocation.  In short, the manufacturing agreement included a non-compete clause stating that Karan would not “manufacture, market or produce products that are the same or similar to the “contract products” or any other processed beef products that it manufactures for I&J”.  The Commission argued that by concluding the manufacturing agreement, I&J and Karan had agreed to not compete and that Karan would cease to produce processed frozen beef products for its own account and would instead utilise its capacity to produce products for I&J.

The Commission’s argument was that textually, once a feature of the prohibition is found in the agreement, the operation of section 4(1)(b) is triggered, and no further enquiry is needed.  As such, on a plain reading of the agreement, the manufacturing agreement contravenes the Competition Act, with “no need to characterise the agreement”.  The Commission also argued, before the CAC, that since there is no reference to the word “characterisation” in section 4(1)(b), the Tribunal (which had ruled against it) should not have embarked on a characterisation exercise.

I&J argued that the manufacturing agreement was simply a good faith commercial agreement and that in any event, Karan had taken a unilateral decision to refrain from the production of processed beef products for third parties even prior to the manufacturing agreement.  On the basis that Karan already sought to exit the market for the production and supply of beef products, the manufacturing agreement could not, upon properly characterising it, “which means looking to its context and nature”, be interpreted to mean that the market was collusively divided.

The central issue for determination by the CAC was therefore whether the Commission had proved the section 4(1)(b)(ii) contravention based solely on the face of the agreement or in the words of the CAC “whether it is possible to look beyond the face of the manufacturing agreement and consider its character, circumstances and nature to ascertain whether [it] amounted to allocating customers and dividing up the markets?”

The CAC held that:

“it is evident from the wide ambit given to interpretations of what constitutes an agreement, that a purely textual, or face-value, analysis is not the proper approach. A facial analysis only continues to be applicable in cases of hardcore cartelism where the wording of the agreement is so clear as to bring it unambiguously, within the ambit of the s 4(1)(b) prohibitions. Ultimately, agreements are conceived within a specific context and it would be relevant to take that context into account when applying the presumption of illegality test where the situation is not one of hardcore and unambiguous cartelism.”

This explanation clearly shows that what is referred to as “characterisation” in the competition law context, is nothing more than the utilisation of principles used in contractual interpretation by other courts.

The Constitutional approach

Recently, in the UJ case the Constitutional Court was asked the crisp question “Is a court of law entitled or required to take cognisance of context when interpreting a contract”.  (Notice how this question mirrors the question on characterisation used in the I&J case).  The Constitutional Court held that, based on Endumeni:

“…Whatever the nature of the document, consideration must be given to the language used in the light of the ordinary rules of grammar and syntax; the context in which the provision appears; the apparent purpose to which it is directed and the material known to those responsible for its production….The ‘inevitable point of departure is the language of the provision itself’, read in context and having regard to the purpose of the provision and the background to the preparation and production of the document“

This approach to interpretation requires that in considering a contract “from the outset one considers the context and the language together, with neither predominating over the other”.  In addition, according to the Constitutional Court, context must be taken into account “whether or not the words used in the contract are ambiguous.”  This requirement for a balancing act between the text, context and purpose, means that the Commission’s evidence is incomplete in cases where it simply relies on the text of the agreement to prove a contravention of the Competition Act.  The proper interpretation of an agreement would require a consideration of the text, as well as the evidence of the context, e.g., evidence of pre-contractual exchanges, and purpose, e.g., the protection of a sold business.

The UJ case goes even further, indicating that “where in a given case, reasonable people may disagree on the admissibility of the contextual evidence in question, the unitary approach to contractual interpretation enjoins a court to err on the side of admitting the evidence.”  Justice Khampepe emphatically states “I repeat: context is always relevant to the interpretation of a contract, whatever its terms.”  The implications of this emphasis are clear and should be noted in the execution of the Commission’s duty to prosecute agreements – context should always be established and pleaded.  The interpretation must be argued on a holistic bases, simultaneously considering the text, context and purpose.

This is characterisation under a different name, but essentially the same. “A rose by any other name would smell as sweet”.

Limitations of contextual evidence

The Cartels Division Manager suggested that characterisation may be used as a means of introducing efficiency or other defences in section 4(1)(b) cases, in circumstances where the Competition Act does not allow for any defences for “hard-core cartel” conduct.  This view is not supported by the facts of any cases that the Commission has failed to prosecute successfully.    In his critique of the I&J decision, the Commission’s Cartels Division Manager stated that parties may simply bring evidence saying that “while the agreement reads thus, it is supposed to read as…” (paraphrased).

This cannot be the case as the admission of contextual evidence is intended only to interpret a contract and its purpose.  Its admission should not be understood as the basis for introducing a defence for offences which are otherwise indefensible.    As observed by Acting Appellate Justice Unterhalter in the Capitec Holdings decision (issued shortly after the UJ case), while important, context “is not licence to contend for meanings unmoored in the text and its structure.  Rather, context and purpose may be used to elucidate the text”. The Commission’s argument that contextual evidence introduces impermissible justifications for hard core collusion cannot be sustained if contextual evidence is being used to give meaning to the words within the four corners of the agreement.

Reflection 

As the Commission has pointed out on various occasions, the word “characterisation” does not appear in the Competition Act, but frankly neither does “collusion” nor “hard-core cartel”, yet all are words used and understood in discussions on local competition law issues.  In any event, characterisation does not need to be specifically provided for in the Competition Act since, properly characterised, it is no more than interpretation of the text of an agreement in line with our common law.  Contrary to the fears underlying the Commission’s strong positions apparent from the first paragraph, characterisation does not sanction for amendments to an agreement which contravenes the Competition Act and does not introduce an avenue to provide substantive defences for offending agreements.  Characterisation simply requires that context be taken into account in interpreting the meaning of the text.

Any approach that seeks to understand agreements in context and in line with their purpose should be welcomed by the Commission as it has the potential to strengthen the cases prosecuted and can assure the public that in the execution of its duties, the Commission pursues true offenders / cartelists.   The aim, at the end of the day, is to find and prosecute the “hard-core cartelists” whose agreements harm the public and social welfare, the so-called “thieves at the dinner table” and not to stymie economic activity by misconstruing commercial agreements.

Written by Simba Rodze, Associate & Petra Krusche, Director; Werksmans