Association of Mineworkers and Construction Union and Others v Anglo Gold Ashanti Limited t/a Anglo Gold Ashanti and Others (CCT 233/20) [2021] ZACC 42; [2022] 2 BLLR 115 (CC); (2022) 43 ILJ 291 (CC) (12 November 2021)
Section 66(2)(c) of the LRA imports the principle of proportionality in assessing the reasonableness and the substantive lawfulness of secondary strikes. It seeks to balance the impact of secondary strikes on secondary employers on the one hand, with their effect on the business of the primary employer on the other hand.
Facts:
In February 2019, AMCU issued notices of secondary strike action on ten mining companies in support of a strike by the union's members at Sibanye Stillwater over wages and other conditions of employment. The secondary strike was to commence on 28 February 2019 and was to last for seven days.
Following the notices, all these companies launched urgent applications to interdict the union from initiating strike action at their operations and to declare the proposed secondary strike to be unprotected. The separate applications were dealt with together by the Labour Court. The court accepted that the procedural requirements in s 66(2)(a) and (b) of the LRA had been fulfilled, and that the application to declare the strike unprotected therefore turned only on the requirement of 'reasonableness' in s 66(2)(c) which provides as follows:
"No person may take part in a secondary strike unless the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer."The LC said the test for reasonableness was ultimately a proportionality assessment to determine whether the harm caused by the secondary strike to the secondary employer was proportional to its impact on the business of the primary employer. The LC held that the harm was not proportional and declared that the secondary strike at all the companies was unprotected.
On appeal the Labour Appeal Court found that the interpretation of s 66(2)(c) was already settled law, and also dismissed the appeal on the grounds that the matter was moot, the primary strike having already been settled at that stage.
The union then applied for leave to appeal to the Constitutional Court. The majority Court found that, although the matter was moot, it was nevertheless in the interests of justice for leave to appeal to be granted. A decision by the Constitutional Court would clarify the approach to secondary strikes for the labour relations community.
The union challenged the LC's interpretation of s 66(2)(c), which imputed a proportionality assessment that took into account the harm suffered by secondary employers. It argued that the effect of the secondary strike on the business of the primary employer should be prioritised, rather than harm to the secondary employer.
The ConCourt interpreted s 66(2)(c), highlighting at least six substantive requirements that regulate secondary strikes. One of these requirements is that, in relation to the primary employer, a secondary strike must have an effect. Also, in relation to the secondary employer, the secondary strike must be reasonable. The phrase 'reasonable in relation to' in s 66(2)(c) imports proportionality in assessing reasonableness. Because secondary employers do not have the same procedural safeguards that primary employers have, such as conciliation and more than seven days' notice of the intended strike, proportionality and reasonableness are shields to safeguard secondary employers. They are needed to preserve the equilibrium that s 66(2)(c) seeks to establish.
The ConCourt accordingly concluded that s 66(2)(c) does import the principle of proportionality in assessing the reasonableness and the substantive lawfulness of secondary strikes. Section 66(2)(c) seeks to balance the impact of secondary strikes on secondary employers on the one hand, with their effect on the business of the primary employer on the other hand.
Turning to the facts of this case, the ConCourt considered whether the secondary strikes in this matter were proportionate. It found that the secondary strikes were unreasonable, firstly, because they had no effect on the business of the primary employer (Sibanye), and secondly, because the impact on the secondary employers was unreasonably destructive.
The ConCourt dismissed AMCU's appeal against the Labour Court and the Labour Appeal Court judgments but reversed the costs orders imposed by those courts, ordering each party to pay its own costs.
Extract from the judgment:
Pillay AJ:
The issues
[31] Essentially, this appeal turns on the substantive requirements for lawful secondary strikes in terms of section 66(2)(c). Specifically, the question raised is whether section 66(2)(c) imports the principle of proportionality in assessing the lawfulness of secondary strikes. More specifically, the question is whether section 66(2)(c) factors in a balancing of the impact of secondary strikes on secondary employers, on the one hand, with their effect on the primary employer on the other hand. Thus, if secondary strikes impact disproportionately harshly on secondary employers, as a party uninvolved in the primary strike, would secondary employers be entitled to interdict the secondary strikes under section 66(3)?
[32] AMCU raises two subsidiary issues relating to the form of secondary strikes. First, in assessing the reasonableness of multi-employer secondary strikes, should the impact on secondary employers be considered individually or collectively? Second, should interdicts against violence in the past and violence during the primary strike be taken as factors informing the assessment of compliance with section 66(2)(c)?
[33] As the matter is moot, the brief summary of the facts and arguments above are intended to set the scene for what is exclusively an exercise in interpreting sections 66(2)(c) and (3). Any references to the facts will be purely to contextualise legal principles. For this interpretative exercise, my starting point is the Constitution, which in turn invites me to look to international law. Together, they underpin the interpretation.
Section 23 of the Constitution
[34] The analysis of section 66(2)(c) begins with an appreciation that the LRA seeks to give effect to section 23 of the Constitution. Contextualising the interpretation within the constitutional framework is necessary to acknowledge the elevated status that collective bargaining and the right to strike enjoy under our current democratic dispensation. The right to strike is constitutionalised because it is a recognised human rights response to slavery, servitude and forced labour.
............
[67] The procedural requirements for a strike are infinitely more onerous than for a secondary strike. A strike must be preceded by conciliation, deadlock and notice to the employer. All this takes time before a strike can commence. Four months lapsed between AMCU commencing negotiations and the primary strike. In the case of secondary strikes, no conciliation, no deadlock, and no dialogue are prescribed as prerequisites. Once the primary strike is lawful, all that remains to be done is for the trade union to give seven days' notice to the secondary employer of its employees' intention to embark on a secondary strike. Cumulatively, the absence of prior engagement, the brevity of the notice and the fact that the secondary employer and its employees have no interest in the outcome of the primary strike, distinguish the secondary employer from the primary employer. Secondary employers having employees who belong to other trade unions that are not engaged in the secondary strike, adds another dynamic to a situation already complicated by multi-dimensional power-play.
[68] Secondary employers, who have no direct or indirect effect on collective bargaining at the primary employer, have no control whatsoever over the process. Without section 66(3), the scheme of the LRA regarding secondary strikes would leave the secondary employer with little leverage to safeguard its interests. Section 66(3) is a shield in the hands of secondary employers who would otherwise have no means of protecting their enterprises. What it is not, is a weapon to annihilate the constitutional right to strike. Thus, when invoking section 66(3), ways of limiting the secondary strike should be investigated first before its outright prohibition is sought. In Samancor, the Labour Court gave its ruling on what would be proportional before anticipating a settlement agreement limiting the secondary strikes, which it made an order of that Court.
[69] Section 66(2)(c) seeks to attain a balance between different models of regulating secondary strikes recognised by the ILO. To section 66(2)(c) I turn.
Section 66(2)(c) of the LRA
[70] SALGA I and SALGA II are currently authoritative decisions which maintain that section 66(2)(c) invites a proportionality assessment when considering whether a secondary strike can be said to be reasonable. Proportionality implies that a secondary strike that impacts more harshly on the secondary employer than on the primary employer would be disproportionate, and therefore, unreasonable and unlawful. Furthermore, if the harm to the secondary employer is so immense, then that fact alone should spare the secondary employer from enduring a secondary strike, even if the secondary strike has an effect on the primary employer. This case seeks to confirm or refute that interpretation of section 66(2)(c), which the Labour Court upheld.
[71] Section 66(2)(a)-(c) reads:
-
"No person may take part in a secondary strike unless-
- the strike to be supported complies with the provisions of sections 64 and 65;
- the employer of the employees taking part in the secondary strike or, where appropriate, the employers' organisation of which that employer is a member, has received written notice of the proposed secondary strike at least seven days prior to its commencement; and
- the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer."
[73] Section 66(2)(a) prescribes the procedural prerequisite that the primary strike must be lawful. This complies with the Committee of Experts' stipulation of "the sole requisite that the original strike is lawful". Typical of international law, that is a minimum standard to maximise assent from member states. The lawfulness of the primary strike is not the only requirement for a lawful secondary strike. Member states are free to regulate secondary strikes according to their national laws. Research has not revealed a single member state that imposes no other limitations for a secondary strike other than the lawfulness of the primary strike.
[74] Embedded in section 66(2)(c) are at least six substantive requirements that regulate secondary strikes. These requirements operate as internal or built-in limitations on the right to participate in a secondary strike.
[75] First, the critical requirement is for a secondary strike to have "an effect". Having an effect on the primary employer is the baseline threshold requirement for a secondary strike. A secondary strike that can have no effect, that is entirely altruistic and amounts to "mindless exercises of worker solidarity for the sake of nothing but worker solidarity", is impermissible. Such solidarity strikes fall on the opposite end of absolute bans on the secondary strike spectrum and are not a proportional exercise of the right to strike. Nor is an absolute ban a reasonable and justifiable limitation on the right to strike. Neither represent any balance whatsoever of the right to strike. Somewhere between these extremes section 66(2)(c) must find its place.
[76] Second, the words "possible" and "indirect", used in section 66(2)(c) in relation to "effect", set the bar low and wide for a lawful secondary strike. The word "possible" lowers the threshold to mean something less than probable - not actual - but more than notional. What is required is that the secondary strike must be capable of having an effect. It is not necessary to predict with certainty that it necessarily would, but merely that it could, have an influence on the business of the primary employer. The word "indirect" expands the scope of the possible effect, thus shifting the balance to favour employee solidarity. However, that is about as far as the section inclines towards the solidarity side of the spectrum. The remaining words and phrases rebalance section 66(2)(c).
[77] Third, the effect of the secondary strike must be "on the business" of the primary employer. Affecting the business of the primary employer anticipates a nexus between the business of the primary employer and the secondary strike, and by extension, the secondary employer and its employees. Impliedly, the nexus must be such that the secondary strike is reasonably capable of influencing the business of the primary employer. Such influence would be present, for instance, if there pre-exists a substantial commercial connection between both businesses. A supplier-customer relationship would establish a connection if the level of mutual interdependence is significant. This principle applied in >Sealy, Samancor, Billiton, and Clidet. In SALGA I, the Labour Court found that a secondary strike at municipalities could have an effect on government at higher levels. Instructive is the Labour Appeal Court's approach to unravelling the interconnectedness of enterprises in Samancor and Billiton II.
[78] If a relationship in which the secondary employer is able to influence market sentiment and ratings agencies that, in turn, impact adversely on the commodity price, the share price and the valuations of the primary employer, that would be another example. But the links must be proven. AMCU's attempt at leading expert evidence in the Labour Court to prove this effect failed because the witness did not qualify as an expert. Furthermore, on the facts, the Labour Court found that so many variables influence share prices - ranging from "macro geopolitical issues to the appointment of a new chief executive" - that it could not pin down any fall off to the intended secondary strike alone. The evidence was lacking in that instance. Another consideration when assessing the impact of strikes is this: a drop in production, and concomitantly, labour costs induced by the "no work, no pay principle", could increase demand and thereby spike the price of the commodity or goods - paradoxically, a most desirable consequence of a strike for the industry. [79] The complexity of relationships between entities in the private and public sectors militates against pinning down hard and fast "one size fits all" principles about what is proportional and reasonable. Furthermore, establishing cause and effect of secondary strikes would be challenging at the best of times. More so in motion proceedings, especially when they are brought urgently, as secondary strike interdicts usually are. This difficulty presented in Billiton II when the Labour Appeal Court had to have recourse to Plascon-Evansto assess the facts to which to apply section 66(2)(c). What the cases cited above show is that it is difficult, but not impossible, to prove the effect, if any, between a secondary strike and the business of the primary employer. In cases in which the courts found an effect, they refused to interdict the secondary strikes. The effect does not have to be equal "upon the businesses of the secondary and primary employers" as suggested in the second judgment but reasonable, as I elaborate below.
[80] Fourth, the assessment of the effect is, unsurprisingly, directed at "the primary employer". After all, a secondary strike is by definition "a strike, or conduct in contemplation or furtherance of a strike, that is in support of a strike by other employees against their employer".
[81] Fifth, the reference to the effect on the primary employer does not imply that the secondary employer is excluded altogether from the assessment. The impact on the secondary employer comes into the analysis in interpreting whether "the nature and extent of the secondary strike is reasonable in relation to the effect that the secondary strike may have".
[82] Adding the words "and the secondary employer" at the end of section 66(2)(c) would have fortified the interpretation that reasonableness also applies to the secondary employer. However, elevating the primary and secondary employers to the same level would have undermined the definition of secondary strikes which targets the primary employer. Importantly, it would have decisively and unambiguously shifted the balance favourably towards the secondary employer. This shift was unlikely to enchant the trade union parties negotiating the draft Bill. Omitting the words "and the secondary employer" may be explained as probable resistance from the trade union negotiators to their inclusion.
[83] Assessing the reasonableness of the secondary strike on the secondary employer is not disloyalty to the text by subversively reading in the words "and the secondary employer" at the end of section 66(2)(c). Quite the contrary. The absence of these words informs the way in which reasonableness of the secondary strike on the secondary employer will be assessed. That is, not based on parity with the primary employer. This approach contextualises the interpretation of the LRA in its genesis as a negotiated draft Bill.
[84] Irrespective of the reasons, omitting the words "and the secondary employer" at the end of section 66(2)(c) distinguishes the primary and secondary employers. In relation to the primary employer, the secondary strike must have an effect. In relation to the secondary employer, the secondary strike must be proportional, even if it has an effect on the primary employer. Importantly, the omission imbues section 66(2)(c) with a degree of flexibility that invites judicial intervention to determine reasonableness case by case.
[85] Reasonableness is not an abstraction. Instead, it harks back to its roots in the Constitution, to give effect to section 36 specifically. To be lawful, the secondary strike must be reasonably capable of having the possible effect. This begins, not with a value judgement, but with a sound enquiry into the facts. In the circumstances of Samancor, SALGA I and SALGA II, the courts decided, on the facts, that the secondary strike could influence the business of the primary employer.
[86] Furthermore, "reasonable" also qualifies "the nature and extent" of the secondary strike. This does involve a value judgement. Conceptually, the phrase "the nature and extent" in section 66(2)(c) is cast widely so that it includes secondary strikes in all sectors and industries, private and public, and may take many forms. Factors to determine the nature and extent of the secondary strike would include the duration and form of the strike, that is, whether it is a go-slow, intermittent or stay?away strike for a day or longer; the number of employees involved; their membership of trade unions; their conduct, including whether it is peaceful or violent; and the sector(s) in which the primary and secondary strikes occur, that is, whether it is in the private or public sector, or the retail, agricultural, manufacturing, or mining sectors, and so on.
[87] It follows that if a secondary strike is incapable of having any effect whatsoever on the business of a primary employer, that would not be reasonable. That would be the end of the enquiry. For, the most basic requirement to qualify for a lawful secondary strike will not have been met. Proving that the nature and extent of a secondary strike could possibly have an effect on the business of the primary employer requires evidence. Such evidence will be harder to establish if the link between the primary employer, and its workers, on the one hand, and the secondary employer, and its workers, on the other hand, is weak. After establishing that the secondary strike has an effect on the business of the primary employer, assessing the nature and effect of the secondary strike calls for a value judgement.
[88] The phrase "in relation to" in section 66(2)(c) unambiguously anticipates a comparison. The comparators are firstly, the secondary strike and business of the primary employer, and secondly, the secondary strike and business of the secondary employer. The standard prescribed for comparison is reasonableness. Thus, the phrase "reasonable in relation to" imports proportionality in assessing reasonableness. Conceptually, proportionality and reasonableness often converge but not necessarily. In balancing the right to strike with the rights of primary and secondary employers, all factors must be considered to determine, not just proportionality, but ultimately, reasonableness. For reasonableness is the standard or attribute to which the LRA aspires for secondary strikes because the right to strike is constitutionally entrenched. Below, an example will illustrate an instance when proportionality and reasonableness can be in tension with each other and the right to strike.
[89] Recognising proportionality and reasonableness of the secondary strike anticipates some safeguards for secondary employers. Otherwise, they would have no means of protecting their businesses. Their employees, who are non-unionised or who belong to trade unions other than the one calling for the secondary strike, may also want protection. Proportionality limits the right to participate in a secondary strike and introduces safeguards to rebalance the rights of secondary employers more favourably relative to the primary employer.
[90] Assessing the impact of the secondary strike on the business of the primary employer is a question of fact determined case by case. Assuming, as the second judgment does, that secondary strikes "may have some impact but to a lesser degree" than the primary strike, that will not be true in every instance. Much would depend, for instance, on the relationship between the primary and secondary employers, the number of employees involved in the strikes and whether the businesses of the primary and secondary employers are interrupted partially or totally. A partial shutdown of the business of a primary employer could be more greatly impacted by a total shutdown of the secondary employer who supplies components to the primary employer. On its own, this fact would discourage interference in the power-play by interdicting the secondary strike, unless after considering all the factors, some of which are mentioned above, the impact of the secondary strike on the secondary employer is not only disproportionate but also unreasonable. In making this assessment, the constitutionalising of the right to strike and to bargain collectively as the primary means of eliminating poverty and inequality must weigh in.
[91] Proportionality and reasonableness enquiries invite judicial supervision of the interpretation and application of the law to the circumstances. Academic opinion favours "an approach that requires the harm caused to the secondary employer to be proportional to its likely impact on the business of the primary employer". In the opinion of Seady and Thompson, the LRA leaves it "to the Labour Court to define the parameters of protected secondary strikes within the meaning of section 66(2)(c)". Cooper concludes that in "the final analysis however, as in the foreign jurisdictions described above, the Labour Court will play a critical role in clearly defining the nature of this balance".
[92] The stance adopted in the second judgment of disavowing a proportionality assessment is excessively inclined towards protecting altruistic sympathy strikes. Trade unions and their members would have considerably more leverage. It would topple the equilibrium that section 66(2)(c) seeks to establish.
[93] Proportionality assessments to determine reasonableness are necessary for reasons that relate to the parties engaged in secondary strikes, and beyond. Persons other than the employers and employees involved in the secondary strikes may be affected if their rights are violated. Jettisoning a proportionality assessment would exclude access to courts to invite judicial scrutiny when rights under the LRA, the Constitution and other laws could be implicated.
[94] In the administrative law context, ouster clauses were notorious for concealing rights violations in our pre-democratic order. In the context of labour law, ousting the courts from assessing proportionality would distort the dispute system design of the LRA. The LRA anticipates and prescribes peaceful and appropriate forms of dispute resolution for almost every conceivable labour dispute. Allowing rights violations to burgeon without judicial scrutiny of secondary strikes would be out of kilter with that design and unwholesome for a constitutional democracy. And denying access to courts to have the proportionality of secondary strikes assessed could have dire consequences if those whose rights are violated have to resort to self-help to protect their interests. Effective management of conflict would be sacrificed.
.................
[109] This matter raises a discrete question of law and two subsidiary issues. Primarily, does section 66(2)(c) import the principle of proportionality in assessing the reasonableness and the substantive lawfulness of secondary strikes? Our response is "yes". Section 66(2)(c) seeks to balance the impact of secondary strikes on secondary employers, on the one hand, with their effect on the business of the primary employer on the other hand. Thus, on the one extreme, if secondary strikes have no effect on the primary employer, or, on the other extreme, if the effect is disproportionately harsh on secondary employers, they would be entitled to interdict the strike under section 66(3). This principle of proportionality derives not only from the reasonableness requirement to be found in section 66(2)(c), but also from the Constitution and international law.