The South African Constitution guarantees workers the right to strike. However, lengthy and often violent strikes have long been a cause of concern, due to their disruptive effect on the economy and associated financial losses, not to mention loss of employment and even loss of property and life, if not managed correctly. As a result, legitimate questions arise as to the suitability of strikes as the primary method of dispute resolution.
The most recent 2023 Industrial Action Report, issued by the Department of Employment and Labour in 2024 highlighted an increase of 11.1% for the 2022/2023 period. The number of workplace disputes increased to 97 strike incidents in 2023, from 87 strike incidents in 2022 (https://lrs.org.za, accessed 30-1-2025).
Section 74 of the Labour Relations Act 66 of 1995 (LRA) provides for an alternative method of dispute resolution, namely interest arbitration. However, there is no general access to interest arbitrations. It is available only to employees engaged in essential services. Moreover, where a minimum service agreement (MSA) has been concluded or minimum services have been determined by the essential services committee (ESC), interest arbitrations are accessible only if the majority of essential service workers vote in favour of it.
As it stands, interest arbitrations are rarely used as a dispute resolution mechanism in South Africa, compared to the more prevalent use of strike action. This article will demonstrate while many factors can account for the infrequent use of interest arbitrations, the structure of the LRA may itself be a contributing factor. Not only are interest arbitrations reserved for essential service workers, but a ballot is also required for access thereto. It should, therefore, come as no surprise that interest arbitrations are not always seen as a viable alternative to strike action. It may even be said that, in some respects, the LRA compels essential service workers to strike.
While ‘the right to strike is a constitutional right afforded to all workers in terms of section 23(2)(c) of the Constitution of South Africa,’ the LRA has, since its inception, placed limitations on essential service workers’ right to strike (K Dreyer ‘The designation of the service of consideration and approval of ex gratia ratifications in the medical scheme environment as an essential service’ (www.linkedin.com, accessed 30-1-2025)). ‘In terms of section 65(d)(1) [of the LRA] any person engaged in an essential service may not take part in a strike’ (S Gon ‘The most despised strike of the year’ (https://dailyfriend.co.za, accessed 30-1-2025)). Such employees must refer unresolved mutual interest disputed for interest arbitration in terms of s 74 of the LRA.
However, essential service workers may conclude an MSA, ‘setting out the minimum number of employees performing an essential service who may not strike to ensure that the life, personal safety, or health of the population is not endangered’ (L Louw and S Ngcamu ‘Public sector strike action may put the new Essential Service Committee Regulations of 2022 to the test’ (www.webberwentzel.com, accessed 30-1-2025)). If an MSA is concluded and ratified by the ESC, ‘the agreed minimum service levels become a minimum essential service, but all other employees, who are not required to provide the minimum service, [are] allowed to strike’ (Louw and Ngcamu (op cit)). However, once an MSA was concluded and ratified by the ESC, the provisions of s 74 did not apply. This meant that essential service workers had to make a choice, either conclude an MSA and obtain a limited right to strike or refer disputes for interest arbitration.
The LRA has, therefore, clearly taken an either-or approach to interest arbitrations. Why it has done so, is less clear. One may surmise that the objective was to prevent essential service workers from hedging their bets, using both strike action and interest arbitrations to resolve the same dispute. Unfortunately, this also meant that once an MSA has been concluded and ratified, essential service workers lose access to interest arbitrations, unless the ratified MSA was somehow set aside. However, since the conclusion of MSAs was not, at the time, a mandatory requirement, essential service workers wishing to retain access to interest arbitrations, could get around this obstacle by simply refusing to conclude an MSA.
This position changed on 1 January 2015, when the Labour Relations Amendment Act 6 of 2014 came into effect. A new s 72 was introduced, which empowers the ESC to order parties to conclude an MSA. If the parties fail to do so, the ESC may itself determine the minimum services by issuing a minimum service determination (MSD). Strangely, however, s 72(3)(b) was retained, meaning that once an MSA is concluded and ratified, the provisions of s 74 still do not apply. A further amendment was added by the Labour Relations Amendment Act 8 of 2018, which provided that once an MSD is issued by the ESC, the provisions of s 74 similarly do not apply.
In other words, the conclusion of an MSA, for all intents and purposes, became a mandatory requirement. But once an MSA is concluded or an MSD is determined, essential service workers are essentially locked into that process and permanently deprived of access to interest arbitrations. The only way to regain access to interest arbitrations is to conduct a ballot in terms of s 72(5), which requires that the majority of essential service workers must vote in favour thereof.
The upshot of these amendments is a labour structure that arguably obstructs access to interest arbitrations. Why not simply provide open access to interest arbitrations as an optional dispute resolution mechanism? Why is a ballot required? Surely, the legislature would want to incentivise the use of interest arbitrations as far and wide as possible.
It is unclear why having an MSA or MSD in place calls for access to interest arbitrations to be restricted in this manner. Interest arbitrations pose no threat to the existence of MSAs and MSDs and, since we do not have a compulsory interest arbitration system, impose no limitation on the constitutional right to strike. A ballot as a precondition for access to interest arbitrations seems an absurd and unnecessary requirement.
Moreover, the ballot process itself is fraught with practical difficulties. Many workplaces and sectors have multiple trade unions operating therein. Section 72(5) is all but geared toward majority trade unions who can more easily achieve the majority vote required. Minority trade unions would find it near impossible to do the same. Moreover, the ballot procedure can only work if the subject of the dispute affects the interests of the majority of the essential service employees. A minority trade union would have to convince not only its own members but also members of other trade unions to vote in favour of interest arbitration. A difficult feat, but exceedingly so, if the subject of the dispute only affects the members of the minority trade union.
Conducting s 72(5) ballots can be infinitely more challenging in sectors like the local government sector. If the dispute relates to an issue reserved for centralised bargaining, such as national wage negotiations, then over 200 000 essential service employees, located in 257 municipalities across the country, may have to be balloted. This is because s 72(5) requires a ballot of employees employed in the essential service, as opposed to essential service employees employed in the workplace. A local government trade union faced with such a gigantic and labour-intensive task, may consider striking as a more sensible course of action.
Against this background, it can be argued that the current structure of the LRA not only obstructs access to interest arbitrations, but at the extreme end, it can even compel essential service employees to strike.
What then, one may ask, was the objective behind the enactment of s 72(5)? Some explanation is provided in the explanatory memorandum of the Labour Relations Amendment Bill, 2012, which states as follows:
‘In order to promote interest arbitration and protect employees from an overly broad minimum service designation, the Bill provides that a minimum service designation will not apply if the majority of employees concerned vote to be covered by the broader essential service designation.’
It would, therefore, seem that the drafters of the LRA mistakenly believed that the s 72(5) ballot process would promote interest arbitrations. It has, however, shown itself to do the exact opposite. There are far better ways to promote interest arbitrations than to require a ballot process simply to gain access to it. The most obvious way is to remove all the unnecessary restrictions and make interest arbitrations generally available to all who wish to use it as an alternative to strike action.
The solution is clear and obvious. Section 72(3)(b) and (5) of the LRA serves no practical or justifiable purpose and should simply be deleted. In its absence, essential service workers would enjoy unrestricted access to interest arbitrations, regardless of whether an MSA or MSD is in place. No pre-ballot processes will be necessary.
Moreover, the deletion of s 72(3)(b) and (5) can more effectively realise the objective of promoting interest arbitrations, as outlined in the explanatory memorandum of the Labour Relations Amendment Bill, 2012.
The Independent Municipal and Allied Trade Union (IMATU) has made strident calls for a fundamental redesign of the labour structure pertaining to interest arbitrations. Together with other parties in the South African Local Government Bargaining Council, IMATU has strongly lobbied the ESC for some time to advocate for the deletion of ss 72(3)(b) and (5) from the LRA. The ESC, in turn, has pledged its support for the aforementioned changes.
It is expected that the Department of Employment and Labour will make the necessary amendments as part of the impending Labour Relations Amendment Bill, which is likely to be gazetted for public comment, by the time of publication of this article.
Craig Adams BA LLB LLM (UCT) is the Deputy General Secretary: Legal and Research at the Independent Municipal and Allied Trade Union (IMATU).
This article was first published in De Rebus in 2025 (March) DR 12.
