Introduction
On 1 January 2019 the most significant changes to South Africa’s labour legislation since 1995 came into effect, fundamentally altering how members of registered trade unions can embark on protected strike action.1 New picketing rules combined with the introduction of secret strike ballots, amongst other provisions, represent a coordinated response from state and capital to increasing levels of working-class mobilisation. The amendments to the Labour Relations Act (LRA) come at a time when internationally the right to strike is under attack. A recent study of 70 countries found that almost all, 68, have adopted or made use of legal provisions that violate the International Labour Organization’s (ILO) principles on the right to strike (Xhafa 2016). In comparison, the rights of South African workers to go on protected strikes remain better than in many other places in the world, a reflection of the militant history of the South African labour movement and the gains it has made for workers since 1994. But the erosion of these rights should be a cause for concern for activists and labour scholars in South Africa and beyond.
What is, perhaps, surprising is that the amendments have been actively supported by the Congress of South African Trade Unions (COSATU), South Africa’s largest trade union federation. Indeed, COSATU’s parliamentary coordinator, Matthew Parks, has hailed the amendments as ‘the most historic and progressive changes to South Africa’s labour legislation … since the end of apartheid and the dawn of our new democracy in 1994' (2018). In contrast, the South African Federation of Trade Unions (SAFTU), which joined a coalition of organisations opposing the Bills, has described the amendments as ‘the most glaring attack on workers since the dawn of democracy … no different to those introduced in Britain by the late Prime Minister Margaret Thatcher' (2018). The differences between how the two federations responded to the Bills represent more than just the political fallout between COSATU and SAFTU. The amendments, and their implications, illustrate significant fault lines within the trade union movement and its relation to working-class struggle in South Africa today. This briefing will analyse the LRA amendments, their implications and what they tell us about the state of the labour movement, the possibilities for trade union revival and the state of working-class struggle.
The amendments and their implications
The 2019 amendments arise from a process set in motion by former president Jacob Zuma in his state of the nation address on 17 June 2014, in the midst of the five-month platinum sector strike. A social dialogue process, led by then deputy president Cyril Ramaphosa, met on 4 November 2014 to consider how to address wage inequality and ‘violent’ strikes. COSATU, alongside the Federation of Unions of South Africa (FEDUSA) and the National Council of Trade Unions (NACTU), were a part of the subsequent negotiations through the National Economic Development and Labour Council (NEDLAC). The amendments to the LRA and the National Minimum Wage Bill were then published for public consultation on 17 November 2017. From the outset, the justification for amending the strike provisions in the LRA has been based on an ideological argument that strikes are increasing and increasingly violent, an assertion that is not borne out at all in the available data (see Runciman 2018a). A close reading reveals that, rather than attempting to minimise violent strikes, the amendments are, in fact, designed to discourage strikes in general.
Overall, the amendments are intended to delay and frustrate the ability of unionised workers to embark on procedural strike action. When a trade union intends to take strike action it refers a dispute of mutual interest to the Commission for Conciliation, Mediation and Arbitration (CCMA) or to the relevant bargaining council. A commissioner then attempts to mediate between the employer and trade union to resolve the dispute at conciliation. If the commissioner is unable to resolve the dispute at conciliation, a certificate of non-resolution is issued, commonly referred to as a strike certificate. After it has been issued the trade union must give the employer 48 hours' notice of any strike action. The conciliation process has now been extended to 35 days. While this may seem a relatively minor change, the impact of this amendment is that it provides employers longer to prepare for strike action through stockpiling and the employment of alternative labour. The likelihood of this is further exacerbated by the introduction of compulsory picketing rules.
Previously, it was not a legal requirement that picketing rules be agreed before the certificate of non-resolution was issued. Now picketing rules have to be agreed prior to the certificate of non-resolution being issued. If the parties are unable to agree the picketing rules then the commissioner will have the power to determine them. The default picketing rules, which were released for public consultation on 17 September 2018 and signed into law on 12 December 2018, further reveal the extent to which the right to strike is under attack. While the default picket rules are intended to be used when unions and employers are unable to mutually agree on picketing rules, the concern is that these regulations will become the norm and impose severe limitations on all strikes.
The default picketing rules are required to state when and where a picket will take place and limit pickets to a maximum number of workers. Members of the public will be banned from showing solidarity with striking workers as pickets will be limited to striking workers and union members only. In addition, commissioners are granted far-ranging powers to decide what workers can and cannot do on a picket. For example, they will be able to decide what songs can be sung, what can be written on placards, whether workers can be addressed by a union official or speak to members of the public about their strike. Commissioners will not be required to justify any limitations they place on the activities of striking workers. This may, potentially, undermine not only the right to strike but the right to freedom of expression. If any of the picketing rules are broken the trade union will be required to suspend the picket until it has satisfied a commissioner that it can control the picket. This opens up trade unions to having to suspend their picket for the slightest reason and fundamentally undermines the collective power of a strike (Runciman 2018b).
In addition to this, the amendments also empower the Labour Court to overturn picketing rules established at the CCMA and interdict any strike where picketing rules are deemed to have been breached (Dor 2018). This comes at a time when the labour court is demonstrably pro-employer in its rulings (see Wesso and Schroeder 2018). The default picketing rules entirely favour employers who will know even before a certificate of non-resolution is awarded when and where a picket will take place, providing employers ample time to stockpile, hire alternative labour and undertake other action to undermine workers’ collective power (Runciman 2018b).
The hand of employers is strengthened even further by the amendments by granting them stronger recourse to resolve strikes through advisory arbitration. An advisory arbitration panel can be established if the Minister of Labour or the Director of the CCMA deems it to be in the ‘public interest’ or if it is requested by any party to the dispute, including employers. The amendment sets out criteria for ‘public interest’ that are so wide, almost any strike could be subjected to advisory arbitration. Furthermore, advisory arbitration can be requested as soon as a certificate of non-resolution is issued and before a strike has started. While the process of advisory arbitration does not suspend a strike it draws union officials into highly technical procedures. Furthermore, the ruling by the arbitration panel is binding unless rejected by the trade union within seven days, after consulting with its members (CWAO 2018).
One of the most contentious amendments has been the introduction of secret ballots for registered trade unions. Section 95 of the LRA has always required registered trade unions to include strike ballot procedures within their constitutions but until now it has not stipulated how the ballot must be conducted. This flexibility has enabled unions to take decisions in line with the direct and participatory forms of democracy that have been integral to the trade union movement in South Africa (Desai 2016).
The fact that balloting provisions already existed in the LRA has allowed COSATU and its allies to present the necessary fiction that the introduction of secret balloting is a clarification rather than a change in the law. In COSATU’s joint submission to parliament with FEDUSA and NACTU, they argued
Unions and employers’ organisations have been required for the past 23 years to have the mandatory requirement to ballot their members when they want to strike. The sole change to the balloting provisions in this bill is to insert the word secret before ballot. The majority of unions, political parties and other organisations elect their leadership through secret ballots … This insertion is a simply [sic] clarification as to what a ballot constitutes. It is line with common law, practise [sic] and the Constitution. Organised labour is comfortable that it has protected workers’ hard won constitutional rights to strike. (COSATU, FEDUSA, and NACTU 2018, 20–21)
While COSATU and its allies may ‘be comfortable’, international experience clearly demonstrates how the introduction of secret balloting has been a key tool with which the state and capital have used to reduce strike action (see Kelly 2015). But COSATU did not even need to look at international experience to understand the implications of the reintroduction of secret balloting for workers. Indeed, only five years earlier COSATU was at the forefront of defending workers against the reintroduction of secret ballots, arguing
that balloting requirements were a distinct feature of the apartheid legislative regime. Its absence in the current LRA was no oversight but rather an acknowledgement of the extensive abuse of technicalities by employers around balloting to prevent industrial action. Its reintroduction reflects a fundamental attack not only on the right to strike but also on collective bargaining. (COSATU 2012, 11)
A contentious decade
Alexander (2015) argues that since 2004 South Africa has been experiencing a ‘rebellion of the poor’ and ‘a rebellion of workers’. Community protests predominately led by the unemployed around basic service provision have become a daily occurrence, with at least 1500 protests occurring in 2017 (see Alexander et al. 2018). The numbers of protests indicate the levels of working-class struggle but also its fragmentation. Community protests are most often isolated from one another and, to date, little in the way of common demands or political perspectives have emerged. The fragmentation of these struggles has led Bond and Mottiar (2013) to characterise such protests as ‘popcorn protests’, reflecting the way in which they spring up and then fade away. However, this term underestimates the political significance of this protest wave regardless of its weaknesses (see Runciman 2016).
Whether there has been a ‘rebellion’ of workers is debatable (see Runciman 2018a), but there have been important shifts in the nature of workers’ struggle, many of which were encapsulated in the 2012 strike wave in the platinum sector. The platinum strikes were notable not only for their intensity but for the fact that they were propelled by independent worker committees (Sinwell and Mbatha 2016), reflecting the disillusionment that workers had with the National Union of Mineworkers' sweetheart unionism and the breakdown in the post-apartheid corporatist framework (Runciman 2019). The fact that the strikes were both unprotected and organised outside of unions is illustrative of some of the important trends in workers’ struggle today. The proportion of unprotected strikes has been rising: between 2000 and 2010 a third, on average, of strikes were unprotected. In recent years, the percentage of unprotected strikes has grown considerably: between 2012 and 2016 on average half of all strikes were unprotected. In addition, although still a minority trend, it is important to note that the number of working days lost by non-unionised workers has been growing (Paret and Runciman 2016, 306–307). The breakdown of the post-apartheid corporatist framework, where COSATU is no longer able to guarantee the discipline of its members, has allowed capital, with support of the state, to pursue conditions favourable to it by restricting the right to strike. This, however, does not explain why COSATU has chosen to support rather than defend the working class from such attacks.
Why COSATU gave away the right to strike
It has become widely acknowledged that COSATU is in crisis (see Satgar and Southall 2015). The challenges it faces are complex and multi-dimensional. For the purposes of the discussion here, three factors are briefly examined: COSATU’s political alignment, changing class composition and bureaucratisation.
COSATU’s alliance with the African National Congress (ANC) and the South African Communist Party (SACP) and its participation within NEDLAC was intended to provide a vehicle for ‘radical reform’ (Von Holdt 1992, 33). But, as Ashman, Levenson, and Ngwane argue, it has largely ‘rendered it [COSATU] toothless’ (2017, 10). Indeed, as Schroeder (2002) highlights, the constraints on COSATU’s militancy could be seen as early as 1995 when, at the height of its power, COSATU failed to win any of its key demands in relation to the re-drafting of the 1995 LRA (see also Lehulere 2015).
Not long after the 1995 LRA was enacted, the ANC abandoned the broadly neo-Keynesian Reconstruction and Development Programme for the explicitly neoliberal Growth, Employment and Redistribution (GEAR) policy (Bond 2000). This was a real political test for COSATU. While vocal in its critique of GEAR, COSATU was also careful to politically distance the ANC from it by arguing that that it had been implemented by technocrats within the state (Lehulere 2003). In addition, COSATU distanced itself from the new social movements that emerged in the early 2000s that were contesting the impact of GEAR. Instead, as Satgar and Southall argue, COSATU remained in ‘an outdated vanguardist model of politics, assuming that change in leadership within the Tripartite Alliance would bring about transformative changes in government economic strategy' (2015, 9).
But, as Lehulere (2003) notes, what was concealed beneath COSATU’s radical rhetoric against GEAR was that its own economic strategy was increasingly drifting to the right. In the political discussion document ‘Advancing social transformation’, it is argued that GEAR had been instituted in the interests of mining and finance capital and had compromised the interests of manufacturing and small-scale producers. The significance of this document, according to Lehulere, is ‘that it represents the first attempt within COSATU to define a politically explicit and programmatic basis for an alliance with capital (or at least sections of it) and the state’ (2003, 37 emphasis in the original), marking the beginnings of COSATU’s drift towards business unionism.
At the same time, the workplace has been restructured through the use of temporary employment services and outsourcing, as had been experienced elsewhere in the world. COSATU recognised as early as 1997 the need to organise atypical workers; the September Commission (1997), a report into the future of unions, dedicated a chapter to thinking about how to organise new sectors. However, proposals that could have assisted in organising atypical workers, such as developing flexible policies on union subs including lower rates, allowing periods of unpaid membership and demanding that employers pay union subscriptions, were never implemented. Overall, the report focuses on how to organise white-collar workers with the chapter dedicated to organising new sectors mentioning white-collar workers 58 times, compared with 11 mentions of non-standard work/employment and 5 mentions of vulnerable layers. Today 90% of COSATU’s membership are permanent workers, with two-thirds of those members engaged in skilled or supervisory work (Bischoff and Tame 2017, 66). The traditional blue-collar worker that was historically the base of COSATU has declined in their significance. As a result, the rate of unionisation has declined from a high of 47% in 1994 to 29% in 2012 (Macun 2014, 43). This has to be understood as not simply a result of the forces of globalisation but also, I argue, as a direct result of strategic choices that COSATU has made over who to organise.
Organising a shrinking section of the working class has combined with processes of bureaucratisation, corruption and a breakdown in democratic traditions within the movement leading to a demobilisation (see Beresford 2015, 97). These are challenges that COSATU has directly acknowledged for over a decade yet little seems to have been undertaken to alter these processes, as the strike wave on the platinum belt in 2012 illustrated.
In the face of this crisis the 2019 LRA amendments offer a partial remedy to the waning influence of organised labour. As trade union membership has declined, the status of unions within bargaining councils has been threatened as they are at risk of no longer being representative. The amendments alter how representivity, for the purposes of establishing a bargaining council or extending agreements, is calculated. Previously, both the employer organisation and trade union must be in the majority, meaning that the employer organisations employ the majority of employees and that trade unions represent the majority of employees who are trade union members. Now only one party will be required to ‘represent’ the majority: either the employer organisations employ the majority of employees or the trade unions represent the majority of union members. Furthermore, the amendments mean the Minister of Labour must consider the composition of the workplace when determining representivity, including the extent to which employees work for temporary employment services, part time and other forms of non-standard work. While this is a potentially progressive move, what is worrying is that COSATU has previously argued that such workers should be excluded from the calculation of representivity largely because of their failure to organise such workers (see COSATU 2012).
Defending the changes to representivity, both the Department of Labour and COSATU have argued that these changes will strengthen collective bargaining (see WWMP 2018). Yet a review of bargaining council agreements undertaken by Bassier (2018) demonstrates that the majority of bargaining council agreements negotiate below the supposed minimum floors set out in the Basic Conditions of Employment Act, calling into question the extent to which collective bargaining is currently advancing workers' interests. The changes to representivity is one that suits both capital and the interests of trade unions. Employers will be able to enter into agreements with unions that may, in reality, undermine workers’ rights, while at the same time minority unions will be able to retain their positions without having to organise workers, particularly precarious workers who constitute a significant section of the workforce.
Overall, the impact of the amendments will increasingly direct industrial disputes away from the hurly-burly of organising workers and strikes and direct them towards the boardroom or arbitration processes. This is beneficial to an increasingly bureaucratised trade union movement that has become increasingly routed in proceduralism and legalism as it seeks to protect its waning status.
Resistance to the Bills: a new hope for the labour movement?
The amendments were not passed without resistance. The Scrap the Labour Bills campaign was initiated at a meeting co-organised by the Casual Workers Advice Office (CWAO) and the General Industrial Workers Union of South Africa (GIWUSA) on 20 January 2018, attended by 26 different organisations. All agreed to form a coalition against the Bills but the primary organisations in the campaign were the National Union of Metalworkers (NUMSA), CWAO, GIWUSA, SAFTU and Outsourcing Must Fall. The campaign organised a march in Johannesburg on 21 March 2018 where an estimated 5000 people participated, followed by a general strike on 25 April 2018, called by SAFTU and supported by the wider campaign, where it is estimated 12,000 workers participated in the march in Johannesburg that day (Wesso 2018). The campaign against the labour bills has the potential to represent an important moment in the processes of trade union renewal, particularly given the attempt to work with civil society organisations, but the campaign also illuminated some of the continuing challenges faced by the trade union movement.
It is noteworthy that the initial meeting of the campaign was largely initiated by CWAO, a non-profit organisation, and not a union. Indeed, in the initial meeting a representative from NUMSA admitted that they had been present at the Department of Labour briefings on the Bills that had happened prior to their public release but they had done nothing to sound the alarm. This, it became clear, was illustrative of the weakened organisational capacity within the trade union movement. Both NUMSA and SAFTU struggled to interpret the Bills correctly. For instance, a joint NUMSA–SAFTU pamphlet for the march of 21 March 2018 was issued with mistakes regarding the legislation around changes to bargaining council representivity. In addition, SAFTU spokesperson Patrick Craven on several occasions inaccurately claimed that the secret ballot provisions would apply to unorganised workers, when they apply only to members of registered trade unions (see Cape Talk 2018). SAFTU also failed to make its submission to parliament on time but was allowed to make a late submission.
The coalition also exposed tensions as contrasting organising traditions between unions and other organisations bumped up against one another. SAFTU did not play a prominent role in the coalition until after the success of the march on 21 March 2018. Indeed, Zwelinzima Vavi, the general secretary of SAFTU, planned to attend an event organised by the Freedom Park community, from the south of Johannesburg, on the day of the march. The lack of participation from SAFTU was openly criticised by Ruth Ntlokotse, the second deputy president of NUMSA, at the march. Following the success of the march SAFTU began to play a much more active role in the campaign. However, its approach was often undemocratic and the memorandum that was delivered on the day of the general strike was done with little consultation with the other organisations in the campaign.
SAFTU also used the campaign to push its own demands that it be admitted into NEDLAC and that the Bills be returned to NEDLAC for further deliberation. This was done without any public reflection, at least, on how NEDLAC has underpinned neoliberal reforms to the labour market and compromised the trade union movement. SAFTU’s desire to enter into NEDLAC calls into question the extent to which SAFTU will challenge the existing arrangement of class forces in South Africa and, in turn, be a vehicle for trade union renewal.
Conclusion
Working-class struggle, although fragmented, has been growing, posing not insignificant challenges to the ANC’s hegemony and, in turn, the position of COSATU. The aim of the amendments is to constrain and contain working-class struggle by making it more difficult for unionised workers to embark on protected strike action and to increase the power of the state to resolve strike action. The amendments have been achieved with the active support of COSATU, which has fully embraced its role as a manager of capital. Indeed, COSATU’s weakened position within the working class politically and organisationally necessitates a greater reliance on its relationship to the ANC and capital through the Tripartite Alliance and the post-apartheid corporatist framework. The 2019 LRA amendments should be viewed, therefore, as part of the attempt to restore the ‘political monopoly’ (Buhlungu 2016, 145) between COSATU, the ANC and capital, thus further entrenching neoliberalism.
This briefing has suggested that while the Scrap the Labour Bills campaign may represent the prospect of trade union renewal, it also exposes some of the critical challenges confronting the trade union movement. The fact that SAFTU used the campaign to demand admittance into NEDLAC demonstrates its desire to re-enter a corporatist framework that has failed to bring about ‘radical reform’ for the working class. This calls into question to what extent SAFTU has broken from the established political and institutional framework. The prospects for trade union renewal will rest, in part, on the extent to which SAFTU-affiliated unions will continue to struggle against the amendments. For instance, will they pose serious challenges to the institutional framework by defying the secret ballot provisions and risk de-registration? Will SAFTU-affiliated unions continue to work with and build alliances outside of the trade union movement?
What is certain is that working-class struggle will continue. In recent years there has been a growing trend of unprotected strikes; it seems likely that this could continue as workers abandon the institutional framework to press their demands. It is also possible that we will see more workers organising outside of trade unions in order to overcome the restrictions that will now be placed on unionised workers. This may lead to a revived labour movement that looks radically different from the one we see today.