Title: Economic Policy from Below
Subtitle: an Anarchist Critique of the COSATU Unions’ “Radical Reform” Project
Author: Warren McGregor
Date: 2019
Source: https://zabalazabooks.net/2020/03/26/economic-policy-from-below-an-anarchist-critique-of-the-cosatu-unions-radical-reform-project/
Notes: **** This pamphlet is an extract from the book Strategy: Debating Politics Within and at a Distance from the State – Eds. John Reynolds & Lucien van der Walt published by the Neil Aggett Labour Studies Unit (NALSU), Rhodes University, Makhanda, South Africa

Whereas many union movements in the world entered the 1990s in a state of political crisis, South African unions not only continued to grow very rapidly – the South African union movement was among the five fastest growing union movements in the world at the time – but also developed an alternative policy framework that bucked the neoliberal trend.

Centred on the Congress of South African Trade Unions (COSATU), and within it, key unions like the National Union of Metalworkers of South Africa (NUMSA), the unions developed an ideological and strategic orientation described by scholars (e.g. Eddie Webster and Glenn Adler, 2000) as “radical reform” or “structural reform.” The thinking of the main unions in South Africa remains, to this day, profoundly shaped by the “radical reform” (RR) model.

The aim of this input is to examine the RR model, which was an attempt to build on the many key progressive gains won by workers and their organisations through struggle in the 1980s, and push through to a deeper transformation in the 1990s. This input defines the key components of RR, and then examines why this innovative response to the parliamentary transition and to capitalist globalisation was not successful. This requires looking at issues of neoliberal capitalist and state domination, the impact of RR on the unions, and the effects of the institutionalisation of trade union activity and dispute processes that have taken place. It raises deeper questions about the unions’ politics as well.

Therefore, this section provides some ideas on:

  • Briefly defining radical reform and noting the assumptions on which the concept is based.

  • The historical and ideological-strategic contexts influencing the development of the approach.

  • The challenges and shortcomings of RR.

Radical Reform (RR) as Economic Policy-from-Below

RR came from the COSATU unions, not the ANC or SACP, and was a strategic trade union and working-class approach to socio-economic transformation of newly-democratic South Africa and the role that organised labour would play in this transformation. It was developed in the fires of struggle against late apartheid and capitalism in South Africa by COSATU and its affiliates and can confidently be considered an example of economic policy development from below that was developed through engagements with the ideas and desires of the rank-and-file in conversation with their elected leaders and officials.

Locally, the black working-class majority and its organisations were defeating the last vestiges of apartheid and racist capitalism, and engaging the newly developing institutions of a democratic South Africa. These struggles had incubated the development of powerful and militant working-class organisations, keenly aware of their power and historical role and responsibility.

At the dawn of the 1990s, COSATU entered into a formal alliance with the ANC. There was then a mighty, radical, mass-based and street-mobilising working class, closely linked to what was then a radical nationalist party, ANC, which was on the verge of state, power, as well as allied to the fastest growing communist party in the world at that time, the SACP.

Defining Radical Reform (RR)

According to the labour scholars Webster and Adler, writing in 2000, radical reform (RR) is a “left version of social democracy.” [1] Social democracy is the idea that the working class can win the existing state, using means like parliament, corporatism and expanded state control of the economy to shift society towards socialism through a series of reforms. A social democratic party is usually a mass party, as it needs maximum numbers to win elections.

What makes RR a “left” version of social democracy is that it was driven by mass, radical unions, who were willing to use rolling mass action to win RR; and, secondly, RR included many demands that were designed to give workers and unions direct power over economic decisions, including in company boardrooms and on the factory floor. The core ideas were that:

  • The restructuring of the post-apartheid economy must be done in ways that benefit the black majority, but the working class and poor especially.

  • South Africa should engage in globalisation after apartheid isolation ended, in ways that empower rather than oppress the working class. Unlike the “low road” of China, based on low wages and brutal suppression of unions, the new South Africa should follow a “high road” closer to Germany, with high-skill, high-wage, high-productivity workplaces in which black workers, in particular, would be empowered.

  • This would include – and this is the “radical” part – rolling back the frontiers of capital by giving workers and unions a greater say in the economy, fighting for policy reforms in the state like universal pensions and a great change in education that would protect labour from markets. At every stage, the working class would win – through the state and through bargaining – more power, leverage and skills.

  • It is not just about higher wages and skills and better conditions – it aims at codetermination of industry via tripartite bargaining and consultative forums with employers and the state.

  • The working class would also use union funds to build a “social economy” including cooperatives.

  • The growing conquests of the working class would be “building blocks” for socialism, each of which would allow further conquests, more ”building blocks,” so that the ultimate outcome would not just be capitalism (even if on “German lines” with co-determination and welfare), but rather a transfer of power to the working class, i.e. socialism.

Therefore RR is radical in that it seeks a future socialist worker-controlled society and economy, but reformist in its approach to transformation, in that it does not seek immediate revolutionary processes, but aims to build worker power by engaging the state and capital in organised industrial forums – e.g. NEDLAC and Industrial Bargaining Councils – pushing for increased worker control over economic management by consistently improving the conditions of work for labour. Reformism is a political strategy focussed solely on winning reforms, and it rejects revolution. In its social democratic version, it is argued that the effect of many reforms is a peaceful shift to a new society, removing the need for revolution.

What makes RR a more powerful application of the classical social democratic model, as per the ideas of key scholars, was COSATU’s presence in a formal alliance with the governing party, the ANC, and the SACP. The ANC’s certain long-term electoral mandate promised consistency regarding governance and policy development. COSATU would have vital access to key decision-makers in the legislative and executive arms of government, access developed not only by formal alliance, but also through comradely personal relationships developed through years of struggle. COSATU and its affiliates would also be able to use this Alliance to send key worker leaders into government as ANC electoral candidates. Further, COSATU would have access to a range of forums beyond parliament to present and win RR proposals, e.g. the Tripartite Alliance itself, ANC congresses, the SACP, NEDLAC and Bargaining Councils.

As such, COSATU could assert pressure on the ruling party to adopt progressive, working class orientated policies of RR, in three key ways:

  • As Alliance members and “inside” the party and the state.

  • As a union involved in government and bargaining forums with employers, both the state as employer, and the private sector.

  • As an independent union movement able to mobilise, when it deems necessary, mass worker protests and contestations “outside” the Alliance on the streets and at workplaces.

The Assumptions of Radical Reform (RR)

This approach to developing working class power and encroaching worker control rests on a few key assumptions, some of which I mention below.

Firstly, it assumes that the union movement is and will continue to be a vibrant, creative force able to respond (i) analytically to non-progressive ideas pushed by the state and private sector, and (ii) physically, via mobilisation, to the actions of the state and private sector and to do so quickly enough to either halt or change the situations facing the working class.

It also assumes that the state is an institution of governance that is able to be manipulated by different class forces – whether capitalist or working class – depending on the relative strengths of these classes in relation to each other. It thus also assumes that class is determined solely by economic relations of ownership of productive means.

Thirdly, RR assumes that ANC policy trajectories can be shifted in favour of policies advocated by the labour movement. COSATU thus acknowledges there are various ideological forces competing inside the party, but it does assume that the ANC, as the self-proclaimed party of the majority of people in South Africa, must then have a working-class bias or sympathy, and that eventually the ANC will come around to meeting the desires of working class people.

Importantly, RR is predicated on a large, organised, united, militant labour movement strong enough to coerce the state and private capital in a pro-worker socio-economic direction. RR also rests on continued ANC rule and a large section of the organised working class united in its desire for longterm and unfettered ANC rule.

Last, it assumes a somewhat one-way direction in change: each victory allows another victory, each building block allows another one. The assumption here is that more and more blocks can be won, until the system is basically socialist.

Ideologically, COSATU’s RR can be located in the sphere of social democracy. As such, its political orientation, including anti-capitalist rhetoric and its working-class bias, is within this framework, even if its political rhetoric draws on Marxism-Leninism and nationalism. It sees a particular role for the union movement, to be sure, but it views progressive transformation as being achieved through the state. It adheres to a stageist approach to achieving socialism, i.e. the idea is that capitalist economic growth (under the ANC) will develop the forces of production, which will enable the shift to socialism. What labour then has to do is make sure this development is used to benefit and empower the working class, so that the transition to socialism becomes possible. Its economic foundation is Keynesian as it seeks a state able to intervene in financial, commodity and labour markets in a way that benefits all classes.

Challenges and Shortcomings

Overall, while COSATU developed a wide range of RR proposals on everything from the chemical industry to pension funds, none were adopted in any serious way by the state or capital. For example, in the main case when a RR proposal was formally accepted – a proposal for reconstructing Spoornet – the state simply ignored the agreement. Achieving some of the desired ends of the RR strategy has faced an ANC increasingly founded on neoliberalism, which COSATU has been unable to shift despite the application of the RR strategy. This helps explain why COSATU keeps asking for a reconfigured Alliance, and for making the Alliance – not the ANC – the centre of policy.

The Global and Local Context

The RR strategy was developed in the contexts of the late 1980s and early 1990s. This was a time of rapid changes in the world. The balance of forces was shifting against the left and the working class internationally, and the local context was a transition that involved major compromises. The new phase of capitalism everywhere was neoliberalism – this was not even new in South Africa where the National Party had privatised ISCOR and SASOL.

It was the end of the era of Marxist states and the foundations of social democracy and trade unionism in the advanced industrial countries were under severe attack. This was the advent of the era of neoliberalism, structural adjustment and free-marketism, not only as regards socio-economic development. Nationalist parties across the poorer countries were embracing neoliberalism. Socialism and trade unionism were considered anachronistic and a wall impeding freedom – admittedly an attitude fostered by the propaganda of ruling and capitalist classes emboldened by their victories against organised labour and the Left. COSATU was growing, but it was an exception to the international trend, and while the SACP was growing, most communist parties worldwide were collapsing.

By the late 1980s, South Africa had become isolated from much of the rest of the world. As such, much of the foundations of RR betray a sense of South African exceptionalism, discounting the dramatic changes that were taking place on various international stages. Its ideas for development and the role of organised labour in the process of societal change seem outmoded when related to international changes. The focus was on South Africa, but South Africa was not an island, and even within South Africa, conditions were arguably challenging for RR.

As time would show, the ANC came under massive pressure to adopt neoliberalism, and did so decisively with GEAR in 1996. The confidence that the ANC would be open to radical projects like RR was shaken. Meanwhile, South African private capital, after flirting with ideas of a new deal for workers, turned back to neoliberalism, gutting jobs, using precarious labour, and expanding internationally.

Decline in Union Power

At the same time, the unions’ capacities and dynamism declined, even as their numbers swelled. By the early 2000s, COSATU had outsourced most of its political education to the SACP, as its own programmes were in crisis. Growing bureaucracy and corruption in unions weakened structures.

Links to political parties work both ways: fights inside the ANC spilled into COSATU, and a growing layer of COSATU leaders saw a job in the ANC as a profitable exit plan. The vibrant, creative, contested and relatively democratic education forums that had been established by the unions during the 1970s and 1980s, were to be reduced to classrooms of workers getting either technical training on the basics of shop-steward work, or narrow ideological and political education.

Increasingly, COSATU’s voice in the public declined, as the ruling party acted as the political filter for the voice of the organised working class. The lack of critical political education has contributed to this situation, imposing an economism on the unions as the ANC has been allowed to dominate the political terrain. Since the ANC itself and the larger Alliance are seen as sacrosanct, COSATU focuses on working with the ANC. In practice, this means – in seeing the ANC as leader of a “national democratic” (NDR) phase of South Africa’s post-apartheid trajectory – that COSATU responses are limited in scope. They cannot envisage the ANC itself as a stumbling block on the road to a proworkers’ society. They often tend to be about criticising certain leaders and policies – not the party. This has led to being entangled in factional battles within the party. This has dramatically reduced the political influence and authority COSATU has on the majority working class and its imaginations, many of the members of which have either sought other unions, political parties and organisations, or have disengaged from political activity altogether.

Problems Internal to the Radical Reform (RR) Project

The RR programme was ambitious, but some of its key ideas were actually quite vague strategically, perhaps because the routes to co-determination and socialism are not easily spelt out strategically. It was never quite clear why the ANC – as a multi-class party working in a capitalist state – should be expected to prioritise the working class. Capitalism, as the unions admitted, was a mighty force with a relentless drive to profit at the expense of workers – how then would worker and union partnerships with capital through co-determination and NEDLAC not end in unions assisting capitalism? Also, the stageist approach meant a long-term alliance with the ANC, which brought its own problems (as we have seen earlier in this chapter).

COSATU’s pathways to achieve co-determination are also not clear. How would this happen? What would it mean? How would workers do this without being made into agents of capitalism? The use of industrial Bargaining Councils, a positive development won through struggle, became an end in itself, rather than a means to push industry in a pro-worker and co-determinist direction.

The desire for tripartite bargaining institutions – structured forums for dialogue with bosses and an unclear desire for eventual co-determination – has fostered the institutionalisation of union activity. In the neoliberal era, these forums are continually under attack and rendered powerless as they are turned into mere consultative arenas, with little to no decision-making power. So unions have become increasingly integrated into forums that are increasingly pointless.

Additionally, effective engagement in these forums requires a specific and high-level skill set, meaning outsourcing of research and legal representation and an increased bureaucratisation of the union. RR policies are technical and cannot be easily developed from the ground-up. They get given over to specialists, which then means ordinary workers have only a limited idea of what is actually being proposed and little space to change it.

Engaging the state and bosses, in the Alliance, parliamentary caucuses and boardrooms cannot be done by all members of a union. As this, as well as the tripartite structures mentioned above, is a major factor of the RR strategy, much power becomes centralised in the hands of leaders of the organisation due to the high-level needs of these forms of elite, individualised engagement. This has led to increased distance between rank-and-file members and elected leaders and a growing bureaucracy and authoritarianism in unions.

The 1990s also saw a huge “brain drain” from the union movement of some its most capable leaders and activists to the ANC, particularly at election time. This has had the obvious negative effect of a reduction in the capacity of the movement to respond – with effective RR proposals – to the socioeconomic changes that took place under ANC rule. Another unfortunate effect has been the increasing numbers of union members viewing union work as less a “calling” than a stepping stone into long-term employment in the state or even private sector.

COSATU is also part of a fracturing and fragmenting labour movement. Many new unions have been formed as breakaways from established unions, or as altogether new worker organisations. There has been much discussion of the external local and international conditions that have caused this. However, new worker formations must also be considered as a direct response by workers to the problems they perceive in the established unions, which are no longer attractive to them. Thus fragmentation must also be seen as a challenge by workers to trade unions and unionism.

Despite the definition offered above, both the literature and the application of the RR strategy shows a distinct lack of ideological and strategic clarity. There is no clear definitive end goal that is established and thus intermediate milestones are not clearly articulated. Much of this can also be related to a clear lack of critical political education in the workers’ movement, thus not allowing for open discussion and debate amongst workers of their organisations’ ideas and strategies, which, in turn, means that workers have little say over the directions their organisations take.


Workers and their organisations have responded to their increased impoverishment and lack of ability to impact society. I have mentioned that fracturing and fragmentation of unions should also be considered as internal worker responses to a distinct lack of union militancy and democracy. These too, though, face serious challenges. Many of those who have left COSATU have formed a new federation, the South African Federation of Trade Unions (SAFTU), and new unions. Many of these new formations are workplace, city and region based. Some worker leaders, particularly those from NUMSA, have also formed a new political party to contest state elections.

Yet, these newer formations exhibit real similarities in organisational structure to the formations that their members have left – for example, big man politics and a centralisation of power are also found in these newer formations. In addition, there seems to be no real ideological shift developing in these new organisations. For example, they may be very critical of COSATU’s alliance with the ANC, but most still see their political futures through the lens of political party power and the state, and many concrete SAFTU and NUMSA proposals remain very much in the RR framework. This limits the imagination of what a trade union can do (and has done) as regards social transformation.

[1] Webster, E. and G. Adler. 2000. “Introduction: Consolidating democracy in a liberalising world – trade unions and democratisation in South Africa.” In Webster, E., and G. Adler. (eds.). Trade Unions and Democratisation in South Africa. Johannesburg: Witwatersrand University Press.