It is true we in live a world where technology and globalisation give the impression of rapid change and constant flux. The mouthpieces of capital, marching under the banner of ‘labour flexibility’, insist that this reality must be reflected in the world of work. Massive, global shifts towards temporary work, subcontracted and ‘disguised’ employment, we are told, are an inevitable feature of the ‘New Economy’.

If there is a fragment of truth to this argument, it’s that rapid technological change, growing machine intensity and computerisation have accelerated capital’s need for ‘re-tooling’ the workforce and have decreased reliance on skills that take a long time to acquire.

But to the extent that flexibility appears as new, it is only that it involves legal and contractual innovations designed to subvert the protections associated with the ‘standard employment relation’ (SER). We may note that it is estimated that at the time of the famous 1973 strikes in Durban, half of the 200 000 industrial workers in the city were ‘contract workers’. In fact, what we are now calling ‘precarious employment’ betrays capital’s primordial drive to reduce workers to a factor of production, this time by rolling back the victories and rights that the labour movement won through the latter part of the last century.

It is no coincidence, then, that the methods and institutions of precariousness in South Africa grew apace with stronger unions and more forceful labour regulation. Labour brokers began to spread in earnest around the beginning of the 1980s, coinciding with the re-emergence of the union movement, but really exploded when amendments to the Labour Relations Act in 1995 provided much greater protection for workers while at the same time reaffirming the position of brokers as legal employers.

Brokers are known formally as ‘temporary employment services’ but there is nothing ‘temporary’ about them – many, if not most, of the workers they employ are full time and permanent and would formerly have been employed directly. By introducing a third party into the employment relation – termed ‘triangular employment’ – brokers’ primary function is to undermine labour legislation and segment the workplace. Workers have a much harder time accessing labour protections if they cannot correctly identify their legal employer or target the source of their grievance. Their collective strength is hugely weakened when there may be up to a dozen employers on a single factory floor with whom they are required to bargain separately.

Brokering has become the central weapon in the assault on decent work in South Africa. It is likely that well over a million, perhaps two million workers are involved in triangular employment. They join an army of workers, many millions strong, who are victims of other forms of precariousness – including those who are forced to become nominally independent ‘subcontractors’ with no market power. Despite what is often claimed, there are not ‘two economies’ in South Africa, but one highly unequal ‘formal sector’ that reproduces and feeds off a vast and growing pool of informality.

Agriculture is a case in point. From the early 1990s until now it was witness to one of the largest human relocations in South African history: almost two million people were evicted as farms concentrated, mechanised and oriented towards international markets. Agribusiness reorganised labour usage, shifting en masse to seasonal, casualised workers and labour brokers of the most exploitative form, known as the ‘bakkie brigade’. It is estimated that only around 3% of farm workers are unionised.

Many brokers, often those employing cleaners and security guards, claim they are ‘service providers’ and not brokers as such, but in substance there is little difference. Economic research by Amandla! staff has provided statistical confirmation of what unionists know well: triangular employment imposes a significant wage penalty on workers. In light of this, the fact that the wage share of GDP began decline precisely after the imposition of stricter regulation in the mid-1990s was a first indication of labour’s failure to resist this onslaught.

The most recent surveys suggest that the overwhelming proportion of COSATU members, perhaps more than 80%, are SER workers – a statistic starkly at odds with circumstances of the broader class. While COSATU’s membership has climbed to over two million workers, union density in the economy as a whole (the proportion of ‘organisable’ workers that are part of a union) has decreased from a high of 58% in 1996 to around 30% today. Many in COSATU have now acknowledged that the failure to organise the precarious is a principal cause of the crisis facing the union movement.

That failure has fed criticisms that COSATU increasingly represents a privileged ‘labour aristocracy’ whose interests have diverged from the masses of un- and underemployed workers. There is no doubt that the composition of COSATU’s membership has shifted from predominantly unskilled and semi-skilled mining and manufacturing workers towards better educated, particularly public sector, workers. At an extreme, some have theorised the creation of an entirely new class – the ‘precariat’ – whose political life conflicts with the traditional proletariat and demands whole new strategies that depart from the factory floor.

The historical view of precariousness advanced at the start of this article should raise some doubts about these theories. Their proponents, moreover, tend to elide the significant overlap and interdependencies of different sections of the toiling classes – not least the fact that the vast majority of unemployed either live with or depend directly on the wages of formal workers.

However, the real common interests of workers and the unemployed will be meaningless if not organisationally articulated. Segmentation on the factory floor and in the broader workforce acts as a millstone around the neck of the entire class. In many places bosses consciously use this segmentation to entrench division – convincing a core of formal workers that their interests can only be safeguarded by excluding the precarious. Unionists have at times been complicit in this, refusing to organise casualised workers for fear of ‘informalising’ their membership.

These practices have accelerated the degenerative trends that have so far prevented the union movement from developing the creative responses demanded by the challenge of precariousness. By all accounts, the proud traditions of worker control and shopfloor democracy of the South Africa labour movement have weakened consistently through the democratic era. According to the latest survey, over 45% of COSATU workers did not attend a union meeting in the last year.

The use of union positions as a transmission belt into government and the private sector has affected the labour movement at every level – right up to the tendency for General Secretaries to join the executive of government. Shop stewards seeking a ladder out of indecent work have understandably leveraged the skills obtained through union work and often become brokers themselves or joined ‘human resource’ departments, draining the union of precious capacity. Labour brokering’s links with BEE, (allowing firms to win procurement points), have created a politically connected lobby behind the industry.

In many ways Marikana and its antecedents represent an open rebellion against the disjuncture that has grown between ordinary workers and the union bureaucracy. But amid the tragedy, there is also an opportunity: to use the energy and introspection occasioned by the crisis to revitalise the union movement, reanimate worker control and adapt to rapidly evolving conditions.

Some have suggested that answers lie in new organisational forms, such as general or regional unionism, that can more effectively combine formal, precarious and unemployed workers. But other struggles have shown that industrial unionism can work if the power and representation of precarious workers are actively cultivated. It is clear that diverse responses are required. The thousands forced into self-employed sub-contracting, for example, create the potential for co-operatives to make real contributions to working class power. In the embattled agricultural sector, securing victories may require a shift to social movement-unionism that unites broader sections of the rural poor in struggles linking immediate issues to demands for agrarian reform.

There will be no silver bullets. But as Guatam Mody, Secretary General of the New Trade Union Initiative (at the forefront of precarious organising in India) reminds us: just as precariousness is not wholly new, neither is ‘new unionism’. The process of re-inventing stands to draw on a rich history of organising from the era before labour had won significant rights and protections – not least in South Africa, where the union movement emerged in the most adverse circumstances.

Struggles against precariousness must be rooted in a reconstitution of worker unity on the shopfloor but cannot ultimately avoid addressing themselves to the macroeconomic and legislative structures of South Africa’s low wage regime. COSATU’s corporatist approach has so far yielded little from a state that is committed to a profit-led economy. But the most successful international struggles, those in Latin America, have shown that government participation in enforcing legislation and enabling the institutions of wage-led growth have been crucial to securing decent work. As government officials across the board swear fealty to a document advocating a social contract for ‘wage compression’, COSATU may soon be forced to reconsider its terms of engagement with the state.