Reshaping skills policy in South Africa: structures, policies and processes

The reforms of state policy in the early period of the transition were developed in the context of a broader response to a deepening capitalist crisis globally and in South Africa. A radical black working class organized in COSATU had gained momentum throughout the 1980s and the trade unions were becoming stronger given their mass power. Union strength meant that the terms of negotiating new labour reforms were crucial in ensuring that policies to address historical imbalances were developed. Such policies included skills development legislation that needed to be overhauled to take into account the imperatives of equity and growth, while also ensuring democratization of policymaking processes (Ngcwangu, 2016).

Policymakers should consider reformulating skills policy that address economic requirements that are clearly defined and are relevant to the indicators shown by industry. It is difficult to achieve this specific alignment as indication of skills requirements from industry are often not clearly defined or quantified. Research can be used to improve knowledge of skills requirements through reliable data analysis and predictions of future skills requirements which would aid the state in planning. On the other hand, the state should use research and policy to respond to community development through skills provision, especially given the rising unemployment levels in the country. The state should strengthen the work of funding bodies such as the National Skills Fund (NSF) to support NGO, CBO and local skills development initiatives that can also bolster skills training in the informal sector of the economy.

Through a method of periodization, this article provides an overview and critique of the structural changes, processes and policy developments in the arena of skills policy in South Africa since the early 1990s. The objective is to locate the proposed new changes (which will come to effect in 2020) against the background of changes to skills policy that have occurred in the period of the democratic transition. The article has four segments: (1) skills development – a contested concept; (2) an overview of skills policy changes: 1990-2008; (3) key policy changes in NSDS III (2009-2016); (4) evaluation of NSDS IV proposals: 2020-2025.  



The contextual character of production and the nature of relations between management and employees shape skills development. Tilly (1988) opposes a narrow definition of skills and maintains that ‘Skill is a social product, a negotiated identity. Although knowledge, experience and cleverness all contribute to skills, ultimately skill lies not in characteristics of individual workers, but in relations between workers and employers’.  In a sense, the notion of skill is intricately connected to the work or production process in a given work environment or industrial sector. According to Vorwerk (2005), ‘The concept of skills has been rehabilitated since the early 1990s. At that time it was mostly applied to manual labour and low-level activities. But since then the term has surfaced far more often and is now linked with far more complex behaviours: people skills, negotiation skills, problem-solving skills, ICT skills etc. These shifts in definition and understanding are reflective of deeper shifts in the language associated with the broad field of education-training and skills.  

The South African skills system is a complex interaction of institutions (educational and labour market) and policies which govern these institutions, have numerous sub-systems, all of which can be generally referred to as part of the skills system. Kraak (2004a ) notes that the new institutional architecture of skills development as a whole departed from the apartheid system by envisioning training across occupations and qualifications rather than just apprentices; it focuses on sectors rather than industries; incorporates SMMEs and accommodates a wide variety of people in the labour market critically – the pre-employed and the unemployed (ibid, 2004: 118). This is a significant departure which has resulted in the broadening of the concept of skills beyond just apprenticeship and artisanal training.


An Overview Of Skills Policy Changes: 1990- 2008

By the early 1990s the reforming apartheid government had begun to explore ways of engaging the growing black trade union movement in attempts to build a social accord. The National Training Board (NTB) developed the National Training Strategy Initiative (NTSI) with the so called ‘blue book’ as its foundational document. The Manpower Training Act, 1981 was passed following the decisions of the Wiehan Commission. The thinking in the state at the time was that a principle of allowing some into the labour market and skilled trades would diffuse the ‘many’. The South African economy was experiencing a rising demand for skilled labour and these could not only be drawn from the white community as many white artisanal skilled workers had been gradually moving into the professions and ‘white collar’ work in the civil service (Crankshaw, 1994; McGrath, 1996).

It was against this backdrop that under the provisions of the Manpower Act of 1991, 33 Industry Training Boards (ITBs) were established by the government at the time. The ITBs were focused on artisan training. The NTB together with the Human Sciences Research Council (HSRC) produced a number of reviews of the training system. The main interest groups on these review processes were the apartheid state and the mining sector. The apartheid governments prioritized elements of reforms given the growing popular local and international resistance to the system of racial segregation. Apartheid governments also had interests in education reforms and building of manpower resources to contribute to the rebuilding of the South African economy given the years of its exclusion through international sanctions and the pariah status of the country in international affairs.  

The transition from the Industry Training Boards (ITBs) to the SETAs (Sector Education and Training Authorities) had implications for all the policy actors in the skills system. Kraak (2004 a) has described the main differences between the old and new institutional forms as consisting of the following factors:

  • ITBs had a narrower focus of training apprentices. The SETAs expand training at all occupational and qualification levels;
  • The new institutional environment to be set up around ‘Learnerships’ is intended to turn   around the decline in training due to the demise of the old apprenticeship system;
  • “Sectors” are larger than “industries” and include a number of industries previously untouched by ITBs or demarcated by separate ITBs;
  • SETAs are also responsible for including people who are not in formal employment, for example, in SMMEs, in job creation programmes, pre employed youth and the unemployed’ (Kraak, ibid:118).

The SETA model of funding skills training arose within the context of these considerations. The ITB system had been more employer dominated with very little participation by employees. Training under ITBs was largely focused on artisanal training, was done nationally, had a coverage of 50% to 100% in their sectors and, like the SETAs, had a low coverage of small businesses. SETAs have been criticized for being highly bureaucratic, being mostly regulatory, accused of poor management of funds and inability to keep closer relations with sectors. However these assessments vary between SETAs as some are seen to be more efficient than others.

Since the promulgation of the Skills Development Act in 1998, and until 2008, skills development was the responsibility of the Department of Labour (DoL) and was framed in government policy as part of labour market interventions to bring equity and overcome the apartheid legacy in the economy. Criticisms are that in the post-apartheid era there has been a weakening of the state’s capacity to intervene in skills development resulting in a lower number of artisans being produced in the country. Lolwana and Ngcwangu (2015) argue that consequently, there was a significant drop in the number of qualified artisans (from 13500 in 1985 to 5145 in 1999 to less than a 1000 by 2000).  With the recent campaigns on artisan development2, the numbers are slowly climbing up again. However, placement rates after training are still problematic.

A central concern about the new skills development dispensation is the introduction of the learnerships system which is seen by business to have resulted in the weakening of artisanal training. The learnership system is seen to have been useful for training in sectors that did not have industry training boards such as banking and finance as it was a system that gave training opportunities to people at the lower levels of the NQF such as bank clerks, hairdressers, retail jobs etc. But in the artisanal type industries the learnerships model is seen to have weakened the apprenticeship system. What this means is that rather than build on the numbers of apprentices the new skills system actually weakened the capacity of public institutions to train more artisans.

The decline in jobs and the need for entrepreneurs to create employment are critical factors, which can be supported by the production of more artisans for the country. Other more recent developments aimed at addressing the weaknesses of skills development include improvements to workplace training and simplifying the work integrated learning requirements by employers. The NSDS III strategy highlights the need for employers to open up their workplaces for more training, particularly for those students that need the workplace component in order to complete their studies.  



By 2009 with political changes inside the ANC precipitated by the historic Polokwane conference, skills development within the state was shifted to the Department of Higher Education and Training (DHET) which has articulated skills development within a framework of a Post School Education and Training system rather than just as a labour market issue. Priority has been given to improving co-ordination within DHET and across the government. As it is common in South Africa, the tendency is to create more policy and more structures. The difficulty with this approach is that due to the short-term nature of political cycles, policy implementation tends to be held back by bureaucratic management. However, what is evident in the NSDS III (2011-2016) is a shift towards an expansionary approach and more financing for vocational training.

Other changes in the NSDS III are related to grant regulation reforms (reduction of the mandatory grant), more funding for full academic qualifications, strengthening of artisanal training and pressure on employers to make workplaces available for training purposes. The new skills development landscape as articulated in the NSDS III has a more direct explanation of the problem, which is to address the socio-economic crisis of the ‘NEET’ (Not in Employment, Education or Training) category of youth, which is estimated at approximately 3 million young people. Not only is co-ordination a challenge for the state but also the disarticulation of government programmes to the extent that the effect of well-conceptualized policies is not fully achieved.

Various attempts at reforming the image of SETAs and the delivery model are currently under way.  In general, the skills system has been a complex and often ‘messy’ affair to manage, given the competing interests of private service providers, intellectual communities of researchers, employer interests and poor co-ordination from the state resulting in this set of problems leading to a crisis. It is against this backdrop that the DHET has issued proposals to restructure the skills development landscape particularly to redirect the SETAs work, improve governance and centralise the management of the skills system to the DHET through a national skills council.  But will mere bureaucratic rearrangements address the key structural challenges confronting the poor and the working class?



At the core of discussions about the SETA system are the bureaucratic inefficiencies, perceived corruption, governance problems that arise due to resistance to change by some within the business community, misalignment of skills policy with industrial policy and potential ‘scope creep’  by the SETAs. Meaning that expectations on what SETAs should do and what their mandate requires is often a cause of conflict between stakeholders within the skills development community. The SETAs are central to the South African model of skills development; however, in their current institutional character SETAs play a more of a regulatory role than an actual training role. This dominates the work of SETAs with the actual training largely outsourced to private training providers. There are three aspects of the new proposals that will have a far-reaching impact on the skills system should they be implemented.

Firstly, in their problematisation of the challenges facing the skills system in South Africa , the DHET has presented four options for consideration by the public and stakeholders.

  • Option 1: Minimal Change
  • Option 2: Cluster SETAs
  • Option 3: Skills Council
  • Option 4: SETAs as an integral part of the larger PSET system

Options 2 and 3 rest on the assumption that the co-determinist model of engagement will be retained but in different sectors of the economy. The overarching framework of the proposal is to increase centralization of the skills system to the DHET which could be perceived to undermine the involvement of other stakeholders such as labour, civil society and business in shaping skills policy. Enhanced stakeholder participation ensures that the original vision of skills policy in South Africa which is anchored on co-determination is achieved even through there have been structural weaknesses in labour and civil society’s involvement in conceptualizing skills policy. Workers at the shop-floor level are critical to the skills debate as their constituency are the ones who objectively stand to gain from progressive changes to the skills landscape.

Option 4 proposes an integrated PSET system in which SETAs have stronger alignment to government departments ensuring that their research plans are in line with the plans of the ‘line function’ government departments.  This proposal has pros and cons. The ‘pros’ are that in critical areas such as industrial policy the SETAs that work within industrial sectors will have clear measurable plans that are consistent with government plans , it will also ensure that data being sourced and analysed is consistent in order to make skills planning simpler but more reliable.  The ‘cons’ are that in a government driven system a clear line of engagement with stakeholders needs to be clarified in advance. Business for example has become critical of the skills system because of governance matters however, it should be noted that business has not contributed much in identifying the potential challenges it faces and the reality of its own restructuring processes that undermine attempts at improving wages of workers and creating more employment in the economy. The antagonisms that already exist could sharpen in a context where the skills system is seen to be purely dominated by government.

Secondly, another proposed change to skills policy is the introduction of a new language of occupations rather than that of skills. The argument in the proposal is that occupational teams should ensure that broad generic interventions are developed and capacitate service providers in the specific occupational areas of training.  While on the face of it these proposals are plausible what is clear is that the dimension of power and control over the process of training is critical. Occupational teams would work in an autonomous way driven by experts. This would weaken the capacity of the government in technical issues while strengthening the role of independent providers. In a way re-running the narrative of the period between 1998 and 2008 where skills training was in the hands of private providers resulting in government merely playing a regulatory role.  Another complex system with bureaucratic processes would emerge from these changes with the actual impact not very clearly identifiable given the layers of institutional processes which would be created.

Thirdly, changes to the funding model propose that 80% of the current SETA discretionary grant be located in the National Skills Fund (NSF) with the SETAs automatically receiving funding to cover the allocation from the grant, which covers their administrative costs. In order to appreciate the complexity of the proposed changes to the funding model it is important to reflect on how the current model emerged and what logic informed it. The introduction of the 1% levy in 1999 to fund skills development was revolutionary in the sense that it was aimed at enhancing the role of employers in skills training while ensuring that SETAs were funded adequately. The levy provides funding up to the amount of approximately R5 billion per annum which ensures that skills development is funded outside of the normal taxation system.

This phase of the reorientation of skills policy in South Africa is characterized by technical changes on the structures and institutions that govern skills development. These proposals are different from those in NSDS III which were largely aimed at addressing the ‘NEET’ youth unemployment problem and reducing the dominance of private provision. The proposed changes will have far-reaching implications for the structures, processes and policies of skills development in South Africa.

This article has briefly outlined the developments within South African skills policy since the early 1990s and the constraints to realising goals of employment growth through skills training. In the first phase between 1990 and 2008 skills policy emphasised institutional reconfiguration, a shift to the SETA system and the implementation of the NQF which was shaped by neoliberal policy assumptions. The second phase of skills policy between 2009-2016 saw a shift from pure market led assumptions to a model of rebuilding state capacity to respond to challenges such as the ‘NEET’ problem of youth unemployment. The current proposals for future skills policy changes (2020-2025) suggest that South Africa’s skills dispensation would have a stronger role for the state through a combination of centralisation of authority over skills institutions and also through greater co-ordination capacity within the Department of Higher Education and Training (DHET). The key issue about this reorientation is that despite the competing demands within the skills system, all efforts need to be directed to responding to both economic development needs and community initiatives of assisting the unemployed. The deepening capitalist crisis is resulting in rising unemployment and worsening socio-economic conditions within communities. Skills development should be one way of overcoming these challenges.

By Dr Siphelo Ngcwangu, originally appeared in New Agenda: South African Journal of Social and Economic Policy Issue 67. To subscribe to New Agenda click here




1. Gazette 405055 (Dec 2016) Extension of the National Skills Development Strategy for a period of two years. This implies that the new phase of the NSDS will begin in March 2020 for another five-year cycle following the extension which starts on 1 April 2018 to March 2020.
2 The DHET has embarked on a number of campaigns such as ‘The decade of the artisan’ in order to promote interest in careers in artisanal fields amongst the youth.
Crankshaw, O. (1994). Race, Class and the Changing Division of Labour under
Apartheid. Johannesburg: Doctoral Dissertation, Wits University.
Kraak, A. (2004 a). Training policies under late apartheid: the Historical imprint of a low
skills regime in. In S. McGrath, A. Badroodien, A. Kraak & L. Unwin (Eds.), Shifting Understandings of Skills in South Africa: overcoming the historical imprint of a low skills regime. Cape Town: HSRC Press.
Kraak, A. (2004 b). The National Skills Development Strategy: a new institutional regime for
skills formation in post-apartheid South Africa. In S. McGrath, A. Badroodien, A. Kraak & L. Unwin (Eds.), Shifting Understandings of Skills in South Africa: overcoming the
historical imprint of a low skills regime. Cape Town: HSRC Press.
Lolwana P and Ngcwangu S (2015). Understanding Barriers to Youth Skills Development
and Employment in South Africa Draft Report Commissioned by NORRAG. Johannesburg Wits University.
McGrath, S. (1996). Learning to Work: Changing Discourses on South African Education and Training, 1976-1996 (Unpublished doctoral dissertation). University of Edinburgh, Edinburgh.
McGrath, S. (2002). Skills for development: a new approach to international cooperation in skills development? Journal of Vocational Education and Training, 54(3), 413-430.
National Training Board (NTB) & Nedlac (1995). South African Funding Mechanism
Research. Industry Training-Supply and Competition Phase 1: Study 2 of 2. Pretoria: Coopers  & Lybrand.
Ngcwangu, SU. (2016) A Sociological Assessment of South Africa’s Skills Development
Regime:1990-2008, (Unpublished doctoral dissertation). University of Johannesburg, Johannesburg.
Tilly, C. (1988). Solidary Logics: Conclusions. Theory and Society 17(3), 451-458.
Vorwerk, C. (2005). Occupational Qualifications on the NQF: Communities of Mistrust?
Paper presented at the Q-Africa 2005 Conference, Cape Town.

Leave a Reply

Your email address will not be published. Required fields are marked *