Stephanie Allais and Siphelo Ngcwangu | 18 July 2024
The clumsy rules for employers to get training money back actually encourages them to give poor data.
Continually tweaking behaviour via incentives, rules, and tools has led to a system that is simply too complex for its own good. Read Part Two of the series here.
Dear Minister Nobuhle Nkabane,
As academics who have worked closely with your department over many years, we welcome your appointment to lead higher education and training, convinced that you will bring renewed energy and fresh insights to an area critical to the future of our country’s young people and its economic success.
It is often said that South Africa has a skills crisis, and that our education system doesn’t meet the needs of the economy. But we have an elaborate national system that asks employers every year what skills they need.
So, why is training not meeting the country’s needs?
Simply put, the rules and tools set up over the years to consult employers and fund their training requirements are too complex. Well-intentioned policymakers have added unnecessary complexity time and again, creating too many tools to do too many things.
They end up doing none of them well and sometimes incentivise negative outcomes. Clumsy and tedious funding processes and complex, overly elaborate regulatory oversight complete the picture. The result is little coherent direction from the state in priority areas, and an inability by employers to prioritise the training they need.
Effective leadership
As the new minister, you could provide refreshed and effective leadership with a few simple changes. This is not about new policy, just simplification and streamlining.
The system was designed to understand the skills employers need within the context of specific industries, and to encourage and support them to do the required training.
Employers with an annual payroll exceeding R500,000 pay a levy allocated via the tax system to Sector Education and Training Authorities (Setas). They in turn give 20% of the money back when employers submit an annual workplace skills plan and training report.
The idea is for these documents to provide insight into employers’ skills needs and show what training is taking place. Setas and the government can then direct funds to meeting training gaps by funding areas not covered by workplace training.
The assumption is that because this data identifies the skills employers need, it can be used to formulate sectoral training planned by Setas. It also informs the list of Occupations in High Demand developed by the government, and a critical skills list that guides the allocation of certain work visas.
The problem in the real world is that current data from employers gives very poor insight into skills needs. In some cases, employers only get their money back from the Setas if their training report matches their previous plan. This incentivises them to propose only training that will be easily achieved.
The clumsy rules for employers to get training money back actually encourages them to give poor data. Some employers also only list those training programmes that will be funded by Seta. This might sound crazy, but it is the result of overly complex policy.
Labour force trends
Plans developed by Setas have two main funding mechanisms for skills provision: earmarked “funding windows” during which employers and training providers can apply, and bursaries. Setas decide on these based on the aggregated data obtained from employers, as well as analysis of labour force trends and other research.
So, for example, employers list production workers as scarce even when they are not because they know they will need to train new employees, and grants are made available for scarce skills training. This way, they get back their levy funds to do the core training they need to do every year.
Another example: where employers have crucial training by international providers often linked to suppliers of specific technology, they can’t get funding because these trainers are not accredited in South Africa. This essential training by highly regarded international providers is valued and needed by employers.
So, some employers simply don’t report skills gaps in these areas because it’s a waste of their time as they won’t get money back for it.
Broad-based black economic empowerment (BBBEE) employment scorecard reporting adds further complexity, and additional perverse incentives. Some employers report on skills gaps to obtain funded learnerships as cheap labour, with no plan to hire trainees once they have completed their courses. With no real skills gaps, training funding gives them subsidised labour.
The system does try to deal with this: the BBBEE scorecard rewards employers if they hire staff they train. The result is that large employers reduce the amount of training they do in crucial trades, because they don’t want to be reporting training above their own skills needs for fear of being penalised for not hiring trainees once they qualify.
This way, fewer young people are trained as artisans who could work in small businesses inside big companies’ value chains, and are lost to the relevant industries.
Continually tweaking behaviour via incentives, rules, and tools has led to a system that is simply too complex for its own good.
There are other problems facing employers in reporting their skills needs. They have to describe their workforce against the codes of the Organising Framework for Occupations, which many find cumbersome and hard to relate to their own organograms, job descriptions, and concrete requirements.
So, transaction costs are huge. Additionally, the annual nature of the process mitigates against long-term planning.
Equally significantly, many employers don’t participate at all. No insights are obtained from non-levy-paying members, and there is very little engagement with smaller companies, making the national data incomplete in significant ways.
As a result, the government and the Setas collect data from employers that is essentially meaningless and incomplete within the context of the overall economy, then use it for planning through clumsy funding systems.
Setas are supposed to be tripartite structures, accounting to employers, unions, and the government. So why are employers’ voices not stronger in the design of these systems?
Again, we have good intentions backfiring. In reality, Setas are ultimately responsible to the Department of Higher Education and Training (DHET), not the tripartite structures. The DHET oversees their annual performance plans and adjudicates the strength of their sector skills programmes.
The aim is positive – to make sure the Setas are doing what they should. But it does so within a complex and over-designed system that reduces oversight to box ticking instead of strategic assessment. The state is left checking allocations against designated windows, instead of making strategic decisions about economic and educational priorities.
Recently gazetted changes to the rules and tools for skills planning address little of this because they don’t reduce complexity or change lines of accountability.
Anticipating skills requirements across a complex economy like South Africa’s is not a simple exercise. International and local research shows that employers struggle to articulate their skills needs.
But some small changes could substantially improve the data South Africa collects and how it is used. For example, Stats SA could regularly assess and analyse workplace skills needs by industry and for the whole economy.
Developing insight at a sectoral level also requires understanding the skills needs of non-levy-paying companies. Industry associations could be helpful here.
SETAs could be important intermediaries between the worlds of education and work, but only if they spend more time in high-level engagement with employers on their real training needs, planning and funding accordingly.
Stephanie Allais is Professor of Education and Research Chair of Skills Development at the Centre for Researching Education and Labour, University of the Witwatersrand. Siphelo Ngcwangu is Associate Professor of Sociology at the University of Johannesburg.
‘Disclaimer - The views expressed here are not necessarily those of the BEE CHAMBER’.