SA-TIED Seminar Series: Beyond compliance: rebuilding South Africa’s skills development institutions

thoughts and and our research on a set of fairly complicated issues. Um, the I've put the title of the full paper with all the publication details on the opening slide there. And um, for those who are interested in the issues, I would really um, encourage you to read the whole paper because the issues are pretty complicated. And um, you know, I'm not going to be able to go into all of it in the presentation today, but I'll try and give an overview of some of the key parts of arguments and and findings of the paper. I want to start by taking a bit of a step back. Firstly, to think about the aims of our structures and policies and look at the international research on where there are systems and models that have made some progress in achieving some of these aims. Um, then I want to show some of the reasons why we're not making as much progress as we would like to be. And a lot of it is down to some seriously flawed assumptions as well as design flaws in our institutional architecture. And my argument is that this does not mean we should throw everything out and think that designing a new set of structures will magically do a better job. If in principle we can design new institutions to do the work that we think should needs to be done, we can also fix the current ones. And I think one lesson we really need to take from the last 30 years is that institutions are not easy to build. So, moving on to the skills system, I think that there's a lot of noise that there's been a lot of media coverage, there's been a lot of debate in the public sphere about fixing our systems. And for me as someone who has been researching skills formation and skills policy in South Africa and internationally for many years, over 20 years, what's frustrating about a lot of the noise that we hear in this space is the constant reinvention of the same solutions. Often by very well-intentioned reformers or policy makers or commentators who don't realize that they're simply recreating a different version of what we already had. And the reason this is really important is because what we already have hasn't worked the way we wanted it to and we need to understand why before we just jump towards um reinventing things. You know, an example that always sticks in my head from when I was working for the International Labor Organization many years ago researching qualifications frameworks. Um was the policy makers in Bangladesh with extensive assistance from international organizations changed the Bangladeshi National Council for Skills Development and Training, which was a representative body created to oversee training, to the National Skills Development Council, also a representative body created to oversee training. And you know, for me it really captured the logic and and when I spoke to people and I was trying to understand what they thought was different, they would say they were saying, "No, there's employers involved in this one and employers are now going to be change in controlling the system." But employers had been involved in the previous one and the previous one hadn't worked. The logic is always, "It's going wrong. The system is too dominated by the supply side. We need to hand it over to the demand side, create a new structure and this new structure will solve the problems." And this policy repetition can be seen in country after country around the world and it's often because we don't think carefully enough about the underlying problems, but also because we think the solutions are simpler than they are. And in South Africa today, we are living through a new version of this this same story. And so, that's why I really um I'm keen to share our research with you, colleagues. So, I mean, we can see it in in our um skills and TVET space in um South Africa, we have seen um really uh the same policy goals for well over 30 years. So, for example, um a major policy goal in our system has been the so-called um inverted pyramid, where we want to shift from um what is seen as a university-dominated system to a um to more students enrolling in TVET. And this has been policy our policy goal. It's stated over and over and over again in many different um policy forums, in many different policies. And we've never achieved it. So, you know, the question is the we I think we we really need to take a step back and and ask ourselves, why aren't we achieving it? And is it actually achievable? Um it's yet to happen. We have had reform after reform introduced to make education so-called more responsive, to strengthen employer engagement, to build capacity of institutions, and we have made some progress in some areas. And where we haven't made progress, it isn't because policy makers were trying to do something different and coming in and saying, "Oh, what we really want is a demand-driven system is is going to solve the problem." No one has been trying to build a small irrelevant TVET system. So, let's let's try start by thinking about our goals. And I think we can all agree that we want as a country, we want people with the relevant expertise and skills to do the work needed in the in the economy and society. The expertise and skills needed by employers, including the state, um and when I say the state, I'm also thinking about public works programs, work for the common good. So, not not necessarily only waged work in the formal economy. We also want individuals to obtain good employment. We want individuals to be able to progress within workplaces. Where the state pays for education and manages education institutions, we want to ensure that these programs meet the needs of the economy. We want individuals to have information about what choices to make choices about what to enroll in. And all of this requires insight into the needs of the economy. Now, that is one of the main goals of the controversial CETA system that has um absorbed so much time and energy in um in the public domain. The CETAs were set up to do two main sets of things. Firstly, to gain insight into the the skills needs of the economy. And secondly, to uh and and this is both at the level of individual workplaces as well as at the at the sectoral level. And secondly, to improve the amount, quality, and relevance of the training that is available to make sure that the demand that we identify is met. So, key to the first part was obtaining data directly from employers about their skills needs. Other employment uh other strategies to engage employers. Research at a sectoral level. And key to the second part was the idea of pooling funds to ensure that that we could actually fund the necessary training. Um and that collectively, if employers pool their funds together, we would be able to build the institutions that could actually provide the training that would meet their needs as a whole, and and would go far further than what they could simply do if they were doing individual on-the-job training themselves. There was, um, of course, an additional role in for the SETAs initially of quality assurance, which was later partially removed. It's a complicated story for those of you who don't work in the sector. I'm not going to go into it today. But, I think, you know, we can agree that the basic logic was sensible. We need industry-wide coordination. The levy grant mechanism is a common one. Um, but yet, there are many problems in the ways in which the systems have been designed. The TVET colleges are not new institutions. So, the SETAs were created, um, just under 30 years ago. The TVET colleges have been in existence for much longer. But, over the past 30 years, we have subjected them to an enormous amount of reform. We've changed their institutional form. We've merged them. We've changed their name. We've changed their core qualification offerings. We've added new offerings. We've changed how they're funded. We've changed how they report. We've changed how they employ staff. And all of this, arguably, um, with quite a lot of contestation and perhaps insufficient agreement, um, nationally about what TVET is, what it can and should do in South Africa today. And we know that the college system as a whole is very patchy. That, um, some of the colleges are doing well. Many of them are struggling. Uh, throughput rates in many of the qualifications are quite low. Um, and many of the reforms, and I'm going to go into them have not helped them to become stronger institutions. So So, those are the two sets of institutions which are both key to our skills problems and our skills needs, our skills challenges um that I'm going to be talking about in this presentation. Both institutions, I argue, have been very have been undermined by an oven and and a very complicated set of rules and tools of an over-engineered regulatory framework, which is derived at its core from new public management. For those of you who don't know what that is, it's a it's an approach to public sector governance or public or managing the public sector, which sees the role of the state primarily as arms-length governance and quality assurance and focuses on contracting out as many goods and services as possible. So, rather than building institutional capacity, successive policy interventions inst- have when where we see that things aren't working, we add more bureaucratic compliance mechanisms. We add more regulations. We add more requirements. We add more incentives. We add more reporting requirements onto fragile institutions that quite frankly are just not coping. So, my argument is that we need to think systematically through what are our goals, which are the institutions, mechanisms, and policies most likely to be able to achieve them, and then make very careful, slow changes to our current institutional landscape without thinking if we abolish this, replace it by that. We've seen some of some some recommendations like that in the in the public media. The methodology of this particular paper um is a is a is a range of things. So, I did an extensive review and analysis of the international research related to some some more directly related, for example, on sector skills count bodies internationally, but also the broader literature on skill formation systems. I did an extensive review and analysis of the research that has been done on the seaters, on the skill system, on the colleges, um including [snorts] of course much of our research at the Vits Centre for Research in Education and Labour. In particular, some recent research that we've done in manufacturing companies, previous research that we did in other sectors, and this is all detailed in the paper. Um as I said up front, I have been working in and researching these systems for many years since their inception. Um and so, I also drew on many years of engagement in policy processes, reviews, dialogues, and so on. Um and also at the beginning of this year, we held two forums with some key stakeholders and we tested and engaged on some of the ideas um presented here. >> [snorts] >> So, yeah, and as I said, it's it's it's all in the paper. And the question, I think we need to foreground, and particularly this is what I looked at in the international literature, is why do countries struggle so much to build the skill formation systems that they want, that they are committed to building? Why does reform after reform claim to be shifting from a supply-led to a demand-led system, and this is never actually achieved? And it's not a South African problem, colleagues. I can absolutely assure you. And one of the main reasons for this persistent difficulty that policy makers experience is that while we try and reform things by changing the skill system, actually, it's mainly or in many cases, factors that are outside of the formal education and training system that shape the ways in which skill formation happens. Skill formation is not something that can happen outside of economic and social systems and dynamics. It's endogenous to them. Um for example, the nature of labor markets and the extent to which labor markets are structured around occupations or individual jobs has a huge impact on training. The extent to which um jobs are long-term or short-term has a huge impact on training. The nature of industrial relations, the extent of the of of of of public welfare, the nature of corporate governance because in countries with more patient capital, you see much better training outcomes. The nature of the general education system and this obviously relates to public provision and of course state capacity. Active states can coordinate training. Weak states um struggle and all of these elements reinforce each other over time and and this is also why it's very difficult to you know policy makers around the world and we are no exception constantly go on study tours to Germany or Switzerland and come back and say to their department, "Oh, we want we need a we need the German system." And you know, we really have to understand that the German system is embedded in the German economy. So, you know, there are countries that have achieved some successes and of course we all know the most well-known ones are Germany, Austria, Switzerland, um where they have this, you know, very famous dual apprenticeship system. In Switzerland, up to 65% of upper secondary students on a vocational route, strong occupational labor markets, strong employer engagement and coordination on training content. But [snorts] also other factors, industry-level collective bargaining which enables workers, you know, salary stability, companies don't have to poach workers from each other and so on and so on. So, um I can't go into all of the factors. There's a huge body of research on this. But what's important is dense employer coordination, long-standing occupational bodies, sustained state investment. It's not You can't just export one little aspect of the system. We have seen some success factors in other countries. So, in South Korea, for example, and Taiwan, uh Hong Kong, um to a limited extent Singapore, um we we you we the there was a very clear example of developmental states coordinating rapid industrialization and directing the skill system to match that industrialization. So, here we saw very state-led, sector-specific, linked to industrial strategy that was actively being driven by the state, very strong state capacity. Um in Latin America, and you know, Brazil was the leading example, starting from the 1940s and it it it was kind of in the system was at its strongest from the '40s to the '80s. Indus- industry actually got together and at a national level pulled funds, created national institutions that provided the training that the rapidly growing industrial sector um needed. And the you know, they offered training to a population that was moving from the rural areas into the cities, hadn't didn't have a strong educational background, weren't able to access general education, and um were therefore, you know, this this this had a a very strong appeal to them. So, it was built during a very specific economic period and has faced challenges subsequently, but these are these are examples of countries that have at at periods at moments in time, some of them more sustainably than others, built vocational education and training systems that really seem to have worked. This is just honing in on the German system, which in the lit- research literature is referred to as collective skill formation systems. And, you know, these These are the systems with the strongest vocational education systems. Large percentage of the upper secondary youth cohort, as I said, in Switzerland up to 65% In Germany, it used to be more. It's It's It's gone down a lot, and it may well go down more now that they are deregulating their labor markets. >> [snorts] >> Um And so, what you saw in these countries was a good school system with fairly equal outcomes, and the majority of students who complete junior secondary school move into an apprenticeship-based secondary schooling pathway. Much smaller component going to an academic pathway, and then to university. And [snorts] the students who are in the apprenticeship-based pathway are both in a school and in a workplace. They're being trained for occupations that have fairly strong occupational norms and identities. In Germany, a more regulated labor market. Switzerland, less regulated, but still very very strong occupational norms and um education programs developed by strong occupational communities. Of course, those economies also have less desirable jobs, and some people do go from the school system straight into those less desirable jobs. And, of course, they also have a migrant population. Um And that is now, of course, growing with climate change and other um kinds of refugees. So, this this picture that I'm showing you now um is changing a lot in Germany. The The contrary to this is what we see in the wealthy, um, English-speaking countries. And consistently, they have much weaker vocational education and training systems. Um, this is where vet systems are weak. This so-called skills matches and skills shortages are seen as persistent. >> [snorts] >> Um, employers do not coordinate on training. There's less collective bargaining. Certainly, much less collective bargaining at an industry-wide level, which creates collective action problems for taking on trainees. Because, you know, all all employers know that other employers are likely to poach from them. Lack of occupational labor markets and license to practice requirements for many areas of work means that there's no agreement on the on core expertise and skills for these areas of work. Little incentive for individuals to train. Um, and so, individuals tend to enroll in as much general education as they can. You see a much larger university system, up to 60% even getting to 80% in some countries. So, whoops, sorry. Um, So, you know, this is just a, um, kind of two, um, pulling out the kind of core factors which lead to these systems looking the way they do. And they're not factors that are internal. They're not factors that are amenable to change by an education and training minister. Um, [snorts] they are very much embedded in other parts of the economy, the labor market, the, um, you know, the the the pub- the political system, and so on. So, in the coordinated more more coordinated market economies, Germany, Austria, Switzerland, etc., as I said, lots of coordination from employers, social partners, etc. Um, and and the the inverse in the more liberalized markets. And unfortunately, the broader set of economic policies and imperatives that underpin the larger model have been predominant for over 30 years, continue to dominate in many cases today because, you know, so we see many countries continuing to to deregulate their labor markets. Germany are discussing doing it as we speak. Um perhaps it's true, we can debate whether or not so-called flexible labor markets are good for economic growth. That's a separate debate. But what is indisputable from a training point of view is that they make for training very difficult and lead to employers being frustrated because they can't find the skills that they need. So instead what policy makers um the real policy response, starting in the UK in the '80s, was a set of reforms to skill systems, perhaps based on the assumption that the only correct approach to the economy was deregulation, liberalizing, marketizing as many services and goods as possible. And so reform the skill system, built um employer coordination on builds builds um organizations to coordinate an employer perspective on skills and training. So this is where we first see the emergence of sector skills bodies, get employers to specify what they want from education programs. This is where we first see the emergence of competence-based training with a view to informing education institutions about program decisions and curriculum design and to shape funding. The idea was that if the state makes funding available in the same way to public and private institutions and we in stop talking about education institutions and start talking about providers as if every service is just the same. Um and if funding is allocated for the programs that employers have agreed that they need, education institutions will become responsive. So this is a package of policies my colleague Simon McGrath and I have done various pieces of research on it and we call it the global toolkit. It moved from the UK to Australia and then to many low and middle low and middle income countries. We were an early adopter and it very much underpins the original design of our seaters as well as our approach to funding colleges and it's not a set of policies that has great success anywhere, colleagues. It did not work in the UK. It's had enormous problems in Australia. There aren't any examples of it really being very successful because it assumes that you can recreate the through a set of institutions that the underlying logics and relationships that you have in those collective skill formation countries. It assumes that employers know what their skills needs are and that they can easily articulate them. It assumes that if you create a sector skills body to to coordinate employers, you've done your job. You've created the body, the body is at a sectoral level and so that's it. Employers are now coordinated. It is It assumes that qualification frameworks solve information asymmetries. It assumes that market discipline makes providers responsive when actually it strips them of capacity for cross subsidization and long-term planning. It assumes that a state that fails to deliver training is capable of monitoring and regulating the quality of training. Now, you know, all of the these problems have a particular they've you know, we we took the set of policies and they have a particular manifestation, [snorts] a particular set of problems in how we implemented them in South Africa for for a range of reasons. These are just a few of the key challenges that we find in South Africa. We don't have occupational labor markets in many cases. Vet is very stigmatized. Employers, with some exceptions, don't really coordinate with each other um, and we all know that we have a lot of problems with state capacity. Now, I want to, um, hone in a little bit to some of the specific mechanisms because otherwise I'm going to run out of time. So, [snorts] that that's the broad perspective from the international literature and I think it's I it's really important because, um, I I really want to emphasize that, um, the the the [clears throat] the the challenges are difficult to solve and they're particularly difficult to solve at the level of the skills system. Um, we took those institutions, that that model, that global toolkit, and we made it very, very complicated. So, for example, we created a workplace skills plan that employers have to submit their training needs on and a report on what training they've done. Um, and that form is very complicated, the ways in which they need to report on it is complicated, and it creates perverse incentives because they need to report on train the the training that they do against the training the train the they reported skills needs, and so they tend to list skills that they can already address. Um, they tend to withhold information about longer-term needs because they don't want to be held accountable for things that they may struggle to deliver. Um, SETAs create what the so-called funding windows where employers and providers can apply for training against so-called scarce skills. So, employers list the skills that they the that they want to do training against as scarce, even when they tell us as researchers that they're not scarce at all because they want to be able to access training. Um, sometimes they want to be able to access, you know, learnerships as cheap labor at all. Someone has a mic on. Um, but but certainly we see many ways in which this data, which our systems were designed to get, reflects gaming and not the real needs. There's problems in terms of the organizing framework for occupations, which wasn't designed around how companies think about skills. Um and um this data then gets aggregated by SETAs. They supplement it with other data. They do research. They look at labor They look at um you know, the labor force modeling. They look at trends. Um they look at so-called change drivers in their sector, and they submit reports to the Department of Higher Education and Training. And their reports are evaluated by people with actual expertise who who are using a set of tick boxes. And so, the whole thing becomes a big compliance exercise that is defeating our central and really important goal of gathering information from employers about what their skills needs are. If we turn to the other side of the equation, so remember I said at the start, one of the key roles of these institutions is to get information from the from employers about what skills they need. The other side of the of the picture is to put push funding into the training that's needed to meet these needs. And here we also see And and again, it's beca- it's a lot of it is in in the detail of how the systems are designed that the that this isn't working. Very high transaction costs. Um it's hard for employers to So, employers have to put in applications to the SETA. They want to run a training program. They have to find the candidates. They have to find a provider. They They have to jump through bureaucratic windows. Um sometimes providers put in applications to SETAs. Only those with a lot of expertise in SETA systems are going to be able to make progress in that. The funding window model assumes that we have the institutions that we need and all of these institutions have the capacity that's needed and they can just apply for funding as and when there's a pot of money available for a particular program that has been designated as as important. Informal training at a company level can't easily be refunded to companies. I I know that there's a policy move to change this and increase the mandatory grant which I support. Um a key key issue with the funding window model is that funding flows to programs, not institutions. Short-term allocations per program. It's very difficult for institutions to plan. It's very difficult to build capacity of training providers. And so what we end up seeing and this has been in the media a lot is an underspend in in what is viewed as a skills-starved economy where whether it really is skills-starved or not is a is a separate debate because honestly, if you look at our educational achievements of the past 30 years, I think our education system is has improved much faster than our economy has. Um so our systems for skills anticipation don't work very well. Um again, and I'm just going back to the international literature here, we're not alone here. You know, we've just done a a large review of what countries are doing, what international organizations are doing in the area of skills anticipation. And there's lots of methodologies out there. There's new and exciting and people are using AI and web scraping and there's all sorts of methodologies that have been available for many years. Um what what there's almost no evidence of is how they are used. There's almost no evidence of how what kinds of educational planning decisions different forms of data actually uh are used for. And there's very little evidence from anywhere in the world that they've actually helped countries to to fix their so-called skills gaps. And and there's a lot of reasons why this is the case. It's not something that's easy to do. And this is partly why it's so important to not think to you know to to move away from this kind of factory model. If we can just tell the education system on a regular basis, these are the things you need to do. We just find out from the economy, ask the employers, tell the education institutions they can just um change their production line for the year. >> just juxtapose the view of finance and um and muted. You can go ahead. >> [laughter] >> Okay. Um people are clearly multitasking >> [snorts] >> while they're listening. Um yeah, so just just to foreground that there are limits to what labor market intelligence can tell skills planners. This doesn't mean that we should abandon the idea of getting insights from employers. I think it's really important. I think it's one of the key strengths of our system. Um we should not abandon a built-in mechanism for doing this. Employer surveys are really expensive and there's all sorts of methodological problems, but we must start by dramatically simplifying the tools that we use to do this. Of course, supplemented by other tools, but the most important thing is much more careful thought about, "Okay, once we've got this information, which kinds of educational planning decisions is it going to inform? Is it going to Are we going to push more money into employers because they need to do urgent upskilling? Are we going to do more long-term planning? Which of Which of the data sets help us to make which of these decisions?" And I think we we really go wrong there. The The next serious problem with the institutional design is the governance model of the SETAs. Um where this um a confusion between fiduciary oversight and the industry voice. So, we kind of assumed that we would create these bodies and industry players, union representatives, and employer representatives would sit on the boards, and that would make them you know, demand representing the voice of the economy somehow. But yet, these boards also are focused on fiduciary oversight in a very complex environment in which they have to report on a whole series of very highly specified um tools and and and mechanisms. So, clearly, these boards are not sufficient to create industry leadership. There's often vested interests. Um we've We know that there's corruption, and we know that there's large, costly staff bodies, sometimes without sector expertise, sometimes with a lot of sector expertise. And you know, we'd also really emphasize that the SETAs are a very mixed um bag, and some of them have have had a lot of successes. And I think that the problems that we see in them are very much related to how they've been designed. An- another problem is the fact that they were all designed exactly the same, whereas the sectors of the economy are very different. Um then we've seen intense micromanagement from DHET, and this really relates to um government priorities overriding um industry priorit- priorities creating alienation from employers in these bodies, whose prime role is to engage with industry, and to be the the kind of intermediary between um industry and the education and training system. Um and every time something has gone wrong, we've added more layers of reporting requirements. In terms of at a national level, there's duplication of structures, structures with unclear mandates, etc. I'm not going to I won't get more into that. Um, [snorts] the TVET colleges, you know, I think we've made their job really very difficult. Um, they they have been up till now offering three different sets of qualifications. The NATED qualifications were the qualifications that were the theory component of the old apprenticeship currently being phased out. The National Certificate Vocational, which was introduced when when it was first announced by the Department of Labor, it was the Department of Labor back then. Um, that they were going to get rid of the apprenticeship system. Um, at that point the then Department of Education started developing the National Certificate Vocational, which is a general vocational qualification. The idea of it was that it's not narrowly designed at one at at specific occupations, but but gives um learners math, science, etc. focused on areas of the world of work. And then the new occupational qualifications, which I'm sure um those of you who follow these things have seen a lot of noise about in the media. Um, there were there were articles in the Sunday papers, um and so on, which are currently being phased in, but um in ways that uh that we we we might want to express concerns about. So, just to wrap up, um in in my analysis in in my analysis, having worked in this in the system, but also having researched very extensively on a range of different aspects of this system. Um, we've I've put a a number of recommendations into the report, but this is just a few of the headline ones. Dramatically simplify the workplace skills plan and annual training report requirements, preferably into one report, focusing on what are the most useful things that we can ask employers that they can actually answer on and that we can use. >> [snorts] >> Um diversify the models of the SETAs. They don't all need to look the same. They don't all need the same governance structure. Um some could be employer-led, some could be state-supported. Make employer engagement and and union engagement where where in sectors that are unionized the most important performance measure. Real participation, not not submission of forms. Um in in some of the SETAs that we've worked with the they the SETA can't even give you the contact details of the of the employers in their sector. Fund strategic provision and and institutional capacity and don't just make funding windows available. Um and as I mentioned already, separate the mandatory grant from the the data collection incentive. And um you know, this is a maybe it's a bit of a finicky issue. It is explained in the report. Um The TVET colleges, I think um both the TVET colleges and the community colleges need to reflect local economies and and have different mixes. >> [snorts] >> We need much more state support, particularly for work placements. I think, you know, that and I know that there's a team of colleagues working on these issues trying to figure out how how this roll out of occupational qualifications is going to unfold. I know it's a it's a tricky issue. But for me, the most important thing is with these institutions, we shouldn't treat them as private providers that, you know, they must just that come and apply for grants and then we give them grants because they don't have the capacity to do it and they never will. And actually, the private sector doesn't step in and offer um vocational education in in because vocational education is expensive and difficult to offer. And so, really, that's what you see in many countries around the world. It it just it just doesn't exist. Where the state really doesn't play a very strong role in building it. So key takeaways our problems are not unique. Many countries have struggled. Skills anticipation can't substitute for building institutions. The the particular models of the global toolkit and the the kind of rational of new public management isn't what vocational education requires. Central coordination is really important. We should not get rid of it. But also vocational education is not a mass solution to youth unemployment. It's expensive. It's difficult to get right. It's really important. And and and so we need to really think through where it's needed, what's needed. The the massification of the TVET system has not made it easy to offer high quality provision. And and the system was struggling before that. So that's just my research center. The center for researching education and labor. I've put a few references to some of our work, some of the other work in this area. And of course all the detailed references are in the slides. Sorry, in the in the full paper which I really do hope that colleagues will read. So I'll I'll stop sharing now and hand back. >> Thank you very much Stephanie for that thoughtful and important presentation. I think you've given us a great deal to think about particularly around the limits of compliance driven approaches. The institutional challenges facing the skills development system and what meaningful reform would require in practice. So thank you very much for that. We are now open for Q&A. I would ask that if you have a question you please raise your virtual hand. I'll look out for the hands. Alternatively, you're welcome to type a question in the chat box. Are there any questions from the floor? Okay. >> An an an interesting one from Tunde in the in the chat. I don't know if someone >> I can read it out. Um so Tunde says a lot of public debate around skills development ask whether programs work. Your paper suggests that this is much harder to measure than it appears. What kinds of evidence would give us more mean a more meaningful picture of whether the skills levy system is contributing to institutional capability rather than only measuring output such as enrollments, um completions, or short-term employment. >> It's a really good question, Tunde. Um and I'm not sure if I am the right person to answer it. I mean, I'm very pleased that my colleague Carmen Morack is on the call because she's an evaluation expert. Um but, you know, I I what I would just say is that I think um we do need to the what works discourse in relation to TVET and skills programs is a very limited discourse because firstly, the cause the causal chains are almost impossible to unpack. But secondly, you know, a program might work in the sense that a few young people got employed, but we don't know whether other people got pushed out of the labor market. We don't know Did the company become more productive? Um did the com Was the company able to grow? We tried to test that in a research project. It was very, very complicated to do. But I think um Tunde, what I would really say is that the the point that you are raising is one that we need to have very serious conversations about and really try and think through more systematically and more carefully in in the design of our systems and in terms of how we evaluate our systems and when we evaluate our systems, what are we looking at them for? What are we expecting them to do? I don't know if Karma wants to comment on that at all. >> Karma, would you like to comment? >> Um no, no, no, not not right now. I'll come in. >> Okay. >> Um later, but it's useful. >> All right. Thank you. We have one from Okay, let's start. Ivan, I see you have your hand up. Please come in. >> Uh thank you very much for a fascinating presentation. Uh my interest is really around cities and places. And I'm struck by um I'm not an expert at all on the skills development system. But um I'm you don't talk much about the centralized nature of of the system. And I'm wondering whether, you know, the focus on sectors the emphasis on sectors rather than the spatial dimension, places, and particularly local uh the locality, is is part It's not a panacea, obviously, but is it part of the way forward? Because it it ticks some of your boxes around better matching of supply and demand, employers and local training TVET colleges, um a kind of much stronger role for our uh local institutions. I hesitate hesitate to say municipalities we cuz we know municipalities already struggling to deliver their their existing mandates. But if they had some kind of role at the local level to to to encourage participation, to strengthen relationships between uh institutions, and to try to build the sort of networks and the the systems and the relationships that that are so I think so strong in Germany and other places that you didn't mention. I wonder whether that would be part of the of the way forward trying to move away from these very centralized entities to more um more of a devolved approach. >> Yeah, I mean I think I think you need Yeah. >> Just before you come in, should we take I see we have a hand up from Tanya and Nangamso. Should we take those two questions and then you can just respond because there may be some overlaps. Are you happy with >> I'd I'd rather respond one-on-one because the questions are quite complex if you don't mind. >> That's fine. >> I mean I I I think Ivan you're the I I don't think it's an either or. I think that we need a much more place-based focus and I think that that's where the that should be the logic of managing the colleges. I don't think we really need a distinction between TVET colleges and community colleges. I think that the colleges, you know, the advantage the strength of the colleges is that they are located they're so widely spread in rural areas, in you know, in across municipalities. The the TVET colleges, to some extent, the community colleges, to the extent that they exist, the idea is that they really are there in every municipality. And so I think that there should be a very very strong role for place-based skills development and engaging with the local economy. I don't think that that should be an alternative to sector-based coordination because you you have you it doesn't help you to think about how to train occupational workers. You know, um you need um the the training needs of a plumber or an electrician, of course there's things that they need to understand about a local community, but you also do need to have occupational coherence across those kinds of occupations. And so, you know, and and you know, the dairy industry for example, it is quite place-based. We don't have dairy factories everywhere. They're they're mainly in in very specific places, a lot as we know in the Western Cape, KZN. So, even there, even at a sector level, at an industry level, there is a place-based component to it, but I think the sector the sector dynamic and the sector role is very important. What I do also really agree with in terms of what you said is we need to move away from over central prescription because at the moment we're not allowing each industry to follow the logic and the needs and the dynamics of its its own industry. Um there's a lot of complicated questions in the chat, uh but I I think we want to but I would love to talk about it further. We've done a fair amount of work on skills ecosystems and it would be really good to actually talk about these issues with you. >> Good. I see a thumbs-up from Ivan, so I think he's covered. Um we'll take the other two questions and I'll I've seen the chat as well. Tanya, you're welcome to unmute yourself. >> Thank Thank you so much and thanks for a great presentation and I'm really looking forward to reading the paper. Um the area that I want to just hone in on is um a point that you've made, but I just wanted sort of take it to the next level and understand what the solution is. Um it It seems you know, you mentioned that um many times it's factors outside the skills system that are challenging. Um I think one of the areas that we see a lot of is that the skills um system doesn't talk to the industrial development system, and the new industrial development strategy has just been launched by the DTIC. Um it references skills that quite superficially, and there certainly hasn't been this meaningful interaction and engagement between what this industrial strategy is seeking to achieve and what is the skills formation that's going to be required to accompany that. And I'm just wondering, you know, what are the institutional mechanisms that can be strengthened in that regard because the skills institutions sit separate from the industrial institutions and structures. We did have in the last administration master plans which could have served as a sort of bridge between them. But those master plans didn't in most instances. I think in global business services you probably have your best example of where they they they really did try and sort of identify the skills needed to develop the sector. But other than that, you didn't really see it. And the master plan system isn't really that clear in terms of where it's going now. So I'm just wondering, did did you when you sort of looked at augmenting institutions so that they are better designed to deliver what they mean to? Did you think of of how the current systems could be augmented and relate to each other? Thanks. >> Um we I didn't look at it specifically in this paper. I do think that it's an absolute priority. And and I and I think that it's a it's a tragedy and a travesty that that you know that this isn't happening when we've got these structures that are set up to be sectoral. Part of the problem is that the um the demarcation of the seaters is a little bit awkward in some instances. I'm not saying necessarily redemarcating them. I know it was a long process and there's a lot of debate. But you know, for example, agri-processing um is a crucial part of industrial strategy, but it falls into two seaters. Um etc. So um Yeah, I mean I think that um the the master plan process I know that some colleagues from the DHD did participate. I know it was difficult and as you said it was business processing where they where they made the most progress. Um [snorts] my view is that the SETAs have been were were set up and they were given this like little set of prescriptive things that they have to do and they have to account to the board on these things. They have to gather this data, submit that, submit and so they're so busy doing those things, doing the the tasks, trying to meet the incentives, trying to meet the annual performance plans that there's just no focus on the actual goals of the system. And and I think that that is a problem that we see in in many parts of the state where we're where where we we we have sort of governance mechanisms going wrong. So I would say that if we are rethinking the structures of these institutions and make and and building in a stronger role for industry, it also really needs to factor in exactly what you're raising Tanya. But but I haven't looked at exactly how that could happen. >> Thank you Stephanie. Nangamso and Shirley, nice see your hands. Um but I think let's take a question from the chat. We have a question from D Pariachi, um Stephanie. And Prof, where would you see the most immediate reform being bene benefited, lowest hanging fruit in the educational system is the question. >> The things that I would start with, um I would separate the annual training plan from the workplace skills plan. I mean it's a very simple thing. You could just separate them and and make them not count the one against the other and you'd get better data from employers. I would dramatically simplify the workplace skills plan. Maybe it's not that simple, but some people who've got good experience in survey design, firm survey design. We've got a lot of ideas on it that we'd be willing to share. Um you could just start getting better data from employers. I would change the funding window system and um uh think about what the priorities are for different sectors and really let the employers and and unions where where there are are active in the seaters to to make decisions on um institutions institution building, you know, that that are priorities. For example, um you know, it might be building a TVET college in a particular local area to offer a particular program where a particular uh close to an industrial park which is being built. It might be um proverb supporting a private provider in the dairy industry or the automotive sector where they're quite concentrated and the employers do work together and they do have private providers um that operate across. Um So, I think that there there are some quite simple reforms that could be made while we try to see how should we actually change these institutions to make them more productive um and support the goals that the of the system. Um I do think that we need to give the colleges a lot more support particularly in this roll out of occupational qualifications. Um and I know this I know that's a that's a separate set of issues, but um I think there there are ways in which the Department of Higher Education and Training could be giving them a lot of um a lot more support. >> Thank you. And Nqabisa, you're welcome to come in. >> Uh thank you so much um and thank you for the presentation. I think for me what some there's something that keeps on coming up and you've also been mentioning it. It's about the bureaucracy, uh the regulation, the prescriptions, and you've even gone as far as saying boards are not being effective uh because there's so much prescription and tasks that they need to do. So, what I'm not understanding is why is this so hard to fix? Is it not getting into the right ears who are decision-makers? Is the research um relating to this not getting to the right ears? Um because, I mean, where we are with it's South Africa and the youth, there's so much unemployment. It would be sad that one of the reasons is because of of too much regulation. Why is it such a struggle to deregulate? Like, why is it so hard? Where is the bottleneck in your opinion or in your research? Thanks. >> That's [laughter] a very difficult question to answer. Um [snorts] but it's a great question. I mean, I And and and actually, I would say it's not something that um that we've actually really researched. So, I have my views. I have opinions on it. Um and I think it would actually be really useful to to talk to people about that. I mean, I know one of the problems in this whole set of issues is that if you want to make changes to the Skills Development Act, it's got to go through Nedlac. And so, people have tried to make even quite small changes like the DHET did try sometime back to change the Workplace Skills Plan and simplify it in ways that I think would have assisted. I think they could have gone further. Um but many of the many of the um discussions just get completely deadlocked in Nedlac. So, I think that's a that's a tricky area. Although, I think having said that, there are some aspects of this that could be changed without legislative change, but some of the some of the changes that I think are needed to the system would require changes to the legislation. Um and, you know, Nedlac is is not always um a harmonious body. We have a a lot of conflict at Nedlac and um yeah, that's to go back to my very opening point from the international literature, >> [snorts] >> conflictual industrial relations make it very difficult to build skill systems. >> Thanks, Stephanie. Lillian, I see your question. Just please give me a thumbs up. I feel that Stephanie answered that in part when she was answering the Paliatsos question. If you're covered, please just give me a thumbs up. In the interim, Sherilyn, you're welcome to come in. >> Thank you so much. Thank you, Stephanie. My question relates actually to the last two intervention or the last two questions, last three questions. And I'm trying to hold on to some hope that there is a link between skills and youth employment. And maybe we're just not dealing with coordination friction. Perhaps we're not looking for appropriate skilling. Is there a link? And if you if you think it's worth spending time on, what would you do to to drive youth employment from a skills base? >> Yeah, I think there's definitely a link. And I think that the skill system has a really crucial role to play in a broader mix of policies that are focused on youth employment. But I don't think you can drive employment through skills. And I think that I think that approaches that try and do so generally don't work. I mean, we we are just launching a community of practice and we're trying to share ideas with all the other researchers who are working in this space. And and hopefully we will come up with some ideas. We you know, we're really trying to coordinate across all the bodies and and and sets of people who are working on different aspects of these problems. Um I think that, you know, there's obvious roles for targeted short-term skills. There's obvious roles for good long-term provision. There's obvious roles for second chance matric. We know that many young people leave matric with very, very low levels of language and um literacies. Um Shireen Motala is saying, you know, even the the the students who are moving through that TVET programs have very poor quality. And and many of those students have actually Most of them have finished school. And then they go to the TVET college. And then they still have um very weak outputs. So, a very weak educational levels. And you know, if you speak to many employers, um what they will tell you is that they are really looking for people with good language skills, people with good math skills, um etc. So, I think, you know, there's a there's a huge role for second chance um education programs to help the kids. And you know, it's not something that Treasury generally likes because um it's expensive. And you know, people don't like hearing things that sound inefficient. It is inefficient to be um uh providing training to be providing education programs to people who have moved through 12 years of education. It's not great. It's It's very unfortunate that that's what we need to do. But we do need to do it because many of our young people leave school with very low levels of educational achievement. You know, our education system has improved, but not nearly fast enough. Um So, I think there's an enormous role for education and training, but I do not think it can drive youth employment. And I don't think that a mass TVET system will lead to employment. There's just no evidence that the main reason that people are unemployed is that they're lacking skills. You know, obviously there's there's there's needs small patches of targeted needs and and you know, I think we could be much more creative and this goes to Ivan's question about building um skills um skills components into community works programs and really finding ways of supporting municipalities to organize people to do work for the public good um and and get some useful skills as part of that program as part of that process. You know, there's things that can there's there's a lot of things that can be done in that space. Um even if it doesn't lead to formal employment, but it could lead to better outcomes in in community work programs, in public works programs. But it's just not the case that if we suddenly magically have a mass TVET system um those young people are going to walk out of that into jobs. So, I think it's it's a it's a complicated picture and it has to be located in the broader policies and systems for job creation. >> Thanks, Stephanie. I see we have a a lot of engagement and there are many questions coming through on the chat. We may not be able to get to all of them in time or in the time that we have left. Um so, for any questions we're not able to address today, please feel free to continue the conversation with Stephanie directly after the session or you're welcome to reach out to the every tight team and we'll do our best to support a follow-up. Um but before we close, Stephanie, are you open to taking one more question? I see Nazreen has her hand up. >> Sure. >> Okay. Thanks, Stephanie, and I know we're over time, but uh you know, you've made the point several times that the TVET system is not sufficiently funded, but I I on the whole we actually have a pretty decently funded skills system. Uh if you take into account all the entities, uh what would you suggest or who would you suggest we hold accountable for actually looking at a combination of demand-led skilling, but also noting we have to support the supply side because there are young people who need some training. Uh but how do we use this money effectively? And you you know, you you'll know in the research we tell people, international colleagues, the the quantum of funding we have in our system, and they're always shocked. Um so, how do we use our money efficiently and effectively? >> Yeah. Look, I I think that Nazreen, you know, you've just asked like one of the most difficult questions right as we are over time. >> Sorry. >> [laughter] >> But uh um it is true. There's money sitting in the seaters, and you know, part of the whole kind of um reason for the public outcry about it is because of unspent funds. Um but the to get that money out of the seaters, you've got to jump through hoops, you've got to apply against particular funding windows at a particular moment, you've got to put an application in. So, I think that this the state needs to allocate funding. And then um if we had effective industry level, sector level coordination, we could also then take decisions about allocating funding that doesn't require um all those application processes. Um [snorts] but I I think that for me, one of the issues is I'm then I'm just going to This is just one small answer to your big complicated question. But I think that you can't build education and training institutions through funding models that are based only on learner enrollments against qualifications. So, we have to think about because if you think about it, you know, you're a TVET college, you've got particular staff, they have particular capacities, they are all employed to teach on particular programs. New programs are emerging, none of that staff is funded to go and engage with local employers. No one is paid, you know, you don't have institutional capacity to do that kind of work. So, I think we do need to think about the the different roles and the different kinds of core funding that is needed in an education institution that is able to actually then pivot from one kind of offering to another offering, that is able to take staff out of teaching and send them on retraining because now um the technology has changed that they used to train on, etc. When institutions are operating on a kind of a hand-to-mouth um applying for them for the for these learners on this program, they can't do any of those things. So, I think part of the picture for me is moving away from narrowly Of course, there should always be some funding that's based on learner enrollments for against specific programs, and we need to find the right balances. But, I think you know, when you've got a an institution like a university that is large and can cross-subsidize over years, over time, um it's much it's much easier to to to maintain institutional capacity. But, in small, weak institutions, it's impossible. And so, what you see is that the institutions just aren't there. If you don't If you don't create them as a state, they're just not they're just not there. Even if you make funding available for people to apply against. So, I think part of the story, one small part of the answer to your question, is designing different funding mechanisms that can build the capacity of institutions that then can put in applications for the funding that is available for for specific programs, and so on. >> Thank you, Stephanie, and thank you for the very rich presentation. I think a big virtual applause to you. I'm going to Where's my reaction? There we go. Huge virtual applause to you, and uh thank you also for engaging so generously with the questions. I'd also like to just thank everyone who joined us today and who contributed to the discussion. Um we appreciate your engagement. We will be making the recording available um as well as Stephanie's slides. And like I said earlier, you're you're more than welcome to reach out to us, and we we can help address some of the questions that we couldn't answer today. Um yeah, and if you would like to join other ISACA seminars, I encourage you all to go on to the website and just register for the newsletter, and then you'll get notifications of upcoming seminars. But thank you very much, and we wish you a very lovely day. >> everyone also directly. >> All right, thank you. >> Thank you so much. >> Thanks, everyone. >> Thank you.