Context

South Africa's latest Quarterly Labour Force Survey (QLFS), released in August 2026, put national unemployment at 33.6% (8.5 million people out of work) and youth unemployment (ages 15 to 34) at 47.4%, its worst level outside the COVID-19 shock. Among 15 to 24 year-olds, 36.4% are not in employment, education or training. (Statistics South Africa, QLFS Q2 2026).

Behind these numbers sits a longer structural shift. Over the past four decades, South Africa's economy has shifted decisively from manufacturing toward services. Today, services account for the large majority of both output and employment, while manufacturing, mining and agriculture have contracted in relative terms (Bhorat, Lilenstein, Oosthuizen and Thornton, 'Structural Transformation, Inequality, and Inclusive Growth in South Africa,' WIDER Working Paper 2020/50).

Whether this services-led path can meaningfully absorb the large volume of country's young jobseekers, and what that requires of government, business and intermediaries, is one of the central questions facing South African policymakers today.

We pose these questions to Alex Beyers, who has spent close to a decade working on youth employment from inside one of South Africa's most prominent labour-market intermediaries: Harambee.

Conversation

Q: South Africa's Quarterly Labour Force Survey for August 2026 put national unemployment at 33.6% and youth unemployment at 47.4%, its worst level outside the COVID-19 shock. You have worked on this problem at Harambee for close to a decade now. What has changed in your own thinking about how to fix youth unemployment over that time?

Beyers: It's worth acknowledging up front that this is, objectively, a depressing topic. Barring the COVID period, youth unemployment has never been this bad. A few things have become clearer to me over the years. First, South Africa fundamentally has an employment problem. On the broader definition of unemployment, that 33% becomes roughly 40%, and the youth number goes from the high 40s to over 50%. Second, in an economy with this much unemployment, new entrants will always struggle more than people already in the labour market. A million new work seekers arrive every year and they have to compete against people with five or ten years of experience. What has changed for us at Harambee is understanding how much of this is a “systems issue” rather than something the SA Government’s Department of Employment and Labour and a handful of NGOs can solve alone, and how necessary facilitated transition mechanisms, like matching platforms, are for the people furthest from the labour market.

Q: Where does Services fit into South Africa’s employment story?

Beyers: Services has grown a lot more than sectors like manufacturing and construction, particularly over the last decade. Part of that is the lost decade of the 2010s when state capture took hold and growth was sluggish. Enablers like energy, transport and utilities took a real beating, and that disproportionately impacted sectors that depend on physical operations and exports. Services was just less exposed to that damage. So, one view is that services outperformed, both in terms of output and employment, not because it did so well, but because the rest did so poorly.

Q: Community and social services and finance and business services are where almost all of the youth job growth since 2008 shows up, while trade and manufacturing have moved the other way for youth. Why did services specifically absorb young people?

Beyers: Beyond the infrastructure constraints, there is a genuine skills story. In most services roles you do not need a highly specific technical skill. You do not need to weld or drive a forklift. A call centre agent can succeed with a decent grade 12, sometimes without one, if their language skills are strong enough. That matters enormously, because only 15 to 20% of our youth have any qualification beyond grade 12. In manufacturing, mining and telecoms, you need that specific technical skill set. Services do not carry the same constraint.

Q: Global business services and business process outsourcing appear to be the flagship of the services story in South Africa. Business Process Enabling South Africa (BPESA) reported over 26,000 new jobs in this sector in 2025, about 90% of them youth, with a target of half a million by 2030. Is that target credible, and what has to go right to achieve it?

Beyers: You are right, GBS and BPO has been a leading player in terms of youth employment. I do think the target is credible, with one honest caveat: nobody knows how far AI will go in this industry. Beyond that, I think the sector needs better adoption of national matching platforms across employers. It also needs to hold and grow international demand. Lastly, the sector requires stability in the youth employment tax incentive which has recently seen delayed payouts and a debate about its structure.

Q: Are you seeing early signs of AI displacing the entry-level voice and contact-centre work that has been the classic on-ramp into this sector for first-time job seekers?

Beyers: The early indications on our side are that AI is augmenting rather than replacing. A lot of what could be automated with pure automation has already happened. What is left often requires the kind of interaction clients are not yet comfortable moving fully to AI. South Africa may actually be somewhat protected here, at least in the medium term, because of the skills gap on the client side. Someone doing this work can be upskilled with AI tools rather than displaced by them. That said, we have to acknowledge there is a real unknown. The honest answer from most employers we speak to is that this has been a live discussion for years without the numbers really moving yet, but that could look different in another three years.

Q: Trade is still the single largest employer of youth in the country, at roughly a quarter of youth jobs, even though it is one of the sectors that has been losing jobs overall. Is trade still part of the growth story for South Africa, or is it more a legacy stock of jobs that is slowly shrinking?

Beyers: Retail looks like a mature sector. It is unlikely to outgrow overall economic growth, so we have to bargain on proper GDP growth to stimulate it further. There is also some internal displacement happening. A lot of jobs have been added through last mile delivery, and arguably some of those are offset by smaller stores needing fewer staff, or some smaller stores closing altogether because they cannot compete with delivery-enabled retailers. So, at an industry level there is a bit of a zero-sum game underneath the headline number.

Q: Community and social services is the other category that has grown for youth. Much of this is public employment in education, healthcare or administration. Is that growth durable, and does it help people transition into other kinds of work afterward?

Beyers: In our view, yes, it does help, though not at the scale of creating more private-sector jobs directly. That effect is buried further down as a second or third order effect. People who have been through structured public employment programmes are more likely to find work afterward than a random comparison group. In our own follow-up surveys of a schooling assistance programme, participants were more likely to pick up informal tutoring work or find formal administrative roles.

Q: Tourism now accounts for close to a million direct jobs and has been outperforming several traditional sectors. Do you see tourism as part of the same services story as global business services, or a genuinely different pathway?

Beyers: I think it is a similar bet, for a few reasons. There is a similar order of magnitude of jobs available, and tourism probably has more knock-on job potential than business process outsourcing. In the last few years, there has been a real growth in investment in coordinated action in this sector: bringing together private sector stakeholders, civil societies, community and government. One concrete result is the Electronic Travel Authorisation (ETA), which has made it dramatically easier for visitors from large markets like India and China to get in. Most of the jobs underneath this growth, cleaning, driving, hospitality, guiding, are within reach and quick to train people for, which is why the services argument holds at scale here.

Q: Gig and platform work has also grown quickly, particularly taxi hailing services and quick commerce delivery. What is your view of this sector's potential to provide good jobs at scale, and are there constraints or regulatory risks on the horizon?

Beyers: This sector has some genuinely volatile undertones right now, partly tied up with an international migration issue. On the delivery side specifically, which is mostly motorbikes, there is a real constraint on the number of people who hold legal motorbike licenses. Getting access to a bike and then a license is a real bottleneck. Owning a bike isn’t a cultural default in South Africa the way it is in some other countries. There are challenges across the lifecycle; right from capital constraints to lack of driving skills to licensing. That gap has partly been filled, informally, by migrants who already hold motorbike licences from elsewhere, and with an election coming up, migration is becoming a politically sensitive issue that could affect this. There is a real analogy to the Electronic Tourist Authorization (ETA) story here: making access to the vehicle and the license easier could unlock a lot more growth in this sub-sector.

Q: Stats SA's own numbers show the national care economy is actually shrinking as a share of employment, with paid care work falling from 14.2 percent of total employment in 2016 to 12.5 percent in 2026. Does this sub-sector have the potential to provide mass youth employment, and if so, how should South Africa approach it?

Beyers: It is hard to see clearly what is actually happening under the hood of that number. Fewer middle-class households employ full-time domestic workers than they did in the 1990s, and where the demand still exists, it is increasingly intermediated through platforms like SweepSouth. This shifts that employment out of the private households category and into business services. On elder care specifically, we have looked at this carefully and have not seen real demand materialise on the ground yet. I think the potential is genuinely there, but it probably needs medical aids and households to shift how they pay for care before it scales.

Q: How can South Africa apply the tools of productive development policy to make its services sector more productive, not just larger?

Beyers: The classic economic model here is to find something you already have a competitive advantage in and then do that really well. What that model often leaves out is that you need a coordinating mechanism to systematically identify the blockers and remove them. We often already know the blockers at a high level, that there is too much red tape in a given sector, for instance. What is missing is a process and committed people who go through every relevant municipality, identify the specific bylaws that need to change, rewrite them, and put them forward to the right authorities. On the general business environment, there is also a predictable but real problem: starting a small business is difficult because registration, tax paperwork etc. complicated, but businesses often need to be formally registered to qualify for funding. The bigger missing element in South Africa is not the ideas or recommendations for what needs to change. It is the implementing capacity to actually go and do it.

Q: If services-led growth is genuinely the path forward for South Africa, what are the two or three biggest bottlenecks standing in the way of it scaling further, and where does Harambee specifically sit in that system?

Beyers: Services is likely to keep growing faster than the rest of the economy, but it cannot carry the whole economy on its own. We still need the primary and secondary sectors growing too, even if it is at a slower pace. There is a government-business partnership currently working on three national unlocks: Energy, Logistics and Crime. Global business services is probably least exposed to any of the three, since you mainly need electricity and internet access. Tourism and export services need all three. We are pushing to add youth employment as a fourth national priority alongside them. Beyond that, what is underrated is not the ideas. Most of the good recommendations already exist somewhere. It is implementing capacity: a coordinating body, credible with government, business and civil society, whose entire job is to identify and remove sector-specific blockers. That is the kind of unit that sits between the Tourism industry and Home Affairs to get a visa system built, or between the global business services sector and its skills pipeline. That coordinating role is the piece the think-tank world consistently underestimates.