The broken pipeline — why SA’s compliance-driven education system is failing

In April, executives from the Education, Training and Development Practices Seta sat before Parliament to explain why R637-million in skills development funds could not be accounted for. A senior finance manager offered a defence: the missing documentation was the result of “accounting and human error”. The person handling the transactions, he added, had been using the system for the first time.

The authority whose entire mandate is training and development for the education sector could not operate its own financial system — blaming the failure on an untrained staff member.

The missing funds are now the focus of a criminal investigation, and the President has since declared that the Seta model is not working and must be replaced. Across town, the Auditor-General has declared the State Information Technology Agency (Sita) a systemic risk to government IT delivery, having found it operated for more than three years without a permanent chief information officer, a permanent board or a permanent managing director, and with a 54% executive vacancy rate. The department has now formalised a three-year plan to redesign Sita, a reform that was promised for seven years but never concluded.

People are asking: What went wrong at Sita? What went wrong at the Setas? What went wrong with the skills pipeline? Wrong questions. These are not separate scandals. They are fracture points on the Pipeline Disruption Map — a diagnostic that emerged from patterns I kept seeing once I read the education-to-industry pipeline as a single system, from the Foundation Phase classroom through matric and higher education to industry entry. Plotted there, they resolve into one failure, surfacing at different points along the same line.

The map tells you something the audits and hearings cannot: where and when each break was planted. The finance officer who could not interpret her own journals did not fail in April. She failed when somebody taught her to reproduce rather than evaluate procedures, and that occurred 15 or 20 years ago. A broken pipeline does not mean broken people. The candidates emerging from it are not the problem; they are the output of decisions made about them by people who did not anticipate what was coming.

I know, because I spend my days at the far upstream end of that pipeline, and I see what happens when you interrupt it.

This year I have been implementing a phenomenon-based, inquiry-driven curricular framework at a remedial school in Johannesburg. It is fully compliant with Caps (Curriculum and Assessment Policy Statement). The core content aligns with the standard national Grade 6 curriculum, but the architecture differs significantly. Rather than compartmentalising subjects into isolated 40-minute periods, this model organises learning around central phenomena, integrating multiple disciplines into a unified inquiry.

My class consists of 16 boys, all classified as remedial learners; all children the mainstream system has already failed at least once. By the third week of the first term every one of them met the writing assessment target, and by the fourth week the class averaged 92% on an independent vocabulary and comprehension assessment.

The real pressure test

More telling was what happened when the second-term examinations arrived: the real pressure test. The subjects they had learnt through integration held essentially level under exam conditions, the assessment format that typically punishes remedial learners most. Their learning had durability, and durability is exactly what our institutions lack.

These boys are not flourishing because the content is easier. They are flourishing because the architecture makes learning make sense. When I asked them to design their own classroom economy, they independently created an inverted wage hierarchy in which essential services earn more than managerial roles, reasoning that “the class cannot function if trash is not collected, but it can run without a manager”. That is economic reasoning from children the system has labelled remedial. The label describes their instructional history. It says nothing about their cognitive capacity.

How does the break get planted in the first place? I have worked with teachers across every context this country offers (IEB, private and government). When I ask them to name the skills their pupils should demonstrate by year-end, I hear aspirational outcomes instead: pupils will understand photosynthesis, appreciate diversity, know about apartheid. Important as purposes, impossible to observe, therefore impossible to assess. This is not a personal failing.

South Africa’s teacher training reversed the international definitions. It taught educators that outcomes are what you measure and skills are how you get there, when the truth runs the other way. We handed an entire profession a compass that points south, wondered why nobody reached the North Pole, and when the curriculum changed, never rewired the compass. Content delivery without skill development produces people who can pass the assessment but not solve the problem the assessment did not anticipate. It produces, eventually, the finance officer who can process a transaction but cannot interpret what the journal is telling her. The conditioning is the same. Only the uniform changes.

Every institution along this pipeline now faces the same choice: what I call the stewardship–extraction bifurcation. An extractive institution looks at the person who cannot do the job and replaces them, then harvests the same crop from the same pipeline a year later. A stewarding institution traces the gaps upstream and rebuilds the capacity — not as charity, but as strategy, because the people who lived the pipeline failure know exactly where it breaks.

A coherent framework

I have watched stewardship work at the scale of one teacher. A sceptical colleague did her own research and some weeks later told me she finally had valid pedagogical reasoning rather than guesswork; she had simply never been trained to think about why she teaches what she teaches. Given a coherent framework and safe conditions to experiment, her capacity surfaced within weeks. Nobody replaced her. The conditioning is reversible.

Education is not a sector sitting quietly in its own corner. It is the root system, and every other system inherits what it plants. The matric debate, the Sita debate, the Seta debate are all downstream questions. The upstream question is: what kind of training produced these people, and are we willing to rebuild it rather than keep replacing its outputs?

I have two terms of evidence, from one classroom in one small school, that rebuilding works. It is small data. But it shows what the audits and hearings keep obscuring: the problem was never the people; it was what we trained them to do.

If we continue to treat education as a compliance exercise, we will continue to inhabit a country that functions only on paper.

The harvest matches the seeds. It always does. DM