THE BUDGET AS TRUTH 3
What the Plans Say, What the Budget Spends, and What We Cannot Afford to Ignore
Budget as Truth — Article 3 | Thinking About It
Published May 2026
by Shelley Childs
I am not a development economist. I do not work in National Treasury or the Department of Planning, Monitoring and Evaluation. I do not have access to internal risk assessments, to the actual spend-versus-allocation data behind each budget line, or to the municipal Integrated Development Plan (IDP) implementation reports that would show what is happening on the ground rather than on paper. What I have is the ability to read public documents, follow the money where it is visible, and ask the question that three-day conferences consistently avoid: if these plans are real, where are they reflected in the budget?
This is the third article in the Budget as Truth series. In Article 1, we looked at what the budget reveals when you place VIP protection spending next to gender-based violence prevention funding. In Article 2, we examined whether South Africa’s governance ecosystem functions as a system capable of translating plans into delivery of predefined outcomes. Here, we look at the plans themselves — the full stack of development planning documents South Africa has produced — and ask what the national budget says about whether those plans are serious.
The short answer is: the plans are serious documents. The budget is a more honest document than the plans. And the gap between them describes a set of risks that, if unmanaged, point toward a society in which real economic development becomes progressively less likely, wealth becomes more concentrated, and crime fills the space that the state cannot.
The human cost of that gap is examined in Article 4. This article concerns itself with the structure: the plans, the budget, the fiscal risks, and what accountability would actually require.
South Africa Has a Plan for Everything
At the top sits the National Development Plan (NDP) 2030, adopted by Cabinet in 2012. It is South Africa’s long-term blueprint: eliminate poverty, cut unemployment to 6%, grow the economy at 5.4% a year, reduce inequality. It was an ambitious and largely well-argued document. The 10-year review in 2022 found that almost none of the headline targets had been met.
Below the NDP sits the Medium-Term Development Plan (MTDP) 2024–2029, the five-year plan of the current Government of National Unity. It has three priorities: inclusive economic growth and job creation; reducing poverty and tackling the cost of living; building a capable, ethical, developmental state. It quietly sets economic growth at 3% — almost half the NDP target — without drawing attention to the revision.
Below that sit provincial Growth and Development Strategies, and below those sit Integrated Development Plans (IDP) — five-year municipal plans required by law, annually reviewed, meant to translate national priorities into local delivery. Threading across all of this is the District Development Model (DDM), an attempt to force genuine coordination across the three spheres of government.
This is a great deal of planning. The parliamentary review of the Medium Term Strategic Framework (MTSF) found that the link between the NDP and actual implementation “is not as strong as it should be,” that there is “a general lack of inward reflection on the role each plan-creator is playing,” and that the compliance focus makes genuine flexibility almost impossible. In plain language: the plans produce plans. Implementation is a very different activity.
The Budget Is the Reality Check
If you want to know what government really intends to do, you read the budget. Not the speech — the numbers. Here is what the 2025/26 and 2026/27 budgets tell us:
Debt service costs: R420.6 billion in 2025/26, confirmed in the February 2026 Budget Review. The figure rises to R 469.3 by 2028/29. This is the single largest budget commitment, costing more than government spends on health. Twenty cents of every rand of government revenue goes to debt service before a single school is staffed, a single clinic stocked, a single pothole filled. (Debt service refers to the total amount of money required to cover all loan payments, including both principal and interest, over a specific period.)
How we got here: Debt-to-GDP has risen from 31.5% in 2009/10 to approximately 78.9% today. Since 2016/17, the interest rate on government debt has been higher than the economic growth rate — which means debt has been mathematically compounding itself as a share of GDP.
What was cut in the May 2025 budget revision: Infrastructure investment was reduced by R12.95 billion. PRASA — passenger rail — had its allocation removed. Education was cut by R3.11 billion. Health by R2.58 billion. Social protection by R4.05 billion. These are not cuts to waste. They are cuts to the programmes that the plans say are the priorities.
Three Risks That the Budget Must Respond To
Risk One: Fiscal Sustainability
The good news first. South Africa achieved a primary surplus — revenue exceeding non-interest spending — for the first time since 2008/09. A credit rating upgrade followed, the first in 16 years. Bond yields have fallen from an 11% peak toward 8%.
The risk that remains: gross fixed capital formation — productive investment in the economy — stood at approximately 14% of GDP in 2024, the latest year for which World Bank data is available. The NDP target was 30%. Investment contracted by 3.7% in 2024 before a slight uptick in Q4 2025. At full-year GDP growth of 1.1% in 2025 — the highest since 2022, but less than a fifth of the NDP target of 5.4% — the economy does not generate the tax base needed to service debt and expand public services simultaneously. One economist has assessed the current savings target as being approximately one-tenth of what is needed: the goal should be R100 billion per year redirected from low-value to high-value spending.
Risk Two: Youth Unemployment and Social Stability
The Q1 2026 Quarterly Labour Force Survey, released by Stats SA on 12 May 2026, shows the picture worsening. The official unemployment rate rose by 1.3 percentage points to 32.7% — up from 31.4% in Q4 2025 — as 301,000 people lost their jobs, and the number of employed persons fell to 16.8 million. Discouraged jobseekers rose to 3.9 million.
Youth unemployment for those aged 15–24 stands at 60.9%. Among 15–34 year olds, 45.8% are NEET — not in employment, education or training. The NDP target was 6% unemployment by 2030. FEDUSA, the Federation of Unions of South Africa, has described the Q1 2026 figures as evidence that South Africa is “normalising mass exclusion” — no longer a cyclical jobs problem but a structural crisis of economic exclusion, weak labour absorption, and declining confidence in the economy’s ability to create secure livelihoods at scale.
An unemployed young person without training or prospects is not an abstract statistic. The youth unemployment rate at current levels is not only a welfare crisis. It is the supply chain for the risk that follows. Article 4 in this series examines that chain in full.
Risk Three: Crime as Economic Weight
In 2023, the World Bank calculated that crime costs South Africa at least 10% of GDP annually — approximately R700 billion. That is close to three times government’s total spending on healthcare in the same year, and nearly six times the SAPS budget. South Africa ranks in the top five countries globally for homicide rates. The SAPS released its Q4 2025/26 crime statistics on 22 May 2026: murders declined 9.5% year-on-year. The daily average remains 58. Article 4 examines what those 58 murders per day cost beyond the immediate loss of life.
What Agile Looks Like: Three Specific Proposals
The planning system produces documents. What it does not reliably produce is accountability for outcomes. Three changes would shift that, none of which requires new legislation:
1. Fifteen indicators, not fourteen priority outcomes. A delivery unit embedded in the Presidency should track fifteen specific, measurable indicators — with quarterly data, public reporting, and direct presidential accountability when indicators slip. Not the indicators the departments choose. The indicators that reflect what citizens experience: youth employment rate, maternal mortality, rail punctuality, murder rate, water service interruptions, days to business registration, the frustration of dealing with the UIF.
2. Money conditional on results, not documents. Municipal grants should follow verified delivery: connection rates, water quality results, audit outcomes. Not IDP submission. Not compliance checklists. Municipalities that meet targets receive their allocation in full. Municipalities that do not receive it in tranches conditional on a recovery plan.
3. Public dashboards with genuine data. The DPME has committed to public-facing dashboards for MTDP monitoring. If those dashboards show indicator-level data, updated quarterly, with departmental responses required when targets are missed, the accountability architecture changes. Aspirational summary documents do not achieve this. Actual numbers do. (DPME=Department of Planning, Monitoring and Evaluation)
What Success Looks Like in 2030
Not the NDP’s metrics. Honest success markers for 2030: A young South African finishing school in 2025 can find employment, an apprenticeship, or a funded further education place within twelve months. The walk to the taxi rank is safe enough at night. There is a train, and it runs. The clinic has medication in stock. The municipality fixes the water when pipes burst, and bills correctly, the power stays on.
These are not grand ambitions. They are a minimum floor. They are also not currently available to the majority of South Africans.
The three risks described above are not independent. Youth unemployment feeds organised crime. Organised crime suppresses investment. Suppressed investment weakens the tax base. A weakened tax base means the debt burden cannot fall. Debt service wins by arithmetic. The wealthy exit this loop through private alternatives: private schooling, private healthcare, private security, private power. They are already doing so. The middle class and the poor have no exit.
The consequence of that divergence, extended over a decade, is not social tension at the margins. It is the structural entrenchment of a two-tier society in which planning documents are written by one tier about and for the lives of another.
What Those with Access Owe the Rest of Us
I have used public documents, published research, and information available in the public domain. There are things I cannot see: actual spend versus budget allocation, programme by programme; internal Treasury risk assessments; municipal IDP implementation data at ground level; NPA resourcing and capacity data; and which SOE contingent liabilities are closest to being called. (Contingent liabilities are potential financial obligations that may arise depending on the outcome of uncertain future events, such as malpractice lawsuits or those arising out of wrongful arrest or the abuse of police power.)
The argument of this series is simple: the budget is the most honest document government produces, because it forces choices that plans do not. If the people who hold that information can see whether the budget is matching risk mitigation to the actual risk — they should say so. Publicly. In terms that a person in the street can follow.
Not conferences. Not talk shops. Not plans. Money, dates, measurements, and consequences.
Question: The budget does not have a chapter on what happens when the connections between plans and delivery break. Is that no the chapter we need to write — not in a planning document, but in a budget.
This is the third article in the Budget as Truth series on Thinking About It. Article 4 — The Human Cost — follows directly from this piece.
Supplementary Note 1 — What Wasteful and Irregular Expenditure Actually Means, and What It Doesn’t — is a standalone reference piece in the series.
Sources and further reading:
National Planning Commission — NDP 2030: https://www.gov.za/issues/national-development-plan-2030
DPME — Medium-Term Development Plan 2024–2029: https://www.dpme.gov.za
Stats SA — QLFS Q1 2026 (released 12 May 2026): https://www.statssa.gov.za/publications/P0211/P02111stQuarter2026.pdf
Stats SA — GDP Q4 2025 and full year 2025 (released 10 March 2026): https://www.statssa.gov.za/?p=19291
Stats SA — Youth and the Labour Market Q1 2026: https://www.statssa.gov.za/?p=19526
FEDUSA — QLFS Q1 figures show South Africa is normalising mass exclusion (May 2026): https://www.fedusa.org.za/2026/05/13/qlfs-q1-figures-show-south-africa-is-normalising-mass-exclusion/
World Bank — Gross Fixed Capital Formation South Africa 2024: https://data.worldbank.org/indicator/NE.GDI.FTOT.ZS?locations=ZA
World Bank — Safety First: The Economic Cost of Crime in South Africa (November 2023): https://openknowledge.worldbank.org/handle/10986/40499
SAPS Q4 2025/26 Crime Statistics — released 22 May 2026: https://capeargus.co.za/news/crime-and-courts/2026-05-22-saps-to-release-latest-crime-statistics-today/
National Treasury — Budget Review 2026 (February 2026): https://www.treasury.gov.za/documents/national%20budget/2026
Stanlib — Revised Budget 2025/26 analysis: https://www.stanlib.com
IMF Country Report — South Africa Fiscal Framework (January 2025): https://www.imf.org/en/Publications/CR/Issues/2025/01/
The Conversation — South Africa fiscal savings target commentary (March 2026): https://theconversation.com/africa
OECD Economic Survey of South Africa 2025: https://www.oecd.org/economy/south-africa-economic-snapshot/