A democracy that got more corrupt
In June 2022, former Chief Justice Raymond Zondo handed the last volume of his commission's report to President Cyril Ramaphosa. After four years of hearings, the commission put a number on what it called the clearest evidence of looting during the Zuma era. More than R57 billion in public money was reportedly lost to state capture, and over 97 percent of this is linked to the state rail and logistics company Transnet and the power utility Eskom.[1] The commission itself stressed that this was the only part it could prove.
In 1994, South Africa became a democracy with one of the most progressive constitutions in the world. It set up a Public Protector, an independent Auditor-General and a powerful Constitutional Court. Yet three decades later, corruption is not shrinking. In Transparency International's Corruption Perceptions Index, South Africa scored 41 out of 100 in 2025. South Africa ranks 81st out of 186 countries. That puts it well above the sub-Saharan African average of 32, but below the average of 47 for flawed democracies. That is unchanged from the year before, and the score has been stuck below 50 for years The CPI matters because it measures perceptions drawn mainly from Western investor and expert surveys, which turns a liberal "good governance" model into a universal standard and frames corruption as a problem of the Global South.[2]
This article examines why corruption, having been so prevalent in South Africa during the Zuma administration, predates his presidency and may be a consequence of the political pulling and hauling between Washington and the newly elected government of South Africa during the 1990’s. My argument is that GEAR facilitated a state, which was prone to corruption and is not an African pathology.
Not a story about bad men
This dominant framing tends to single out individual people in power. The usual suspects are former South African president Zuma, the corruption-linked Gupta brothers and a handful of non-compliant state-owned enterprises (SOE) executives. Remove them, as the story goes, and the state will heal. This explanation is comfortable for Western investors and the current South African government. It allows the African National Congress (ANC) to present the Ramaphosa era as a clean break and allows investors and international lenders to treat corruption as a problem of African politics. I want to argue something different. State capture was the culmination, not the origin. The Guptas did not have to build the mechanism from scratch they used corrupt dealings and capturing the state. They found a state mechanisation that had already been reorganised. It was fiscally squeezed, fragmented by outsourcing and full of new points where public money meets private interest. That reorganisation began with a policy decision taken in 1996.[3] This does not mean that state capture-implicated individuals are innocent or that corruption is merely imported from Washington. What it means is that we have to ask a different question. Not only who stole, but what kind of state made stealing so easy, and who designed it.
Structural adjustment without the IMF
Unlike Zambia, Ghana or Zimbabwe, South Africa never signed a classic structural adjustment programme (SAP) with the International Monetary Fund (IMF). It did not need to. In June 1996, only two years after the democratic dispensation, the government replaced the redistributive Reconstruction and Development Programme (RDP) with a new macroeconomic framework called Growth, Employment and Redistribution, or GEAR.[4] GEAR read like a Washington Consensus checklist. It promised deficit reduction, tight control of public spending and public-sector wages, the restructuring and partial sale of state-owned enterprises, trade liberalisation and the gradual lifting of exchange controls. Critics on the left, above all Patrick Bond, called it a home-grown structural adjustment.[5] The conditions that elsewhere were imposed through loan agreements were adopted voluntarily, to win the confidence of investors and rating agencies. This matters for how we think about the Global South. Structural adjustment is not only a contract signed in a ministry of finance. It is a set of ideas about what a “modern” state should look like. Such a state is lean, market-friendly and run like a business. These ideas travel through advisers, rating agencies and global capital markets, and once adopted, they outlive any single programme. GEAR as a policy document faded after a few years, but the state it produced did not.
Three gateways to capture
How does a macroeconomic strategy turn into an opportunity for theft? Not directly. The link runs through the institutions of the state. Three changes, each a direct product of the post-1996 reform path, opened the doors that were later walked through.
1. A state that could no longer do things itself
Fiscal discipline was GEAR's core promise. Spending ceilings and wage restraint hit a public service that was, at the same time, supposed to integrate the fragmented apartheid administrations and extend services to millions of people who had never had them.[3] The result was a state with large new tasks and too few skilled people to carry them out. The gap was filled from outside. Consultants, private engineers and contracted service providers took over work that public servants had once done. Departments lost the knowledge needed to plan projects, assess bids and check whether contractors delivered.[6] A state that cannot judge what it buys is a state that can be overcharged, whether by accident or by design.
2. Procurement as the heart of government
As the state did less itself, it bought more. From the late 1990s, outsourcing and public-private partnerships turned procurement from a back-office function into the central mechanism of governance. More and more public money for building houses, maintaining roads, supplying hospitals and running IT systems flowed through tenders.[7] Every tender is a point where public authority meets private interest. Multiply these points by thousands, spread them across national, provincial and municipal levels, and give officials wide discretion over who wins. The result is what South Africans call the tenderpreneur. This is a business person whose main asset is not a product but political access. This figure is not a deviation from the outsourced state – he/she is its logical by-product.
3. Public companies, private logic
GEAR's plans for full privatisation largely stalled, partly due to resistance from trade unions. What happened instead was commercialisation. Eskom, Transnet, Denel and South African Airways stayed in public ownership but were told to operate along private-sector lines. They got corporate boards, performance targets, executive bonuses and large procurement budgets managed with little political oversight.[8] These hybrid entities combined the worst of both worlds. They had the money and the monopoly power of the state, but were shielded from democratic control by their corporate form. Whoever controlled the board and the chief executive's office, controlled billions in contracts. It is no coincidence that Transnet and Eskom account for almost all of the losses the Zondo Commission could document.
The missing brake
In theory, the Auditor-General, Parliament's oversight committees and the Public Protector should have stopped this, and they did produce warnings. Year after year, audit reports recorded growing irregular expenditure and repeat findings at the same departments and municipalities. But warnings without consequences become routine.[9] Oversight bodies, like the Public Protector and the Auditor-General, were themselves under-resourced, and their findings rarely led to dismissals, prosecutions or recovered funds.[10] Put together, the mechanism is simple. Austerity weakened the state's own capacity. Outsourcing and commercialisation created countless lucrative interfaces with private actors. Weak oversight meant that abuse was unlikely to be punished. Political scientists have long described corruption as the product of monopoly plus discretion minus accountability.[11] From 1996 onwards, South Africa's reforms raised the first two and weakened the third.
State capture used the machine, it did not build it
Read against this background, the Zuma years look less like a sudden collapse and more like the systematic use of a ready-made infrastructure. The Gupta network did not need to invent new instruments. It placed allies on SOE boards and in executive posts, steered appointments in ministries and channelled contracts through procurement rules that already allowed wide discretion. The cases documented by the Public Protector's State of Capture report (2016) and the Zondo Commission follow this pattern closely.[12] At Transnet, the commission found that contracts worth R41.2 billion awarded under its leadership were illegal, including the purchase of locomotives at inflated prices. At Eskom, the Gupta-owned company Tegeta received prepayments and guarantees that helped finance its takeover of the Optimum coal mine. Consulting firms such as Trillian and McKinsey received payments the commission described as unlawful.[1] Every one of these schemes ran through the gateways described above. Each involved a commercialised SOE with a large procurement budget, an outsourced function, a consultant filling a capacity gap and oversight that came years too late. The researchers behind the Betrayal of the Promise report (2017), later expanded into the book Shadow State, called this a “silent coup”.[13] But a coup this quiet is only possible when the institutions have already been hollowed out.
What this argument does not say
Three objections deserve an answer. First, the ANC's own practices played a central role. Cadre deployment, the party's policy of placing loyal members in key state posts, was a domestic choice. So was the use of the tenderpreneur procurement process to build a politically connected black business class, neither was a demand by the IMF. Second, the decline of state capacity is uneven. Some institutions, such as the South African Revenue Service (SARS) before 2014, or parts of the judiciary, remained remarkably strong. If institutional decay had been the inevitable product of structural pressures, these bodies should have weakened alongside the rest of the state. Third, outsourcing was sometimes a pragmatic response to real skills shortages rather than an ideological project. After 1994, the state shed many experienced officials through severance packages and emigration, and the apartheid education system had produced too few Black engineers, accountants and IT specialists to replace them. Contracting these functions out was often the only short-term way to keep services running.
All three points are true, and none of them undermines the argument. Patronage networks need resources to distribute, and the reformed state provided them in unprecedented volume. Strong islands within the state show that capacity was a political choice, not fate, and a pragmatic response that becomes permanent still reshapes the state. The point is not that external reforms caused corruption on their own. It is that they created the institutional terrain on which domestic actors could make corruption systemic.
The cure that repeats the disease?
This is not just history. Since 2020, the Presidency and National Treasury have driven a reform programme called Operation Vulindlela, now in its second phase and backed by the Government of National Unity formed in 2024. Its centre piece is the unbundling of Eskom into separate generation, transmission and distribution entities, with a competitive wholesale electricity market. Ports and rail are to be opened to private operators, with the national ports authority owning the infrastructure and private firms competing to run terminals.[14]
The official justification is growth. Implicitly, it is also clean governance. If the state controls fewer contracts, the argument goes, there is less to steal. Some of these reforms may well improve services. But the analysis above should make us cautious. Private participation in the grid and the ports means new concessions, new licences and new long-term contracts. Each of these is another interface between public authority and private interest. These are exactly the kind of gateways that state capture exploited. Without a state that has the capacity to negotiate, regulate and monitor these deals, more market access can mean more opportunities for capture, only in a different legal form.
For readers in Europe there is also a lesson closer to home. Germany and the EU are among South Africa's most important partners, not least through the Just Energy Transition Partnership (JET-P). Development finance and policy advice that push for liberalisation while treating corruption as a purely local governance failure risk repeating the mistake of the 1990s. If we want clean institutions in the Global South, we should stop prescribing the reforms that weakened them. South Africa's corruption is not an African pathology, nor simply the work of a few greedy men. It is, to a significant degree, the product of a particular way of reorganising the state. South Africa adopted it in 1996 and is being encouraged to deepen it today.
End Notes
1. Judicial Commission of Inquiry into Allegations of State Capture (2022): Final Report. Pretoria; Mail & Guardian (2022): R57 billion: How the Guptas played Monopoly with state money,23 June 2022.
2. Corruption Watch (2026): CPI 2025: SA's performance unchanged, global average declines.
3. Presidential Review Commission (1998): Developing a Culture of Good Governance: Report of the Presidential Review Commission on the Reform and Transformation of the Public Service in South Africa. Pretoria: Government of South Africa.
4. Department of Finance(1996): Growth, Employment and Redistribution: A Macroeconomic Strategy.Pretoria: Government of South Africa.
5. Bond, P. (2000): Elite Transition: From Apartheid to Neoliberalism in South Africa. London: Pluto Press.
6. Auditor-General of South Africa (2013): Performance Audit Report on the Use of Consultants at Selected National Departments. Pretoria: Auditor-General of South Africa.
7. Public Affairs Research Institute (PARI) (2014): The Contract State: Outsourcing & Decentralisation in Contemporary South Africa. Johannesburg: PARI.
8. Mostert, C. (2002): Reflections on South Africa's Restructuring of State-Owned Enterprises. Occasional Paper No. 5. Johannesburg: Friedrich-Ebert-Stiftung; Bowman, A. (2020): Parastatals and Economic Transformation in South Africa: The Political Economy of the Eskom Crisis. African Affairs 119(476), 395–431.
9. Auditor-General of South Africa (2019): Consolidated General Report on National and Provincial Audit Outcomes: PFMA 2018–19. Pretoria: Auditor-General of South Africa.
10. Institute for Security Studies (2013): Parliament and Treasury Should Ensure Sufficient Funding for the Public Protector. Press release. Pretoria: ISS; Corruption Watch (2013): Call on Government to Walk Its Anti-Corruption Talk. Johannesburg: Corruption Watch; Republic of South Africa (2018): Public Audit Amendment Act, No. 5 of 2018. Pretoria: Government Printer.
11. Klitgaard, R. (1988): Controlling Corruption. Berkeley: University of California Press.
12. Public Protector South Africa (2016): State of Capture. Pretoria; Judicial Commission of Inquiry into Allegations of State Capture (2022).
13. Bhorat, H. et al. (2017): Betrayal of the Promise: How South Africa Is Being Stolen. State Capacity Research Project; Chipkin, I./Swilling, M. (2018): Shadow State: The Politics of State Capture. Johannesburg: Wits University Press.
14. Financial Mail (2026): Stillmixed results for Operation Vulindlela,21 May 2026.
.png)