Municipal corruption in South Africa damages services through captured procurement and absent consequences

Introduction

When a municipal tender goes wrong, the first sign is usually an asset that is late, over budget or not working, and sometimes the investigators arrive afterwards. The Special Investigating Unit’s 2023/24 annual report lists matters of this kind, among them proclamations on the procurement of a reinforced concrete water reservoir for Amajuba District Municipality and on the Rooiwal wastewater treatment works upgrade in Tshwane (SIU, 2024). This paper asks how corruption of this sort affects service delivery in South African municipalities. Its answer is fairly specific. The harm seems to come mainly from two things working together, procurement that politically connected networks can capture and a system of consequences that rarely bites afterwards, and not from a shortage of rules (Gray, 2021; Zungu et al., 2025b).

Two cautions apply from the start. First, irregular expenditure is not the same as theft, and a good deal of what auditors flag is carelessness or fear of getting a form wrong (Phakathi et al., 2024; Ajam, 2025). Second, the academic studies used here are mostly qualitative and small, while the official figures are national and broad. The paper looks first at what the audit and investigation record shows, then at procurement, then at patronage as a state capture spillover, then at the accountability institutions, and finally at what can and cannot be said about service outcomes (AGSA, 2025).

What the audit and investigation record shows

The Auditor-General’s consolidated report for 2023-24 is the best single picture. Of the 257 municipalities in scope, audits of 247 were complete by the cut-off, and only 41 (16 per cent) received clean audits. Irregular expenditure of about R87.03 billion was incurred over the three years of the current administration (AGSA, 2025). Procurement and contract management findings turned up at 214 municipalities, or 87 per cent, with material non-compliance at 155. The report also counts 446 material irregularities since 2019, 285 of them involving financial loss with an estimated value of R8.74 billion. About 48 per cent had been resolved, and roughly R1.32 billion had been recovered, was being recovered or had been prevented (AGSA, 2025).

The Special Investigating Unit tells a complementary story from a different angle. In 2023/24 it reported about R2.29 billion in cash and assets recovered, contracts and administrative decisions worth about R2.14 billion set aside, 583 criminal referrals and 297 disciplinary referrals (SIU, 2024). Three of its new proclamations concerned municipalities: Chris Hani District Municipality over the appointment of a co-operative development centre, Amajuba over the water reservoir tender, and Tshwane over wastewater works upgrades. These SIU numbers cover all state bodies, not municipalities alone, so they are a rough guide to capacity and not a count of municipal corruption (SIU, 2024; AGSA, 2025).

Reading these figures as a measure of corruption would be a mistake. Phakathi et al. (2024) studied unauthorised, irregular and wasteful expenditure in the City of Tshwane, a metro, and metros are the category the Auditor-General identified as the biggest contributor to such spending. They traced it mainly to cultural and behavioural weaknesses in the internal control environment and to management practices that left oversight thin, though they also found that the absence of measures to discourage misconduct had handed people opportunities to embezzle funds. The mixture matters. Some irregularity is sloppiness, some is theft, and the audit categories do not separate the two (AGSA, 2025).

The spread of the problem also tells us something. Metros accounted for about R33.29 billion of the irregular expenditure incurred since 2021-22, and unauthorised expenditure of R31.79 billion was incurred by 174 municipalities, or 68 per cent of those audited (AGSA, 2025). So the problem is not confined to a few failing small councils. Phakathi et al. (2024) chose Tshwane after looking at five years of trends across metros and finding significant problems there, which suggests that large budgets and complicated procurement do not protect a municipality from the problem.

Procurement: where the money leaves

Procurement is where the state capture investigations found the money moving. Gray (2021) notes that those investigations identified procurement as a central mechanism for generating rents, and argues that South Africa’s anti-corruption policies already meet world-class standards. What fails is the practice. Gray describes behaviour reminiscent of apartheid-era compliance, in which procurement staff prioritise following orders over following policy, which undermines the function’s contribution to organisational effectiveness. Mseleku (2025), working from the literature and the State Capture Commission’s report, adds that decentralising procurement created a breeding ground for corruption and fraud, because political capture and manipulation often undermine what decentralisation is supposed to deliver. Both point away from more rules and towards how decisions actually get made.

There is a common thread in three otherwise different studies. Gray (2021) describes procurement staff who follow orders, Phakathi et al. (2024) describe a control environment shaped by culture and behaviour, and Ajam (2025) describes officials who choose the cautious option because audit findings and judicial review frighten them. In each, the working question for the official seems to be whether a decision is safe for them personally, and not whether it delivers the service. That kind of culture suits capture rather well. A procurement officer who is used to following instructions, and who expects to be blamed for any irregularity, is not well placed to say no to a politically connected bidder.

The local evidence has the same flavour. Zungu et al. (2025a) interviewed ten councillors in KwaZulu-Natal about the causes of irregular expenditure. Weak planning and project management were part of the story, yet in most instances the councillors described intentional violations of procurement policy, abuse of disaster management provisions, fraud and collusion among councillors, politicians, officials and contractors. Some also described fear, intimidation and assassinations of people who opposed fraud. Ten people is a small sample and they spoke about perceptions, not audited cases. Even so, the account fits the SIU’s choice of tender-specific proclamations, such as the Amajuba reservoir bid, and it fits Gray’s point that the rules are not the weak spot (SIU, 2024; Gray, 2021).

Not everyone reads the problem as a lack of enforcement, and the counter-evidence deserves space. Ajam (2025) interviewed about two dozen senior managers and supply chain practitioners in five municipalities in the Western Cape and Gauteng and found that layered, differently interpreted procurement rules produced heavy workloads and delays. A full tender could take up to six months, and one tender advertised for 29 days instead of the required 30 drew an audit finding even though no service was compromised. Some participants said fear of audit findings made officials postpone or withhold decisions, and one described regulatory complexity as increasing exposure to the very fraud it aims to prevent. The sample was small and not random, so the study cannot speak for the whole system. Still, it suggests that adding rules without changing how they are enforced may slow delivery and still leave the real fraud untouched (Ajam, 2025; Phakathi et al., 2024).

State capture spillovers: patronage in appointments

If procurement is the mechanism, appointments are probably the entry point. Swanepoel (2021) scrutinises the ruling party’s policy of cadre deployment from a legal and political angle. Swanepoel argues that the legal foundation for exercising public power lies in the Constitution and its enabling legislation, not in party policy, and that deployment is incompatible with a constitutional state and instead enables a shadow state. His view is contested, and it is an argument built from law and political commentary, not a measurement of anything. It still gives a coherent story about how people end up in the committees and posts that decide tenders (Swanepoel, 2021; Mseleku, 2025).

Shopola and Juta (2024) look at the same problem through the Local Government: Municipal Systems Amendment Act of 2022, which limits the political rights of municipal officials. They conclude that the changes will unsettle intra-party arrangements and uproot some favouritism and nepotism, but they also say that doing so will not be easy given the damage already done, such as bloated organograms and cadre deployment driven by ruling parties in councils. Zungu et al. (2025b) list cadre deployments among the factors behind the absence of accountability in their KwaZulu-Natal interviews. These sources agree on direction. They are less helpful on size, because none of them measures how much procurement manipulation follows from a particular appointment.

That limit is worth stating plainly. The spillover from state capture into municipalities is plausible and consistently described, but in this literature it is inferred more often than traced. Mseleku (2025) works from the Commission’s report plus secondary literature, and Gray (2021) draws on the same investigations, so applying their findings to individual municipalities involves a step of reasoning (Mseleku, 2025; Gray, 2021). It is a reasonable step. It should be labelled as one.

Accountability institutions: detection without deterrence

The institutions that are meant to catch this behaviour do catch a lot. What happens next is the weak part. The Auditor-General found that 132 municipalities, or 53 per cent, did not comply with consequence management legislation in 2023-24, and only about 48 per cent of material irregularities had been resolved (AGSA, 2025). Zungu et al. (2025b) went further in their interviews with ten KwaZulu-Natal councillors. They concluded that consequence management was effectively non-existent, pointing to a lack of political will among councillors, non-functional Municipal Public Accounts Committees, missing tender records, threats and the involvement of criminal groups.

The oversight bodies themselves look under-equipped. Fourie and van der Waldt (2023) surveyed people connected to oversight committees across a random sample of 20 municipalities and concluded that councillors are relatively ill-prepared for oversight even though adequate laws, mechanisms and structures exist. Factionalism and party politics limited what councillors could do, and respondents raised weak consequence management as a concern. Van der Waldt et al. (2024) reach a similar place from the audit side. Their situational analysis of Auditor-General findings argues that municipal capacity has to improve before audit recommendations can be acted on, and that councils should adopt consequence management frameworks backed by disciplinary boards and municipal misconduct regulations.

Put together, the evidence implies that detection is running ahead of deterrence. Phakathi et al. (2024) found that irregular spending persisted in Tshwane despite the Auditor-General’s findings and an extensive oversight role for the municipal public accounts committee. On the enforcement side, the SIU’s recoveries of about R2.29 billion in 2023/24 sit beside R87.03 billion of irregular municipal expenditure over three years (SIU, 2024; AGSA, 2025). They are not like for like, since irregular expenditure is not money lost and the SIU covers the whole state, so the comparison proves nothing exact. The difference in scale is still hard to ignore.

Intimidation weakens the whole chain, too. Zungu et al. (2025a) report fear, intimidation and assassinations of those who opposed fraud and corruption, and Zungu et al. (2025b) add threats and the involvement of criminal groups. When the people closest to a tender, such as junior officials, councillors on oversight committees and honest bidders, expect to be punished for speaking, the auditors and investigators lose their best source of evidence. The Auditor-General can refer unresolved material irregularities to other public bodies for investigation, but the referral only works if there is evidence to follow and someone willing to give it (AGSA, 2025; Zungu et al., 2025a).

What corruption does to services, and how sure we can be

The service consequences are easiest to see in the infrastructure findings. The Auditor-General visited 113 infrastructure projects and found deficiencies, such as delays, cost overruns, poor quality or non-commissioning, at 87 of them, or 77 per cent. It also attributed R37.28 billion in water and electricity losses to infrastructure neglect (AGSA, 2025). The SIU’s municipal proclamations concern a water reservoir and a wastewater works, which are exactly the assets residents depend on (SIU, 2024). Zungu et al. (2025a) open from the premise that irregular expenditure severely affects the delivery of essential services. All of this fits a story in which captured tenders produce poor or incomplete assets.

Wastewater is a useful test case, since the Rooiwal proclamation concerns exactly that kind of asset. The Auditor-General reported 15 material irregularities relating to wastewater treatment works and landfill sites at metros and entities, 56 material irregularities for pollution of water sources, and non-compliance with environmental legislation at five metros, including over wastewater works and landfill licences (AGSA, 2025). These findings describe harm that residents and rivers actually bear. They do not say how much of it began with a corrupt tender and how much with plain neglect, and the audit categories are not built to answer that (AGSA, 2025; SIU, 2024).

None of this proves that story. Neglect and corruption are different things, and the Auditor-General’s loss figures are about the former. None of the academic studies used here measures service outcomes against a direct measure of corruption in the same municipalities. Most rest on officials’ or councillors’ perceptions, and a few rest on documents. Ajam (2025) shows a second route to poor delivery, in which cautious officials delay awards to avoid findings, so slow services are not automatically a sign of theft. A fair reading is that corruption probably damages services through unfinished or substandard projects and through intimidation of honest officials, while compliance fear and weak capacity damage them independently (Ajam, 2025; Phakathi et al., 2024).

That reading has policy consequences. If the main damage comes from captured procurement and the absence of consequences, then the useful measures are the ones that change who decides and what follows a finding. Those include professional appointments under the Municipal Systems Amendment Act, complete and open tender records, functioning municipal public accounts committees and disciplinary processes that actually conclude (Shopola and Juta, 2024; Zungu et al., 2025b; van der Waldt et al., 2024). Another layer of procedural rules would not obviously help, and Ajam (2025) gives some reason to think it could make matters worse. This is a judgement drawn from a thin and mostly qualitative evidence base, so it should be tested in audited, municipality-level data before anyone treats it as settled.

Conclusion

Corruption affects municipal services in South Africa mainly by capturing procurement and by escaping consequences, and the studies reviewed here point to practice and enforcement, not missing legislation, as the weak points (Gray, 2021; Zungu et al., 2025b). State capture spillovers through patronage appointments are well described but poorly measured (Swanepoel, 2021; Shopola and Juta, 2024). The Auditor-General and the SIU detect and recover a good deal, yet the figures suggest that detection outruns deterrence, with more than half of municipalities not complying with consequence management legislation (AGSA, 2025; SIU, 2024). Counter-evidence matters too. Irregularity is not always corruption, and heavy compliance can slow services without any theft (Ajam, 2025). The best next step is research that links audited corruption indicators to service outcomes in the same municipalities, because the present literature mostly infers the connection.

Reference list

Ajam, T. (2025) Supply chain management administrative burdens: a case study of five South African municipalities. Journal of Local Government Research and Innovation, 6, article a206. doi:https://doi.org/10.4102/jolgri.v6i0.206

Fourie, D.J. and van der Waldt, G. (2023) Enhancing municipal councillors’ oversight in local, district, and metropolitan municipalities. Journal of Local Government Research and Innovation, 4, article a145. doi:https://doi.org/10.4102/jolgri.v4i0.145

Gray, N. (2021) When anti-corruption fails: the dynamics of procurement in contemporary South Africa. Review of African Political Economy, 48(169), pp. 369-384. doi:https://doi.org/10.1080/03056244.2021.1932789

Mseleku, Z. (2025) Decentralisation, governance, and corruption: evidence from the Commission of Inquiry into State Capture in South Africa. African Journal of Public Administration and Environmental Studies, 4(1), pp. 66-90. doi:https://doi.org/10.31920/2753-3182/2025/v4n1a4

Phakathi, P.S., Schutte, D.P. and Mulati, S.P. (2024) An assessment of the drivers of unauthorised, irregular, and wasteful expenditure in the City of Tshwane Metropolitan Municipality. Southern African Journal of Accountability and Auditing Research, 26(1), pp. 137-161. doi:https://doi.org/10.54483/sajaar.2024.26.1.8

Shopola, M. and Juta, L. (2024) Political patronage in South Africa local government: is the Local Government: Municipal Systems Amendment Act helping matters? Journal of Public Administration, 59(1), pp. 102-118. doi:https://doi.org/10.53973/jopa.2024.59.1.a9

Swanepoel, C.F. (2021) The slippery slope to State capture: cadre deployment as an enabler of corruption and a contributor to blurred party-State lines. Law, Democracy and Development, 25, pp. 1-23. doi:https://doi.org/10.17159/2077-4907/2021/ldd.v25.15

van der Waldt, G., Fourie, D. and Malan, C. (2024) Implementing auditor-general annual recommendations in South African municipalities: a situational analysis. Public Money & Management, 44(7), pp. 617-627. doi:https://doi.org/10.1080/09540962.2024.2304544

Zungu, A., Sibanda, M. and Nomlala, B. (2025a) Causes of irregular expenditure in KwaZulu-Natal municipalities: perspectives of councillors. International Journal of Applied Research in Business and Management, 6(1). doi:https://doi.org/10.51137/wrp.ijarbm.2025.azcb.45697

Zungu, A., Sibanda, M. and Nomlala, B. (2025b) Consequence management for irregular expenditure and procurement irregularities: a qualitative study of councillors’ insights and perspectives in KwaZulu-Natal. International Journal of Applied Research in Business and Management, 6(1). doi:https://doi.org/10.51137/wrp.ijarbm.2025.azcb.45698

Legislation and case law

Auditor-General of South Africa (AGSA) (2025) Consolidated general report on local government audit outcomes 2023-24. Pretoria: AGSA. Available at: https://mfma-2024.agsa.co.za/report/mfma_report_2023_24.pdf (Accessed: 18 October 2026).

Special Investigating Unit (SIU) (2024) Annual report 2023/24. Pretoria: SIU. Available at: https://www.siu.org.za/wp-content/uploads/2024/10/SIU-202324-Annual-Report_Tabled.pdf (Accessed: 18 October 2026).