Municipal service failure in South Africa owes more to weak maintenance and accountability than to funding

Introduction

Most people meet the South African state through a tap, a toilet, a streetlight or a refuse truck, and in a good many municipalities that meeting goes badly. The usual explanations are money, skills or corruption, and each has some truth in it. This paper takes a narrower line. Drawing on case studies of Buffalo City, Amathole, Chris Hani, Makana, Ekurhuleni, uMgungundlovu and eThekwini, it argues that delivery fails most reliably where municipalities let maintenance slide, run weak billing systems and beneficiary registers, and leave residents without any channel that works. Protest then becomes the route people actually use. Funding matters, but it explains less than the public debate suggests (Balie and Horn, 2021). The same pattern of ageing pipes, estimated bills and unanswered complaints keeps turning up in the Eastern Cape, Gauteng and KwaZulu-Natal (Twalo, 2025; Patience and Nel, 2021; Muthwa and Matsiliza, 2024).

A caution before going further. Most of the studies used here are qualitative, with small samples and a single municipality each, so they show how failure looks from the inside and cannot tell us how common it is (Sofianos and Matthews, 2025; Pillay and Mutereko, 2022). The paper looks first at water in three Eastern Cape districts, then at sanitation, electricity and protest in Buffalo City, then at maintenance and administration in Gauteng and KwaZulu-Natal, and finally asks whether money really is the binding constraint (Hutete and Sibanda, 2022; Ndasana et al., 2022).

Water in the Eastern Cape: backlogs, bills and the Makana crisis

The Eastern Cape districts show how big the physical backlog is. In Amathole District Municipality, the 2019-2020 integrated development plan figures quoted by Hutete and Sibanda (2022) put about 30 per cent of households without formal piped water, and roughly 60 per cent of the population lives in rural areas where the gaps are concentrated. Chris Hani District Municipality looks similar, with around 26 per cent of households lacking formal piped water according to Twalo (2025). A councillor in the Amathole study said four water-carting vehicles had to serve 23 wards, which is basically a shortage of trucks (Hutete and Sibanda, 2022). Nobody should pretend that geography and poverty are not part of this whole thing. Spread-out villages are expensive to reach.

Even so, the two studies also describe failures that have little to do with remoteness. Officials in Twalo’s (2025) Chris Hani study said the municipality often sends no invoices at all, and when bills do arrive some consumers say they are inflated because meters are not read and consumption is estimated. Some residents who could pay have turned to boreholes and tanks instead. Water departments, according to officials, do not have ring-fenced budgets, so they compete with other departments for council money and maintenance is neglected (Twalo, 2025). In Amathole, participants called the integrated development plan consultations a formality, and the district received a disclaimer audit opinion in 2018-2019 (Hutete and Sibanda, 2022). The picture suggests a chain: poor billing weakens revenue, weak revenue starves maintenance, and bad service gives people reasons not to pay.

Makhanda, in the Makana Municipality, is the sharpest case of what happens when that chain snaps. During the 2019 water crisis residents reportedly went up to two weeks without water, and Sofianos and Matthews (2025) trace the crisis to drought, low dam levels, poorly maintained treatment plants and deteriorating infrastructure. The NGO Gift of the Givers arrived on 12 February 2019, and by 20 February the municipality reported 15 boreholes and several tanks installed. The relationship then went wrong. It rested on a verbal understanding, which the authors link to the urgency of the crisis and to frequent changes of municipal manager, and by May the NGO had withdrawn over unpaid costs while the Department of Water and Sanitation had asked it to remove its trucks (Sofianos and Matthews, 2025). Several boreholes were reportedly unused, partly because of poor siting and a municipality unable to maintain what already existed.

What the Makana case adds is a warning about substitutes. NGOs can fill a gap quickly, and Sofianos and Matthews (2025) accept that, but they argue that relief organisations are better suited to emergencies than to running water services. The municipality declined to be interviewed, so its side comes only from public statements, and that is a real limit on the evidence. Still, read next to Twalo (2025), the case implies that the underlying administrative gaps (unclear roles, no written agreements, nobody responsible for maintenance) matter at least as much as the drought that exposed them. The drought set things off, and the administrative gaps decided how badly it hurt.

Quality and reliability make the gap between town and village easy to see. Urban residents in Amathole reported steady supply and advance notice of outages, whereas rural residents depended on carting, and one resident said tanks could go weeks or months without being refilled (Hutete and Sibanda, 2022). Intermittent supply pushes households to store water, which the authors link to hygiene worries, especially during the COVID-19 period. In Chris Hani a consumer described getting up before dawn because taps ran for only about an hour a day, and consumers who did pay felt their payments were not matched by reliable service (Twalo, 2025). Pay for water that does not come and the case for paying weakens quickly.

Buffalo City: sanitation, electricity and the protest as a message

Duncan Village, a township about five kilometres from East London’s central business district in Buffalo City, shows what residents say when asked about the same failures. Ndasana et al. (2022) interviewed about twenty people, the ward councillor among them. Housing came up first: people described leaking shacks, one had waited since 2008 without an RDP house, and another had applied in 2015 with no result. Many households lacked legal electricity connections and used illegal ones, which residents called dangerous for children. Shared municipal toilets were described as dirty, too far from homes and frightening at night for women and children, and refuse was not collected regularly (Ndasana et al., 2022).

The communication gap stands out in this account at least as much as the infrastructure does. Residents said they saw the ward councillor mainly during election campaigns, that meetings produced no feedback, and that promises were not followed up (Ndasana et al., 2022). That sits oddly with the municipality’s reported capital budget of R2.39 billion for 2019-2020, whose priorities included R284 million for wastewater, and with its report that ward committees had been established in all wards. Money was allocated, in other words, and committees existed on paper. The residents’ experience was of toilets nobody cleaned and councillors nobody saw, and that looks like a failure of follow-through, not of allocation (Ndasana et al., 2022).

Protest in this setting works as a message, and a fairly rational one. A resident told the researchers that protests had won a few toilets but not houses or electricity, and the paper cites a 2015 newspaper report of near-weekly blockades of the Mdantsane bypass (Ndasana et al., 2022). Chiwarawara (2024), writing from Gugulethu and Khayelitsha in Cape Town, makes a related point. He argues that newer labels such as local political protests or community protests conceal the service delivery component, which remains large even where protests have mixed motives, and he notes that a house is often the place where several services come together. The two studies are in different provinces and used different methods, but they agree that treating protest as mere disorder misreads it. Residents seem to protest because the ordinary channels, such as ward meetings, produced nothing.

Gauteng and KwaZulu-Natal: neglected assets and hollow registers

Ekurhuleni is a large metro with a deep asset base, which makes its numbers useful. Patience and Nel (2021) examined three technical departments (energy, water and wastewater, and roads and stormwater) over the 2014/15 to 2018/19 financial years. The city’s property, plant and equipment was worth about R60 billion in 2019-2020, yet renewals averaged only about 3.6 per cent of that asset base against a norm of 8 per cent. Unaccounted losses in 2018-2019 were around 1.28 billion kilowatt-hours of electricity, worth roughly R1.18 billion, and 106 million kilolitres of water, worth about R1.0 billion. The authors estimate that the combined R2.2 billion could have funded roughly 8,800 low-cost houses, though that is an illustration and not a budget decision anyone actually faced. The same authors benchmark the city against equally sized metropolitan municipalities, and note that Ekurhuleni does well in some areas of service delivery (Patience and Nel, 2021).

The reasons they give are managerial and political, not mainly financial. Politicians were described as favouring new assets over maintaining existing ones, asset registers lacked credibility, repairs were rarely recorded against individual assets, and oversight tended to ask whether money had been spent rather than what it had bought (Patience and Nel, 2021). Roads and stormwater had underspent its repairs benchmark by about R180 million, which the authors link to visible potholes. Bikam and Chakwizira (2021), looking at metros in Gauteng and rural municipalities in Limpopo, also find ageing infrastructure under pressure from rapid urbanisation, and they stress insufficient funding and a low revenue base. So the two Gauteng-focused studies are not in complete agreement: one leans on management and politics, the other on money.

Bikam and Chakwizira (2021) came at the same problem from the planning documents. They reviewed sector plans, spatial development frameworks and integrated development plans for three Gauteng metros and four rural Limpopo municipalities, and used key informants to test the claim in the literature that poor operations and maintenance drag down basic services. Their abstract reports that insufficient funding and a low revenue base limit asset renewal and basic services in both the metros and the rural municipalities. They recommend better asset management systems and practices, which is close to the Ekurhuleni recommendations of upskilled senior managers, current asset registers and an integrated infrastructure programme (Bikam and Chakwizira, 2021; Patience and Nel, 2021). When a big metro with a R60 billion asset base and four rural municipalities end up with similar advice, the diagnosis of weak asset management is probably not specific to any one place.

In KwaZulu-Natal the evidence is about how municipalities treat people. Muthwa and Matsiliza (2024) surveyed about 300 residents of uMgungundlovu District Municipality and found gaps between expectations and experience on all five service-quality dimensions. Residents reported burst pipes left unrepaired for long periods, late water tankers in villages, and staff said to have been hired through personal or political connections. Twenty-four per cent strongly disagreed that staff behaviour instilled confidence. Pillay and Mutereko (2022) interviewed eleven officials in eThekwini and found an indigent policy undermined by a poorly maintained register, no dedicated office and no single approach across the electricity, water and finance departments, with political influence and corruption said to distort who is listed. Both studies describe administrative weakness at the point where the municipality is supposed to meet the poorest households.

Is money the real constraint?

The funding explanation is not wrong, and the studies give it support. Twalo (2025) says grants for water infrastructure are considered insufficient for the replacement backlog and that participants criticised the national department for handing over old infrastructure and then being reluctant to fund its replacement. Bikam and Chakwizira (2021) link under-performance to a low revenue base. Yet the evidence on revenue is less one-sided than that implies. Balie and Horn (2021), using national trends from 2001 to 2016 and a close look at six Western Cape municipalities, found that overall municipal revenue rose at a substantially higher rate than population and the number of indigent households. They studied revenue, not service outcomes, and the Western Cape is not the Eastern Cape, so the finding cannot simply be carried over. But it does cast doubt on the idea that growth in need has outrun growth in income everywhere.

Put the case studies next to that and a different reading emerges. Ekurhuleni recorded about R2.2 billion of unaccounted electricity and water losses in a single year and spent little on renewal (Patience and Nel, 2021). Chris Hani officials say bills often do not go out at all (Twalo, 2025). Pillay and Mutereko (2022) agree that budget deficits contribute to failure in eThekwini, but they argue that funding alone will not fix an indigent programme with no owner. In each case money is short partly because it is not collected, not protected for maintenance or not spent where it counts. That is not the same as saying funds are plentiful. Rural districts such as Amathole probably would struggle with any budget (Hutete and Sibanda, 2022). Still, the weakest links in the chain seem to be maintenance, billing and registers, which are management tasks, and they are cheaper to fix than a new treatment works.

Non-payment complicates things further, because it is both a cause and a result. Muthwa and Matsiliza (2024) report that about 71 per cent of their respondents were unemployed, students, housewives or pensioners, which limits what they can pay, and that distrust of billing, partly blamed on maladministration, feeds a culture of not paying. Twalo’s (2025) participants gave much the same list: inconsistent supply, high living costs, billing problems and no visible consequences for defaulting. Unpaid accounts then leave municipalities unable to pay bulk suppliers, which can interrupt supply. Free basic allocations of 6 kilolitres of water and 150 kilowatt-hours of electricity a month are supposed to shield the poorest, but only if the indigent register is accurate (Pillay and Mutereko, 2022). The loop is a vicious one, and breaking it needs working meters and honest registers before it needs more cash.

There is a fair objection to all of this. Much of the evidence is perception, collected from officials and residents who may blame management precisely because it is the part of the system they can see. Pillay and Mutereko (2022) spoke to eleven municipal employees, Patience and Nel (2021) drew on seven interviews alongside financial documents, and Sofianos and Matthews (2025) could not hear from the Makana municipality at all. The financial ratios in Ekurhuleni are harder evidence, yet they come from one city in one five-year window. The argument here is a claim about where the weight of these particular studies falls, and a different sample of municipalities could shift it.

Accountability ties the pieces together. Residents of Duncan Village and Amathole described consultation that produced no feedback, and councillors they did not know (Ndasana et al., 2022; Hutete and Sibanda, 2022). In uMgungundlovu, participation meetings reportedly produced promises without follow-through (Muthwa and Matsiliza, 2024). If the channels for complaint are hollow, municipalities lose the information that would tell them where pipes are bursting, and residents lose the reason to pay. That probably helps explain why service delivery protests persist even when capital budgets grow. The studies cannot prove this causal chain, and they were written at different times in very different places, so it should be treated as a well-supported hypothesis rather than a settled finding.

Conclusion

The cases examined here point in one direction. Municipal service failure in the Eastern Cape, Gauteng and KwaZulu-Natal shows up as dry taps, shared toilets and unreliable electricity, but behind it sit unmaintained assets, estimated or missing bills, registers that nobody keeps up and consultation that leads nowhere (Twalo, 2025; Patience and Nel, 2021; Ndasana et al., 2022). Money is part of the story, especially in poor rural districts, and nothing here shows otherwise. Even so, the Ekurhuleni losses and the Western Cape revenue evidence suggest that maintenance and revenue management deserve more weight than new capital grants (Balie and Horn, 2021). The evidence is limited by small qualitative samples and by the absence of comparable outcome data across municipalities. A fair next step would be to test the maintenance and billing argument against audited indicators for many municipalities at once. Until then, the sensible policy emphasis is on asset registers, billing, ring-fenced water revenue and complaint systems that actually answer.

Reference list

Balie, Q. and Horn, A. (2021) The impact of population growth on municipal revenue: implications for South African municipalities. Development Southern Africa, 38(6), pp. 1046-1058. doi:https://doi.org/10.1080/0376835x.2021.1975534

Bikam, P. and Chakwizira, J. (2021) Municipal asset operations and maintenance performance in metropolitan and rural municipalities in Gauteng Province and Vhembe District Local Municipalities, South Africa. Cogent Engineering, 8(1), article 1935409. doi:https://doi.org/10.1080/23311916.2021.1935409

Chiwarawara, K. (2024) The struggle for housing and basic services in South Africa: a case for service delivery protests. Social Dynamics, 50(2), pp. 148-165. doi:https://doi.org/10.1080/02533952.2024.2352193

Hutete, C. and Sibanda, M.M. (2022) Water service provision and social equity in a South African rural district municipality. Africa’s Public Service Delivery and Performance Review, 10(1), article a641. doi:https://doi.org/10.4102/apsdpr.v10i1.641

Muthwa, X. and Matsiliza, N.S. (2024) An assessment of customer perception of water service quality at uMgungundlovu municipality. Journal of Local Government Research and Innovation, 5, article a139. doi:https://doi.org/10.4102/jolgri.v5i0.139

Ndasana, M., Vallabh, D. and Mxunyelwa, S. (2022) Understanding the service delivery protests in South Africa: a case study of Duncan Village. Africa’s Public Service Delivery and Performance Review, 10(1), article a644. doi:https://doi.org/10.4102/apsdpr.v10i1.644

Patience, J.J. and Nel, D. (2021) Municipal infrastructure management and its impact on service delivery in the City of Ekurhuleni. Africa’s Public Service Delivery and Performance Review, 9(1), article a508. doi:https://doi.org/10.4102/apsdpr.v9i1.508

Pillay, B. and Mutereko, S. (2022) Caring for the indigent urban population in South Africa: a case study of the eThekwini municipality. Africa’s Public Service Delivery and Performance Review, 10(1), article a593. doi:https://doi.org/10.4102/apsdpr.v10i1.593

Sofianos, B. and Matthews, S. (2025) When service delivery runs dry: Gift of the Givers and water provision in the Makana Municipality. Politeia, 44(1), 17 pages. doi:https://doi.org/10.25159/2663-6689/19200

Twalo, T. (2025) Water sector value chain challenges: the case of Chris Hani District Municipality. Africa’s Public Service Delivery and Performance Review, 13(1), article a870. doi:https://doi.org/10.4102/apsdpr.v13i1.870