Independent digital publicationZimbabwe · South Africa · Wider African region
The Kopje Journal & ReviewForensic analysis for the African perspective.

Audit

A Tap Is Not a Reliable Water Supply: South Africa’s Service Gap

South Africa’s household survey records a wide gap between access to taps and dependable supply. Provincial comparisons, regulatory findings and the latest Western Cape allocation cuts show why connections, continuity, water quality and resource security must be assessed separately.

The audit question

A Tap Is Not a Reliable Water Supply: South Africa’s Service Gap

South Africa’s water statistics can describe a household as having access to a tap without establishing that water arrives when it is needed. That is not a minor technical distinction. It separates the existence of a delivery route from the service that route actually provides.

The Northern Cape illustrates the difference. In Statistics South Africa’s 2025 General Household Survey, 95.5% of households had access to piped or tap water in their dwelling, on their property or off-site. Yet 56.1% of households receiving municipal water reported interruptions under the survey’s longer-disruption measure. In the Western Cape, tap access was 99.5%, while the municipal-water interruption measure was 6.2%. Statistics South Africa, GHS 2025, figures 9.1 and 9.6.

These are separate indicators with different denominators: tap access covers households generally, while the interruption measure concerns households receiving municipal water. They nevertheless expose the weakness of treating infrastructure coverage as the final test of delivery. A near-universal tap-access figure can coexist with substantial disruption.

Four questions, not one

A useful assessment asks four things: where water can be obtained; whether it is available when needed; whether it is safe; and whether enough water remains available to sustain the system. Success on one does not settle the others.

The WHO/UNICEF Joint Monitoring Programme formalises part of this distinction. Its “safely managed” drinking-water category requires an improved source on the premises, availability when needed, and freedom from specified contamination. An improved source alone does not establish all three conditions. A basic service can involve collecting water away from home, provided the round trip, including queuing, takes no more than 30 minutes. JMP drinking-water service definitions.

This matters when comparing apparently conflicting headlines. A claim about tap access, a claim about basic services and a claim about safely managed water may all describe different thresholds. They should not be substituted for one another. Nor does an interruption automatically prove contamination: continuity and water quality require different evidence.

Provincial averages conceal different failures

The 2025 survey puts national piped or tap access at 87.4%, including off-site sources. Limpopo’s 63.2% and the Eastern Cape’s 67.7% contrast sharply with Gauteng’s 98.3%. The longer-disruption indicator, meanwhile, reached 61.5% in North West and 55.7% in Mpumalanga, against 32.5% in Gauteng. These are household estimates: the access figures cover households generally, while interruption figures concern municipal-water recipients. They are not population shares or counts of failed treatment plants. GHS 2025, figures 9.1 and 9.6.

The comparison points towards two tasks that overlap but are not identical. Where access is low, expanding the reach of supply remains essential. Where access is high but interruptions are widespread, adding another connection cannot by itself demonstrate that the existing service has improved. Provincial totals also cannot identify which municipality, neighbourhood or supply zone needs which intervention.

Stats SA’s accompanying presentation charts the national longer-interruption measure for municipal-water recipients rising from 24.3% in 2012 to 37.6% in 2025. Its label describes interruptions exceeding two days at a time or totalling 15 days over the preceding year. The presentation and main report use slightly different threshold wording, so those figures should not be read as a precise measure of every outage hour. GHS 2025 presentation, slide 31.

Coping is not restoration

The same presentation records stored water as an alternative source for 43.7% of households affected by interruptions lasting two days or longer; municipal tankers accounted for 22.1%, and vendors 7.7%. These percentages concern affected households, not all South African households. GHS presentation, slide 33.

An emergency arrangement and a repaired network are different achievements. A tanker can help meet immediate demand without resolving why the normal supply stopped. Stored water can bridge an interruption without showing that the next interruption will be shorter. Counting either as evidence of household resilience is not the same as counting it as restoration of dependable municipal service.

The available national figures also do not establish each household’s coping costs. A convincing account of that burden would require records of purchases, collection time and the duration of interruptions, rather than assuming that every affected household buys the same quantity of water at the same price.

More than concrete and pipes

The Department of Water and Sanitation’s December 2023 regulatory release offers a warning against treating every failure as a shortage of physical assets. Its Blue Drop assessment covered 958 water supply systems across 144 water services authorities. Although 85% of systems had infrastructure rated average or better, the department identified staff skills and process controls as important contributors to performance. DWS, release of the 2023 Blue Drop report.

That finding is historical, not a November 2026 safety certificate. It does, however, explain why a construction programme and an operating programme need separate scrutiny. An asset must be maintained and competently operated after it has been built.

The accompanying No Drop findings put non-revenue water at 47% in 2023, compared with 37% in 2014. This measure includes physical losses and commercial problems such as metering and billing weaknesses; it is not a claim that 47% simply leaked from pipes. DWS 2023 regulatory release.

Better performance is not immunity

The Western Cape’s comparatively favourable survey result cannot be projected indefinitely. On 16 November 2026, Water and Sanitation Minister Pemmy Majodina announced allocation reductions for the Western Cape Water Supply System: 7% for domestic and industrial use and 15% for irrigation. The department described these as precautionary measures following its annual operating analysis, while saying the system could meet current requirements. Minister’s statement, 16 November 2026.

An allocation reduction is not equivalent to a household outage. The announcement concerns management of a shared resource; the survey describes reported service experience during an earlier period. Keeping those distinctions intact avoids two mistakes: declaring the Western Cape immune because of its past performance, or treating a precautionary allocation decision as proof that taps have already stopped running.

What would demonstrate improvement?

Government’s proposed response recognises that institutional performance matters. The 2025 Water and Sanitation Indaba resolutions called for professionally managed, ring-fenced utilities and clearer separation of oversight from service provision. Those were commitments, not evidence that the changes had already restored supply. DWS, Indaba declaration and resolutions.

The 2026 Budget Speech subsequently announced R27.7 billion over the medium term for performance-linked reform covering metropolitan electricity, water, sanitation and waste services. That is not a water-only allocation. It also described improvement plans intended to retain and reinvest service revenue. National Treasury, 2026 Budget Speech.

The test is therefore not whether another reform has been announced, but whether its operating results become visible: fewer and shorter interruptions, credible water-quality results, functioning maintenance and wider dependable access. Those outcomes must be assessed separately, with consistent definitions and dates.

For a local audit, that means matching the household experience to the supply system serving it. A provincial average cannot establish that a particular repair worked. The useful comparison would follow the same area before and after the intervention, record the number of households served, distinguish planned maintenance from unplanned failures, and measure how long normal supply took to return. A change in the survey average would then be supporting evidence rather than a substitute for operational records.

There is also a reporting trap in counting new connections without showing what happened to existing ones. A delivery ledger should make both visible: additional households reached and the continuity experienced by households already within the network. Otherwise, expansion and deterioration can proceed at the same time while a single headline number conceals the difference. This is an analytical test for future reporting, not a claim that such a matched national ledger has been assembled here.

South Africa’s service gap cannot be reduced to a missing tap or an empty dam. The evidence supports a narrower, more useful conclusion: measuring access is necessary, but judging delivery requires proof that the supply works beyond the day a connection is counted.

Evidence & accountability

Sources & editorial status

  1. Statistics South Africa, GHS 2025, figures 9.1 and 9.6 ↗Source
  2. JMP drinking-water service definitions ↗Source
  3. GHS 2025, figures 9.1 and 9.6 ↗Source
  4. GHS 2025 presentation, slide 31 ↗Source
  5. GHS presentation, slide 33 ↗Source
  6. DWS, release of the 2023 Blue Drop report ↗Source
  7. DWS 2023 regulatory release ↗Source
  8. Minister’s statement, 16 November 2026 ↗Source
  9. DWS, Indaba declaration and resolutions ↗Source
  10. National Treasury, 2026 Budget Speech ↗Source
Source note

By The Kopje Journal & Review. Original editorial research; sources linked in the article.

Send a correction or source ↗

Cookie choices

You can read every audit with optional media switched off. Your choice applies to this browser for up to 180 days.

Essential website functions

Security, Publisher sign-in and remembering your privacy choice.

Always active

No KJR advertising or analytics tags run on this version of the website. Social links and Stripe checkout open only when you follow a link.

Cookie policy · Privacy statement · European & UK rights