Non-Revenue Water: Why Producing More Water Can Still Leave a Utility Weaker
A water utility can expand production and still remain financially and operationally weak when a large share of treated water disappears through physical leakage, inaccurate metering, unauthorized consumption or billing failures before it becomes collected revenue. The key systems question is not only how much water enters the network, but how much becomes reliable service, measured consumption and cash that can finance the next maintenance cycle.
The Brief in One Sentence
Water security is not created at the treatment plant alone: a utility becomes stronger when treated water survives the network, reaches users reliably, is measured correctly, is billed and collected, and the resulting cash returns to maintenance and renewal.
Why It Matters
The World Bank currently describes non-revenue water as a major operational and financial constraint in emerging and developing economies. Its Water for People knowledge catalogue says roughly 40% of water supplied in emerging and developing countries is lost or unaccounted for through leaks, theft and meter inaccuracies. That is a broad cross-country estimate, not a universal utility benchmark.
Recent Southern African work makes the mechanism concrete. At a March 2026 World Bank-supported workshop, all participating utilities reported NRW above 45%. Five utilities mapped an interim reduction toward 35%; the World Bank estimated that this could save 31 million cubic metres annually, enough to serve about one million people, and generate about US$57 million in annual financial savings. These are programme estimates, not audited realized savings.
The important point is not that every utility should target the same percentage. It is that water, energy, revenue, maintenance and service quality are coupled. A leak is therefore not only a hydraulic event.
Explain It Simply
Imagine a bakery that makes 100 loaves every morning but only 60 reach paying customers. Some fall off the delivery truck, some are taken without being recorded, and some are delivered but never put on an invoice. The owner responds by baking 20 more loaves.
That may help temporarily, but it does not fix the holes between the oven and the cash register. Worse, the bakery now buys more flour and electricity while the missing-loaf problem continues.
A water utility can face the same trap. More production is useful only when the network, meters, billing and collection system can convert that production into dependable service and sustainable cash.
System Boundary
This brief examines urban piped-water utilities, especially in emerging and developing economies where high NRW, constrained finance and service expansion interact. It follows water from abstraction and treatment through pumping, distribution, metering, billing and collection, then follows money back toward operations, maintenance and renewal.
NRW includes more than visible leakage. Physical or real losses come from leaking pipes, joints, tanks and failures. Apparent or commercial losses can include meter inaccuracies, unauthorized consumption and data or billing problems. Authorized unbilled consumption can also sit inside the standard water-balance concept. The exact composition matters because a meter problem is not repaired with the same intervention as a burst main.
This is not a claim that zero NRW is technically or economically optimal, nor that every litre classified as NRW could have been sold immediately.
Evidence Map
- Current World Bank cross-country estimate: about 40% of supplied water in emerging and developing countries is lost or unaccounted for. This is a broad portfolio-level statement, not a utility-specific measurement.
- Current Southern Africa programme evidence: participating utilities in the March 2026 workshop reported NRW above 45%; five mapped a reduction toward 35%, with estimated annual savings of 31 million m³ and US$57 million. These are projected benefits.
- Historical physical-loss estimate: a World Bank study cited in later Bank material estimated global physical water losses at 32 billion m³ per year, about half in developing countries. This older estimate is useful for scale, not as a 2026 measurement.
- Performance-contract evidence: World Bank material reports that a Ho Chi Minh City NRW performance-based contract reduced leakage by roughly 100,000 m³/day in part of the city; another Bank account reports 122 million litres/day conserved after six years. The different figures reflect different reporting points and should not be collapsed into false precision.
- Independent practitioner case: IWA published a 2026 account from Murang’a Water and Sanitation Company reporting NRW reduction from 58% to 24% over three years alongside digital asset mapping, sensing, metering and customer systems. This is an operator-authored case, not an independent causal evaluation.
The Loops That Matter
The deterioration loop: high losses reduce billable water and raise avoidable treatment and pumping cost → weaker cash generation constrains maintenance and renewal → ageing assets and weak information increase leakage, failures and billing errors → losses remain high.
The recovery loop: better water balances, district metering, pressure management, leak repair, meter accuracy and commercial controls reduce avoidable losses → more usable water and/or revenue becomes available → stronger cash generation supports maintenance and targeted renewal → network performance and data improve further.
Neither loop is automatic. Tariff policy, collection discipline, energy prices, governance, staff capability, asset age and local water scarcity can strengthen or weaken the feedback.
What Most People Miss
First: NRW is a composite indicator. Two utilities at 40% can have different problems and need different interventions.
Second: reducing physical leakage can create value even when the saved litre is not immediately sold. It can reduce abstraction, treatment, pumping and scarcity pressure, or improve continuity for existing users.
Third: adding supply can sometimes be necessary. A rapidly growing city may need new sources and NRW reduction at the same time. The mistake is treating new production as a substitute for fixing a distribution and revenue system that cannot convert water into dependable service.
Fourth: percentage NRW alone can mislead. Network pressure, connection density, supply continuity, pipe length, input volume and the mix of real versus apparent losses affect interpretation. Operational decisions need component-level measurements.
Sidy’s Synthesis
The useful unit of analysis is not the treatment plant. It is the complete conversion chain:
Source → Treat → Pump → Deliver → Measure → Bill → Collect → Maintain.
A utility leaks value whenever one of those handoffs fails. Physical leakage loses water and embedded energy. Bad metering loses information. Billing failures lose recognized revenue. Weak collection loses cash. Deferred maintenance then feeds the next round of physical and commercial loss.
Canonical principle: A utility is not strengthened by the water it produces, but by the water it can reliably convert into service, trustworthy measurement and reinvestable cash.
This is intentionally an unnamed synthesis. The mechanism matters more than creating another framework label.
Critical View
NRW reduction can be oversold. Some leakage is uneconomic to eliminate; locating the final increments can cost more than the water, energy and revenue they save. Apparent losses may dominate in one network while real leakage dominates another. Intermittent supply complicates pressure and leakage measurement. Weak tariffs or collection can leave a utility financially fragile even after technical losses fall.
Performance-based contracts can align incentives, but they require credible baselines, measurable zones, data integrity, procurement capacity and rules for changes in demand, pressure and network configuration. A contractor cannot repair governance by contract wording alone.
Digital technology is also not the thesis. Sensors, GIS and smart meters can improve visibility, but poor asset data, weak field response or badly designed incentives can turn dashboards into expensive observation systems.
What Would Reopen the Thesis?
- New comparable utility data showing that high NRW is no longer materially associated with operating cost, cash generation or service reliability in the target settings.
- Evidence that supply augmentation consistently outperforms loss reduction on lifecycle cost in water-constrained utilities with high avoidable NRW.
- Better decomposition showing that the current 40% emerging/developing-country estimate is materially outdated or methodologically non-comparable.
- Independent evaluations showing that performance-based NRW programmes systematically fail to sustain gains after contract completion.
- Material changes in metering, leak detection, pipe rehabilitation or distributed supply economics that alter the cost-optimal balance between reducing losses and adding supply.
Remember This
Do not ask only how much water a utility can produce. Ask how much of that water survives every handoff from source to service to cash—and whether the cash comes back quickly enough to keep the system healthy.
Primary sources
Facts, figures and quotations should be traceable to the sources below. Sidy's synthesis is labeled as synthesis and does not replace sourced facts.
- World Bank — Water for People knowledge products catalogue
- World Bank — Fix the Leaks: How Water Utilities Build Financial Sustainability
- World Bank — Performance-based contracts offer a pathway to efficient water management
- World Bank — What is non-revenue water?
- International Water Association — From Data to Decision
- International Water Association — Water Loss 2024