The number that reorders the problem
Water sessions about small islands usually run on scarcity, and scarcity usually points at a desalination plant. This one produced a different set of numbers. Jamaica’s Minister of Water, Environment and Climate Change told the room his country loses roughly 70% of its potable water to non-revenue water, one of the worst rates in the world. Jamaica is not a water-scarce country. It is a wet one.
Every other national figure on the table pointed the same way. St Kitts and Nevis, which its own delegate called one of the most water-scarce countries in the world, reported non-revenue water of about 50%, after having just built enough production capacity to exceed demand. Fiji put its distribution losses at 49%. Antigua and Barbuda, which has no rivers, no lakes and no permanent freshwater at all, reported non-revenue water fluctuating between 45% and 65%. Four islands with entirely different hydrologies reported the same failure at the same point in the system, and it is not the plant.
Fiji drew the conclusion the panel was circling.
Malta’s number, and what it bought
Malta was in the room as the worked example, and it is the only speaker that brought a before and after. Malta has no rivers and no mountains, its aquifer covers about a third of national demand, and uncontrolled abstraction had already salted that aquifer toward undrinkable. It went to reverse-osmosis desalination early, and more than 70% of its water now comes from desalination plants. That is the reflex answer, and Malta’s own account is that it did not work on its own. Its Chief Officer for Production and Treatment described building five more plants into a network that leaked so badly that up to 75% of the water produced was never billed, which starved the utility of the revenue it needed to run the plants it had just built.
So the strategy changed from producing more to keeping what was already produced. Malta cut its Infrastructure Leakage Index from 20 to 1.6 in 2025, an ILI of 20 having meant roughly 4,000 cubic metres per hour disappearing into the ground. The 1.6 works out to about 8% real losses. The method was unglamorous and repeatable: send people out with pipe locators to find out where the network actually is, cut the island into metered districts, manage pressure down (losses scale with pressure), and repair every detected leak immediately.
The payoff is the part that cuts against the build reflex. Malta’s CEO said the change in strategy let the country defer investment, avoid commissioning new plants, and take existing plants offline. The leak fix substituted for the build. Malta now runs four plants and is only beginning to look at a fifth, and it produces a cubic metre of water for about 2.8 kilowatt hours, which matters when energy is more than 30% of the corporation’s costs.
The same presentation was candid about where the curve ends. Malta’s remaining 8% real loss sits close to the point where the cost of finding the next leak exceeds the water saved, which he estimated at around 5% or 6%. Of Malta’s 20% total unbilled water, only 8% is physically leaking; the other 12% is apparent loss, meaning metering error and theft, which is a billing and enforcement problem rather than a pipes problem. That distinction is the kind of thing a country only learns after it has instrumented the network, and it is precisely what the countries reporting 50% and 70% cannot yet separate.
A live transfer, not a slogan
HLPF side events are full of offers to share lessons. This one had a receipt. Malta’s CEO framed the offer in the only terms that make peer learning an argument rather than a courtesy.
Antigua and Barbuda then read out what the bridge has actually carried. Technical cooperation with Malta began in 2022, under a memorandum of understanding between the two island states, when a team of five technicians travelled to Malta and found, in the mission’s own summary, that Antigua and Barbuda’s present challenges were the challenges Malta had already overcome. A second mission followed in 2025, working alongside Malta’s Energy and Water Agency and Water Services Corporation on leakage detection, pressure management and network optimisation. The cooperation has since extended into joint procurement and digital mapping of Antigua’s network, and the national priorities that came back from it are Malta’s playbook: pressure management zones, pressure-reducing valves, new metering technology, and better hydrological data.
Aruba, which has run on seawater desalination for close to a century, made the structural version of the same point: what small islands need is not the project, it is the relationship that outlives it.
The thing no grant buys
The quietest finding of the session was not about money or engineering. St Kitts and Nevis has done the building: a two-million-gallon-per-day desalination plant commissioned in 2026 at EC$50 million, a well integrated into the national system, an old plant decommissioned, national production now at 8 million gallons a day, which exceeds current demand. Its national water budget line has risen by more than 400% since 2021, and it expects roughly US$50 million from the Green Climate Fund in 2027 for the efficiency phase: smart metering, SCADA, and cutting that 50% non-revenue water.
Then it reported its willingness-to-pay survey. About 95% of households have piped water to the home. About 24% expressed any willingness to pay for reliable improvements. In Nevis, where water already costs more, the figure was about 6%.
That is a measurement of trust, and it is the binding constraint on the tariff reform that every other part of the plan depends on. A utility that cannot raise tariffs cannot fund the maintenance that stops the leaks, which is what would produce the visible improvement that earns the trust that permits the tariff. No Green Climate Fund grant closes that loop. The delegate’s own conclusion was that tariff reform “must therefore be approached as part of a broader social partnership”, and Singapore, speaking last, effectively confirmed the diagnosis from the other end: its water strategy has been “underpinned by sustained efforts to secure public buy-in”, with community education and consistent conservation messaging running for decades alongside the infrastructure. Malta’s version was a two-tier tariff that prices the first 33 cubic metres at around €1.60 and penalises consumption above it at more than €5, plus a quality argument: better-tasting tap water raised confidence, cut bottled-water use by more than two million plastic bottles last year, and brought revenue back to the utility.
What the room could not answer
The panel was unanimous that the know-how exists and is transferable. It was equally unanimous, and much vaguer, on why the money still does not move. Fiji named it directly: it is looking to the Green Climate Fund, the Adaptation Fund and regional partners for resilient infrastructure, and those mechanisms “remain slow”. Its framing of ABAS was that the gap is no longer diagnostic.
UNEP, as custodian of SDG indicator 6.5.1 on integrated water resources management, said the same thing from the global vantage point, and it is a striking admission from a measurement agency: the constraint it observes is not knowledge.
Nobody on the panel said who fixes the disbursement problem, and there was no time for an open discussion. The institutional answers offered were forums: a SIDS Centre of Excellence hosted by Antigua and Barbuda, an Island Investment Forum to be launched at the SIDS Global Business Network Forum in the Maldives from 23 to 25 August, Jamaica’s second Island Water Congress in January, Singapore’s biennial water week, and the 2026 UN Water Conference. All useful. None of them is a disbursement mechanism.
Our read
The implicit verdict of this session is that a meaningful share of SIDS water finance is aimed at the wrong end of the pipe. Production is the visible, fundable, ribbon-cutting half of the problem; leakage is the invisible half, and on the numbers presented here it is the larger one. Malta is the proof in both directions: it built its way into a financial crisis, then fixed the network and found it could decommission plants and defer the next build. St Kitts and Nevis has just made national production exceed national demand while half the water still disappears before it is billed.
The measurement point follows directly. Non-revenue water is not one number, it is two, and only Malta on this panel could split real losses (8%) from apparent ones (12%). Until a utility can, it cannot know whether it is buying pipes or meters, and “70%” is a symptom rather than a diagnosis. That is the transferable asset in the Malta to Antigua channel, and it costs far less than a plant.
The tension the session did not resolve is that the cheapest fix is the one with the least political constituency. Leak repair delivers no launch, and it depends on tariff revenue from households that, in St Kitts and Nevis at least, are three quarters unwilling to pay for a promise. Jamaica’s minister made the case for water as the base of everything else. The panel made the case that the hardest input is not water, technology or even finance, but the confidence of the person at the tap.
Why it matters for the SDGs
This is SDG 6 seen from the states with the least margin for error, and it lands on the same fault line as the Forum’s own water sessions. The SDG 6 Synthesis Report launch concluded that water is well counted and running six to eight times too slow, with the binding constraint in finance and governance rather than data. This panel is what that looks like at national scale, with one addition: some of the missing acceleration is already inside the network, unbilled and unmeasured. It also puts a hard operational number under the argument the SIDS review and the State of SIDS report made politically about vulnerability and access to concessional finance.
The linkages are not decorative here. Jamaica reported three of its four worst droughts on record inside five years, and in 2022 received 8% of the 30-year average rainfall in what is usually the driest month, which is SDG 13 arriving as a water-planning problem. Antigua and Barbuda reported reservoirs that once produced nine months of water a year now producing three to five, with some years of complete depletion. Malta’s reclaimed class A water now serves more than 1,500 farmers and takes pressure off the aquifer, which is SDG 2 and SDG 12. Desalination is roughly a third of Malta’s energy bill, which is SDG 7. And the ABAS-shaped ask, predictable finance and real technology transfer rather than more diagnosis, is SDG 17. The next markers are the SIDS Global Business Network Forum in August and the December 2026 UN Water Conference, where the question is whether “fix the leaks first” is fundable at the same speed as a plant.
Watch & read
- Recording of the side event (UN Web TV, 16 July 2026).
- The SDG 6 special event, where UN-Water’s third synthesis report put global water progress six to eight times off pace, and the SDG 6 in-depth review that opened the Forum.
- The SIDS review (HLPF 5th meeting) and the first State of SIDS report, the political frame this session’s engineering sits inside · the SIDS partnerships steering committee.
- SIDS4 (Antigua and Barbuda, 2024), where the Antigua and Barbuda Agenda for SIDS (ABAS) was adopted, and the UN’s SIDS topic page.
- UN-OHRLLS, the co-convening office, which supports the 92 most vulnerable countries and the SIDS Centre of Excellence.
- 2026 UN Water Conference (Abu Dhabi, 8 to 10 December 2026), the forum every speaker pointed to next.
- Full HLPF 2026 coverage · HLPF 2026, official site.
Speakers are cited by role and country rather than by name. Country figures are as each country reported them and were not independently verified. The room is not confirmed by any official source and is not stated here. Quotations are lightly edited from an automated transcript of the UN Web TV recording and should be read as close paraphrase.