In the second week of January 2023, the remnants of ex-Tropical Cyclone Ellie dumped between roughly 200 and 800 millimetres of rain across the Kimberley in a matter of days, producing what the state's flood recovery agency has described as a once-in-a-hundred-year event. The Fitzroy River rose higher than at any point in living memory, submerging the bridge that carried the only sealed road north through Fitzroy Crossing and cutting communities along its banks off from the highway entirely. Hundreds of people were evacuated by air and by road from Looma, Noonkanbah, Muludja, Karnparrmi and Burawa, among others; homes that had already waited years for basic repairs were destroyed outright. More than a year later, modular houses were still being trucked in to replace what the water took, under the Kimberley Floods Housing Repair and Rebuild Program, jointly funded by Canberra and Perth (Kimberley Floods Recovery, WA Department of Fire and Emergency Services, ongoing; WA Government, "Fitzroy Crossing rebuild underway", 2024). The flood did not create Western Australia's remote-communities problem. It exposed one that had existed for a decade: essential services and housing maintained at the edge of what the state has been prepared to spend, in places for which the state alone has been financially responsible since 2015.
What the government did
That responsibility has a specific origin. In November 2014, then-Premier Colin Barnett announced that as many as 150 of the state's roughly 274 remote Aboriginal communities could face closure, after the Commonwealth government confirmed it would stop funding municipal and essential services in remote WA communities from mid-2015 (approx.; contemporaneous reporting of the Barnett government's November 2014 announcement). The threat triggered national protests and international attention, and Aboriginal leaders said at the time they had not been consulted before the closures were flagged publicly. That crisis was not of WA Labor's making, nor of Roger Cook's: it was a decision of a federal Coalition government and a state Liberal-National government that Labor, returned to office under Mark McGowan in March 2017, inherited along with its unresolved aftermath. In 2015 the Barnett government struck a deal to keep the lights and water on: WA took over full responsibility for essential services to remote communities in exchange for a $90 million exit payment from Canberra, extending the state-funded Remote Area Essential Services Program to around 165 communities that had previously relied on the jointly funded municipal services program (Office of the Auditor General WA, background to the 2015 and 2021 essential services reports).
What Labor did with that inheritance is the proper subject of this audit. A "Resilient Families, Strong Communities" roadmap, finalised in 2016 by the state's Regional Services Reform Unit, set out the framework the McGowan and then Cook governments have operated within since: continued state funding of remote power, water and wastewater services; a stated preference for concentrating investment in larger, more viable communities; and, on the government's own account, an expectation of "supporting fewer communities over time" as residents drift toward towns (WA Government, "Resilient Families, Strong Communities", Regional Services Reform Unit, 2016). The McGowan government established the $350 million Remote Communities Fund in the 2022-23 State Budget, aimed at housing, water and power upgrades after Canberra separately wound back its remote housing funding commitments (WA Department of Housing and Works, "Remote Communities Fund" program page). Roger Cook, Premier since June 2023, has since added to it: a $140 million housing top-up announced in December 2023, and a further $127 million and $82 million added across two funding lines in the 2026-27 State Budget handed down in May 2026 (Cook Labor Government media statements, 19-Dec-2023 and 07-May-2026). Cook's government is also drafting legislation to transfer the 20-million-hectare Aboriginal Lands Trust estate, comprising 284 land parcels, five pastoral leases and around 142 communities, to direct Traditional Owner control, intended to unlock home ownership and outside investment on Aboriginal land for the first time (reported by The Australian and National Indigenous Times, January 2024, citing WA Aboriginal Affairs Minister Tony Buti).
The strongest case
No other state pays this much, for this few, this far from anywhere
Put the government's defence at its strongest, because the underlying arithmetic is real. Western Australia is larger than Western Europe, and its remote Aboriginal communities are scattered across the harshest and most isolated parts of that landmass, hundreds of kilometres from the nearest town, often accessible only by unsealed roads that close for months every wet season. Trucking in diesel, maintaining desalination and bore-water treatment plants, and keeping tradespeople rostered to fly out and fix a burst pipe or a failed pump in a community of thirty people costs far more per household than the same service in Perth or Karratha, and no amount of budget discipline changes that basic geography. The state did not choose to become the sole funder of this system; it was left holding it when the Commonwealth walked away in 2015, and it has kept services running in communities a purely economic calculation might long ago have written off. Since taking that responsibility on, the state has kept power and water flowing with very few outages, has grown remote housing and infrastructure funding substantially rather than let it shrink with each budget cycle, and is now attempting a genuine structural reform, handing land tenure itself to Traditional Owners, rather than repeating Barnett's 2015 approach of announcing closures from Perth. That is a materially different posture from the government whose threat still shadows this debate, and it deserves to be judged as such.
The evidence
The primary record, chiefly the state's own Auditor General, shows a system that has improved in reliability while remaining seriously deficient in quality and unevenly monitored. The Auditor General's original 2015 report on the Remote Area Essential Services Program found that 68 of the communities then serviced had recorded at least one water-quality test failure in the two years to June 2014, and identified weaknesses in the Department of Housing's contracting and information-sharing (Office of the Auditor General WA, "Delivering Essential Services to Remote Aboriginal Communities", Report 8, May 2015). A follow-up audit tabled six years later found progress on reliability: power outages were restored within 24 hours 95 per cent of the time, up from 85 per cent in 2015, and both power and water services averaged fewer than one outage a year per community. But it also found that water contamination remained widespread. In the two years to 2019-20, 37 communities recorded contamination, including E. coli or Naegleria detections in 21 communities, elevated nitrates in 19, and elevated uranium in four. Most strikingly, the 51 smallest communities added to the program received no water-quality testing at all until November 2019, four years after they were brought under state responsibility (Office of the Auditor General WA, "Delivering Essential Services to Remote Aboriginal Communities — Follow-up", Report 25, 2020-21, tabled 2-Jun-2021).
The same follow-up audit flagged a funding structure that risks turning today's reliability into tomorrow's failure. Of $383 million in essential-services funding allocated across 2019 to 2024, only $14 million was earmarked for replacing ageing assets, even though the program's own manager had identified $84.4 million in immediate capital works needs and a further $103.5 million in medium-priority replacement work (Office of the Auditor General WA, follow-up report, 2021). The Auditor General also found that one of the 2015 report's central recommendations, expanding water-quality testing to all communities, remained only "accepted in principle" six years on, with implementation deferred pending a government funding decision. Water quality did improve markedly in some regions over the same period: contaminated Kimberley communities fell from around 26 to roughly 8, the trend shown below. But an improving average sits alongside a maintenance backlog the state has not yet chosen to fund.
The critique
The government's defence explains why this is expensive. It does not explain why, eleven years after Perth took full control of the essential-services task, the state's own auditor still found more than fifty communities going years without a single water test, or why a $383 million funding line could find only $14 million for the asset replacement its own program manager said was urgently needed. Reliability statistics measure whether the taps and generators kept working during the audit period; they say nothing about whether the water coming out of those taps was safe to drink, and on that measure the record is genuinely poor. A government that inherited a crisis in 2015 does not get an indefinite pass for how it has run the system since.
The pattern extends to housing. A Federal Court class action filed by Slater and Gordon in August 2024, on behalf of thousands of Aboriginal tenants of state-owned housing across the East Kimberley and Pilbara between 2010 and 2024, alleges properties inspected by the firm were found without functioning toilets, showers or safe drinking water, with exposed wiring, mould and inadequate fencing that let livestock and snakes onto the property (Slater and Gordon; reported by National Indigenous Times, 22-Aug-2024). Those allegations are yet to be tested in court, but they describe conditions consistent with the overcrowding data the state is failing to move. Nationally, the Productivity Commission's Closing the Gap dashboard found in July 2025 that the target of 88 per cent of Aboriginal and Torres Strait Islander people living in appropriately sized housing by 2031 is improving too slowly to be met on current trends, and remote WA communities are among the most affected by overcrowding of any region in the country (approx.; Productivity Commission, Closing the Gap Annual Data Compilation Report, Outcome Area 9, Jul-2025).
A government that took full responsibility for these communities in 2015 has still not funded the replacement of the assets it took on.
Consultation is the thread that runs through both the 2015 crisis and the current reform agenda. The 2016 roadmap that still frames policy states plainly that the government "expects to support fewer communities over time" as people move to larger towns — a managed-attrition logic that echoes the calculation behind Barnett's 2015 closure list, only quieter and spread across budget cycles instead of announced in one press conference. The Aboriginal Lands Trust land-tenure reform now before the Cook government is, on its face, a genuine self-determination measure: handing 20 million hectares back to Traditional Owner control is not a small thing, and ministers have pointed to consultation already undertaken with residents and native title holders. But the same reporting that describes that consultation also warns the changes could trigger a backlash on the scale of the one that sank the state's 2023 Aboriginal Cultural Heritage Act within weeks of it taking effect, suggesting the government's habit of designing major Aboriginal policy first and testing community consent afterward has not fully changed (reported by The Australian and National Indigenous Times, Jan-2024). Genuine self-determination is measured by who sets the pace and the terms, not only by who benefits from the outcome.
The risk to the future
An ageing system meeting a harder climate
The 2023 Kimberley floods were a preview, not an outlier. A remote essential-services network already running on a $14 million asset-replacement budget against an identified $84 million-plus of immediate need has little slack for a region where cyclones, flooding and extreme heat are becoming more frequent and more damaging. If funding for replacement, not just maintenance, does not catch up with the scale the Auditor General has already documented, the choice future governments face will not be between good and adequate service — it will be between crisis-driven emergency spending after the next flood or bushfire and a managed, planned upgrade now. Land tenure reform carries its own version of the same risk: handled without visible, community-paced consultation, it could hand opponents of Aboriginal policy reform in WA another rallying point, as happened with the Cultural Heritage Act, and set back the very self-determination goals it is meant to advance. The lesson of 2015 was that essential services to remote communities cannot be assumed; the lesson of the decade since is that neither can the money to look after them properly, once the crisis that created the headlines has faded.