Labor states and the debt mountain

A retired couple in Ballarat bought a small investment unit in 2019, a modest top-up for their pension. They had never paid land tax in their lives — their family home was exempt, and the unit's site value sat comfortably under the old $300,000 tax-free threshold. In January 2024 a new bill arrived. The threshold had been cut to $50,000, and a flat surcharge now applied on top of the ordinary rate. Their bill went from zero to real money, and it will keep coming every year until 2033, whether the unit is tenanted or not.

They are not being punished for anything they did. They are paying down debt someone else ran up — debt that funded genuinely popular things: new train lines, new hospitals, wage support for businesses that would otherwise have folded during lockdown. Victoria calls the new charges its "COVID Debt Repayment Plan," legislated to run until 30 June 2033 and designed to retire $31.5 billion of pandemic-era borrowing (approx.; State Revenue Office Victoria, 2023–24 Budget measures). But Victoria's net debt was never going to stop at $31.5 billion, and it hasn't. It is now the largest state-government debt load in the country, and it is still climbing. Queensland and Western Australia tell a more complicated version of the same story — one state riding a royalties boom to the strongest balance sheet in the nation, the other now working through the bill left by a run of Labor governments that lost office in 2024. This is the state-level mirror of the federal fiscal thesis this site sets out elsewhere: money borrowed against the future has to be repaid by someone, and the question is always who, and for what.

What Labor's state governments built

Victoria has been governed continuously by Labor since November 2014, under Daniel Andrews and, since September 2023, Jacinta Allan. Over that decade the government ran the largest transport and social-infrastructure program in the state's history, branded "Victoria's Big Build." It bundles the Metro Tunnel under central Melbourne, the West Gate Tunnel, the North East Link freeway, the Suburban Rail Loop, Regional Rail Revival upgrades, and a long list of new hospitals and school rebuilds. By the government's own account the pipeline has employed more than 50,000 workers at a time and represents a combined project value in the order of $100 billion or more (approx.; Victorian Government, Big Build project disclosures). Separately, Victoria endured some of the longest cumulative COVID-19 lockdowns of any jurisdiction in the world, and the state government layered its own business support on top of Commonwealth JobKeeper payments — more than $10 billion in direct grants, tax deferrals and cashflow support between 2020 and 2021 (approx.; Victorian Government business-support announcements, 2020–21).

Queensland ran a parallel build under a Labor government that held office from 2015 until it lost the October 2024 state election to the Liberal National Party under David Crisafulli. Its pipeline included the Cross River Rail underground line through Brisbane, a large hospital capital works program, an ambitious (and since-scaled-back) publicly owned renewable-energy and pumped-hydro build, and — from 2021 — the successful bid to host the 2032 Brisbane Olympic and Paralympic Games, which brought its own multi-decade venues and transport bill. Western Australia, under Labor since March 2017 (Mark McGowan, then Roger Cook from June 2023), funded its own Metronet rail expansion and hospital upgrades, but did so while running an extraordinary run of budget surpluses fuelled by iron-ore royalties — the seventh consecutive operating surplus was recorded in the 2024–25 Annual Report on State Finances (WA Department of Treasury, September 2025).

The scale is real. Victoria's population passed seven million in December 2025 (ABS, National, State and Territory Population), and Melbourne absorbed a large share of Australia's post-pandemic migration intake; a growing city needs new hospital beds, new train capacity and new schools regardless of which party governs it. Every project named above appears, in isolation, in the kind of infrastructure plan an economist would recommend for a fast-growing state. The question this article puts to the record is not whether Victoria, Queensland and Western Australia needed to build — it is whether what they told the public these projects would cost bore any reliable relationship to what they actually spent.

All three governments describe this era in similar terms: catching up on decades of underinvestment in fast-growing states, while cushioning households and businesses through the sharpest economic shock in a generation. The record varies enormously in how it has turned out.

The evidence: debt, interest and the projects behind it

Victoria's net debt — what the state owes after netting off its financial assets — was budgeted at $167.6 billion for 2025–26, rising to about $194 billion by 2028–29 under the state's own forward estimates (Victorian 2025–26 State Budget, Budget Paper No. 2, and 2025–26 Budget Update, February and December 2025). As a share of Gross State Product (GSP, the value of everything the Victorian economy produces in a year — the state-level equivalent of GDP), net debt sits at 25.1 per cent in 2025–26 and is forecast to peak at 25.2 per cent the following year, before easing only slightly (Victorian Budget Papers, 2025–26). On a per-person basis, that works out to roughly $23,400 for every Victorian, adult or child (approx.; author's calculation from Budget Paper net-debt figures and ABS population data, Victoria's population passed 7 million in December 2025). No other Australian state carries a comparable debt load relative to the size of its economy or its population.

Victoria's debt was not always this size. Net debt sat at around $25 billion in 2019, before the pandemic — a level Victorian treasurers of both parties had kept broadly stable for years. It has since grown roughly sixfold, past $150 billion in recent budget outcomes and heading toward $200 billion just beyond the current forward estimates. Independent analysis by the e61 Institute, a Sydney-based economic research body, estimates that only around $75 billion of that increase is directly attributable to COVID-19 spending and revenue lost during the pandemic (approx.; e61 Institute analysis of Victorian Budget papers). The remainder reflects other spending decisions made across the same years, chiefly the Big Build pipeline described above. Even if the pandemic had never happened, in other words, Victoria's net debt would very likely still have grown by tens of billions of dollars over this decade.

Debt has to be serviced. Victoria's finance costs — the interest paid on what it owes — were reported at around $6.8 billion in 2024–25 and are budgeted to climb toward $10.6 billion a year within the forward estimates as both debt and interest rates work through the numbers (approx.; 2025–26 Victorian Budget and Budget Update). Government and opposition figures in Parliament have both cited daily equivalents in the order of $18 million to $28 million, depending on the year and which budget update is used (approx.; Hansard, Parliament of Victoria, 2025). On the e61 Institute's analysis of the state's own budget papers, Victoria now spends more each year on interest than it spends running Victoria Police, and interest costs are forecast to grow by more than 40 per cent over the four years to 2028–29 — faster than any other line in the state budget, including health and education (approx.; e61 Institute analysis of Victorian Budget papers). Put simply: Victoria now spends more servicing debt each year than many mid-sized government departments' entire budgets, and that bill is still growing faster than the economy that has to pay it.

The rating agencies moved early. S&P Global Ratings cut Victoria two notches from AAA to AA in December 2020, citing the fiscal damage from the state's second COVID wave and its 112-day lockdown; Moody's followed, stripping Victoria of its Aaa rating in February 2021 and cutting it a further notch, to Aa2, in 2022 (S&P Global Ratings and Moody's Investors Service rating actions, 2020–2022). Both ratings have since been affirmed at that lower level rather than cut further, and both remain solidly investment-grade — but Victoria has held the lowest credit rating of any Australian state or territory continuously since that first downgrade. A lower rating is not just a symbol. It generally means a government pays a slightly higher interest rate to borrow, because lenders price in the marginally higher risk — a small, compounding tax on every future dollar Victoria borrows.

Some of that debt funded assets that are now delivering real service to the public: the West Gate Tunnel finally opened on 14 December 2025, having been budgeted in 2015 at $5.5 billion and ultimately costing somewhere in the vicinity of $10 billion once a dispute with contractors over contaminated soil was resolved (approx.; project cost disclosures reported against the original 2015 contract value). But the pattern across the Big Build is one of consistent, large cost growth. The North East Link freeway, approved at around $10 billion, is now expected to cost substantially more — government disclosures and independent analysis put the range anywhere from the high teens to more than $25 billion once scope additions, financing costs and a rail and busway extension are included (approx.; Major Transport Infrastructure Authority disclosures and media reporting on project cost revisions). The Metro Tunnel, originally budgeted at around $11 billion, has needed at least a further $1.36 billion since (approx.; Victorian Government project cost disclosures). The Suburban Rail Loop is the starkest case: the Victorian Parliamentary Budget Office put the nominal lifetime cost of building and operating just its first two sections, East and North, at $200.3 billion in 2022, and updated that to $216.7 billion in more recent advice, covering the period from 2019 to 2084 (Victorian Parliamentary Budget Office, Suburban Rail Loop East and North — cost to build and operate). The government's own Business and Investment Case, released in August 2021, covers only those first two sections; it has never published the full underlying cost workings for the remaining stages of the project it originally announced, a transparency gap independent analysts, including the Grattan Institute, have criticised directly.

A state that could not afford to host the Commonwealth Games spent $589 million cancelling it.

No single episode illustrates the pattern better than the 2026 Commonwealth Games. Victoria announced in March 2022 that it would host a regional, multi-city 2026 Commonwealth Games, with Cabinet approving a budget of $2.6–2.7 billion — a figure the state's own Auditor-General later found was based on "unrealistic assumptions" even though it had been presented to Cabinet as a best-case scenario. By August 2023 the government said costs had grown to $6.9 billion and withdrew from hosting altogether — but the Victorian Auditor-General's Office found that figure itself was "overstated and not transparent," containing double-counted contingency allowances; a more accurate estimate at the time of withdrawal was closer to $5.9 billion (Victorian Auditor-General's Office, Withdrawal from the 2026 Commonwealth Games, 20 March 2024). Withdrawing did not make the cost disappear. The audit found the decision to pull out cost Victorian taxpayers more than $589 million with no discernible benefit: $380 million to settle the host contract with the Commonwealth Games Federation, $112 million in departmental costs already incurred, $38 million for an Organising Committee that never ran a Games, and $42 million on planning work by the state's development agency. The Auditor-General also found that the departments responsible did not give ministers "frank, full and timely advice" about the project's feasibility until mid-2023 — more than a year after Cabinet had committed to it.

Queensland's numbers tell a messier story, partly because the state changed government mid-stream and partly because different reports use different debt measures — general government, non-financial public sector, gross versus net — that are not always directly comparable. What is clear is the direction. The outgoing Labor government's final budget update forecast total debt reaching as high as $252 billion by 2028–29 on some measures; the incoming Crisafulli government's first budget, handed down in June 2025, revised the trajectory down but still projected gross debt of roughly $205.7 billion by 2028–29, with its December 2025 Mid-Year Fiscal and Economic Review putting net debt at about $204.9 billion by June 2029 (approx.; Queensland Budget 2025–26 and 2025–26 Mid-Year Fiscal and Economic Review, Queensland Treasury). Queensland's independent Auditor-General, in an April 2025 report, found net debt for the state's non-financial public sector had stayed relatively stable from 2015–16 through 2023–24 before beginning a sharp rise from 2024–25 onward — a finding from the state's own auditor, not from either side of politics (Queensland Audit Office, Managing Queensland's Debt and Investments 2024, Report 12: 2024–25, April 2025). S&P Global Ratings, which holds Queensland at AA+ (Aa1 at Moody's), revised its outlook on the state from stable to negative in early 2025, a judgement on the fiscal trajectory the outgoing government had left behind.

Queensland's own flagship project shows an identical pattern to Victoria's Big Build. Cross River Rail, an underground rail line through central Brisbane, was promised by the Palaszczuk Labor government at a cost of $5.4 billion, with services running by 2024. In December 2024, within weeks of taking office, the incoming Crisafulli government said the true cost had reached more than $17 billion; the official figure has since been put at $19.041 billion, more than three-and-a-half times the original promise (Queensland Government ministerial statements, December 2024, and subsequent project cost updates). The tunnel and its ongoing maintenance costs alone grew from $5.4 billion to more than $10.5 billion; a further $5 billion was needed for integration work — new signalling, stabling yards, replacement bus services and connections to the existing network — that the incoming government says the previous administration had not properly disclosed; and delays and contractor claims added roughly another $1.5 billion (approx.; Queensland Government ministerial statements, 2024–25). First passenger services are now expected in 2029, five years later than originally promised. Whichever government's account of who under-costed what is accepted, the underlying fact is not contested: a rail line sold to Queenslanders at $5.4 billion will cost close to four times that, and arrive years late.

Queensland's Pioneer-Burdekin Pumped Hydro project shows how far a single business case can drift from reality. The former Labor government proposed it as the cornerstone of the state's renewable-energy firming strategy, costed publicly at $12 billion. Within 48 hours of taking office in October 2024, the incoming Crisafulli government said its own Treasury briefing put the true cost at $36.77 billion — more than triple the public figure — and cancelled the project, citing cost, environmental concerns and a lack of consultation with Traditional Owners (Queensland Government ministerial statement, 2024). On the government's own Treasury population projections, proceeding at that revised cost would have worked out to roughly $15,653 for every Queensland household (approx.; Queensland Government ministerial statement, 2024). Whatever its merits as energy policy, a project publicly costed at $12 billion that was privately understood to cost more like $37 billion by the time of a change of government is the same transparency failure this article has already documented in Victoria, in a different sector.

Queensland's 2032 Brisbane Olympics venues plan shows the same cost dynamic that produced Victoria's Commonwealth Games write-off, just with a different ending. The redevelopment of the Gabba stadium, first floated under Labor at around $1 billion, had grown to an estimated $2.7 billion by February 2023. After the change of government, Premier Crisafulli ordered a 100-day independent review, which found extending the Gabba's life to and beyond 2032 would itself cost close to $1 billion, on top of hundreds of millions just to keep it usable in the meantime. Rather than proceed with the Gabba rebuild, the new government announced in March 2025 that it would demolish the stadium after the Games and instead build a new 63,000-seat venue at Victoria Park, at an estimated cost of around $3.8 billion — a bigger, more expensive project than the one it inherited, and one that broke Crisafulli's own pre-election promise not to build a new stadium (approx.; Queensland Government announcements and independent stadia review, March 2025).

Western Australia sits at the other end of the spectrum. Total public-sector net debt was $30.2 billion at 30 June 2025, more than $13 billion lower than the previous Liberal-National government had forecast when it lost office in 2017, and net debt as a share of GSP is around 7.5 per cent — well under a third of Victoria's ratio, and forecast to stay below 10 per cent across the forward estimates (approx.; WA 2024–25 Annual Report on State Finances, WA Department of Treasury, September 2025, and independent analysis of state budget papers). On roughly 3 million residents, that is around $10,000 of net debt per person — less than half Victoria's figure. The 2025–26 Budget projects an operating surplus of $2.5 billion, the state's seventh consecutive surplus. Moody's reaffirmed Western Australia's Aaa rating in May 2026 and S&P Global reaffirmed its AAA rating in September 2025; Western Australia is the only Australian state or territory currently holding the top rating from both agencies simultaneously.

The critique: who pays, and what it crowds out

Start with the plainest fact: interest is not a service. Every dollar Victoria spends on finance costs — heading toward $10.6 billion a year (approx.) — is a dollar that funds no nurse, no teacher, no train driver, and builds no asset. It simply repays lenders for money already spent. As that bill grows faster than the economy funding it, it structurally narrows what any future Victorian budget, of either political colour, can afford to do for schools, hospitals or policing without either raising taxes or borrowing still more. The COVID Debt Repayment Plan is Victoria's own government acknowledging this arithmetic: it introduced a payroll-tax surcharge on large employers and cut the land-tax tax-free threshold from $300,000 to $50,000, specifically to find new recurring revenue to service debt (State Revenue Office Victoria, COVID Debt Repayment Plan, 2023–24 Budget measures). Land tax on investment properties is a cost landlords typically pass through, in whole or in part, to tenants over time — meaning a debt run up during the pandemic is now flowing into rents years later, on top of the separate rental pressures documented elsewhere on this site.

The project cost overruns are not a side issue from the debt — they are a major cause of it. When a $5.5 billion tunnel becomes a $10 billion tunnel, when a $2.6 billion Games budget becomes a $589 million write-off for nothing, when a rail loop's full cost is never published at all, the gap between the borrowing a government told voters it needed and the borrowing it actually undertook has to be filled somewhere, and it is filled with more debt. The Auditor-General's own language on the Commonwealth Games — a business case that "underestimated the costs and overstated the benefits," and departments that failed to give "frank, full and timely advice" — describes a decision-making culture, not a one-off accident. The Suburban Rail Loop was approved and construction begun on its first stage before the government published a full business case covering the project it had actually announced; the Parliamentary Budget Office, an independent body designed to give Parliament exactly this kind of costing, has had to build much of the public cost picture itself, years after the political commitment was made. A government that commits to a $200 billion, six-decade project before publishing the underlying costings has made a decision that later parliaments, and later taxpayers, cannot meaningfully unwind.

Queensland's experience adds a second layer to the same problem: contested numbers. When a state's own auditor has to publish a report clarifying how much debt the government actually carries, and outgoing and incoming governments cite figures that differ by tens of billions of dollars for the same forward year, voters lose the ability to hold anyone accountable for the trajectory in real time. That is true regardless of which side of politics is in office — Queensland's new Olympics stadium plan, settled on by the government that inherited Labor's Games bid, is itself now a bigger and more expensive project than the one it replaced. The debt-funded, cost-overrun pattern identified in Victoria under a continuing Labor government has, in Queensland, survived a change of government entirely.

There is also a structural mismatch between how these decisions are made and who pays for them. State elections run on three- or four-year cycles; the debt behind the Suburban Rail Loop is costed out to 2084, and interest on Victoria's borrowing will still be compounding long after every minister who approved the projects behind it has left politics. A government facing re-election within eighteen months has every incentive to announce a popular project at an optimistic headline price, and comparatively little incentive to publish the full multi-decade cost before voters go to the polls — the Commonwealth Games and the Suburban Rail Loop both fit this pattern precisely. Voters, in turn, cannot easily punish a government at the ballot box for a cost blow-out that only becomes visible in an Auditor-General's report tabled a year after the relevant election, or in a business-case update published a decade into a sixty-year project. None of the checks meant to catch this in real time — a Parliamentary Budget Office costing, an Auditor-General's audit, a rating agency's downgrade — operates fast enough to change the decision before the money is committed. Each can only tell the public, after the fact, how large the gap between the promise and the reality turned out to be.

What they got right — and the honest caveats

None of this amounts to a state defaulting on its debt, and it would be wrong to suggest otherwise. Every state discussed here retains an investment-grade credit rating, several notches above the point at which a government would struggle to borrow at all. State debt, unlike a household's, does not need to be repaid on any fixed schedule; it can be rolled over indefinitely provided lenders retain confidence in the borrower, and long-lived infrastructure assets are conventionally financed over decades, not years — a hospital or a rail line will still be serving Victorians in 2060. Some of the cost growth reflects genuinely global pressures, not local mismanagement: construction costs rose sharply across the developed world after 2020, driven by materials, energy and skilled-labour shortages that hit every large infrastructure project, public or private, in the same period. The ABS's own producer price index for road and bridge construction — a measure of what governments actually pay contractors — rose by close to a fifth cumulatively between early 2021 and mid-2023 (approx.; ABS Producer Price Indexes, Road and Bridge Construction series, 2021–2023), and every state government building transport infrastructure through that window absorbed some version of that shock, regardless of which party was in office or how disciplined its costing process was. Victoria's pandemic business-support spending, while expensive, was broadly consistent with what every state and the Commonwealth did to avoid a deeper downturn, and the state did keep unemployment from rising as far as some forecasts feared. Western Australia's turnaround shows the underlying model is not inherently broken — a Labor government there converted a resources boom into the strongest state balance sheet in the country, though it is worth noting plainly that WA's position rests heavily on iron-ore prices and a 2018 Commonwealth GST floor deal it did not itself design, factors a change in the commodity cycle or in the GST agreement could unwind. The projects themselves are also, increasingly, real and operating rather than merely promised: the West Gate Tunnel and the first stage of Cross River Rail are due to be carrying passengers and traffic within the life of this article, and Victoria's and Queensland's hospital and school rebuilds have added genuine capacity in fast-growing outer suburbs that badly needed it. A reader is entitled to weigh that delivered capacity in the balance against the cost overruns documented above — the argument here is that both are true at once, not that the projects were worthless because they ran over budget. And even Victoria's harshest infrastructure critics generally do not dispute that the underlying transport and hospital capacity was needed; the argument is about cost control and transparency in how it was delivered, not about whether it should have been built at all.

One further caveat cuts against Queensland's new government, not for it. An incoming government publicising the worst possible version of the figures it inherited is a familiar move in Australian politics, on both sides, and the $36.77 billion pumped-hydro number and the $252 billion debt forecast both originate from briefings and reviews the Crisafulli government itself commissioned and released. That does not make them wrong — the Queensland Audit Office's independent finding that net debt began rising sharply from 2024–25 corroborates the broad direction — but a fair reading applies the same scrutiny to an incoming government's account of what it inherited as it applies to the government that spent the money in the first place.

Debt taken on to build the future can just as easily crowd it out, if the projects it funds do not stay on budget.