Cheryl runs a two-person bookkeeping practice in Wagga Wagga, and every year the federal budget lands, she does the same thing before she reads a single headline. She opens Budget Paper No. 1 and finds two numbers: receipts as a share of the economy, and payments as a share of the economy. She has tracked this pair since well before the pandemic, because the gap between them tells her, in about a page, whether the government her small-business clients depend on for grants, rebates and tax settings is pulling in its belt or letting it out — and that, more than any minister's press conference, is what tells her whether the year ahead is going to feel tight or comfortable for the people who ring her office.
Since the 2022 election, Cheryl has read four of these documents line by line: October 2022, May 2023, May 2024 and March 2025, each one followed a few months later by a Mid-Year Economic and Fiscal Outlook update that checks the numbers against where the economy has actually gone. What she has watched happen to those two lines — receipts and payments, both measured against the size of the whole economy — is the plainest, least disputed record of what four years of budgets under this government actually did. It does not depend on which side of politics a reader favours. It is arithmetic anyone can check by opening the same document she does.
What the government did: four budgets, one shape
The Albanese government has handed down four full federal budgets since taking office in May 2022: the October 2022 Budget, delivered five months after the election under caretaker-transition rules; the May 2023 Budget, the government's first on the normal annual cycle; the May 2024 Budget; and the March 2025 Budget, brought forward from its usual May slot ahead of that year's election (Budget Paper No. 1, various editions, 2022–2025). Layered between them are Mid-Year Economic and Fiscal Outlook (MYEFO) updates each December, and Final Budget Outcome statements each September that report what actually happened against what was forecast. Read together, these documents are the government's own, audited account of two things: how much it took in, and how much it paid out, each expressed as a share of gross domestic product (GDP) — the standard way budget documents make dollar figures from different years comparable, since a dollar buys less each year as prices and the economy grow.
On the payments side, four years produced a long list of new and expanded programs, alongside the winding-down of emergency pandemic support. Cost-of-living relief — energy bill rebates, cheaper medicines under the Pharmaceutical Benefits Scheme, a tripled bulk-billing incentive for GPs, an increase to Commonwealth Rent Assistance — was layered on top of already-growing structural programs: the National Disability Insurance Scheme (NDIS), aged care (including a historic wage increase for aged-care workers following a Fair Work Commission decision), Medicare, and defence spending lifted toward the AUKUS submarine program's long build-out. The government also stood up several large investment vehicles — the Housing Australia Future Fund, Rewiring the Nation, the National Reconstruction Fund and Future Made in Australia programs — some of which sit off the main budget ledger in ways examined in a separate audit on this site. On the receipts side, the story was dominated less by new taxes than by a commodity-and-employment windfall in the first two budgets, examined in detail in a companion piece on the structural deficit, and by bracket creep — inflation pushing wage earners into higher tax brackets without those brackets being adjusted — doing more of the quiet work in later years as the windfall faded.
Two numbers anchor the whole four-year record. Receipts and payments, each as a share of GDP, moved from a wide pandemic-era gap in 2020-21 through a rare period where receipts sat above payments in 2022-23 and 2023-24 — the first Commonwealth surpluses since 2007-08 — before payments moved back above receipts from 2024-25. That is the shape at the centre of this piece, and it is worth being precise about what the shape does, and does not, show before drawing conclusions from it.
The strongest case for the policy
Put the government's fiscal record at its best
The government's defence of its four budgets rests on a simple, defensible claim: when an unexpected wave of extra tax revenue arrived in 2022 and 2023, it banked most of it rather than spending it, and the result was the first Commonwealth surpluses in fifteen years — an outcome no government had delivered since before the Global Financial Crisis. Treasurer Jim Chalmers said in the October 2022 Budget Speech that the government had banked almost all of the improvement in receipts recorded since the pre-election fiscal outlook, a figure reported at close to 99 per cent (approx.) of that upgrade (Budget Speech, 25 October 2022; Budget Paper No. 1, Statement 3, October 2022-23). That is not the easy path. A government sitting on an unexpected revenue upgrade the size Australia saw in 2022 has every political incentive to spend it — on new programs, on tax relief beyond anything already promised, on visible projects that build goodwill before the next election. Choosing largely not to do that, in a treasurer's first budget, is a real and checkable act of restraint.
At the same time, the government did not hold the line by cutting services. Across the same period it funded a historic wage increase for aged-care workers, tripled the bulk-billing incentive for GPs — the largest increase in the incentive's history — expanded cheaper medicines, lifted Commonwealth Rent Assistance, and delivered energy bill rebates during the worst of the cost-of-living squeeze, all while unemployment held near a fifty-year low (Australian Bureau of Statistics, Labour Force, 2022–2024). On the government's own account, that is the point: fiscal repair and essential services were not treated as a trade-off. Net debt came in lower than the pre-election forecasts had projected, meaning a smaller ongoing interest bill than the alternative path in which the windfall had been spent in full. Whatever one concludes about what happened after 2023-24, two consecutive surpluses delivered without a program of service cuts is, on the historical record, an unusual combination — and the primary documents show the government made a genuine choice to bank the upgrade rather than spend it, which is more than can be said for every government that has faced a similar windfall.
The evidence: reading receipts and payments across four budgets
The two lines Canberra has tracked for decades
Every Budget Paper No. 1 carries a chart the Treasury has published in some form for decades: receipts and payments, both as a share of GDP, running back over the long run and forward across the four-year forward estimates. The reason budget documents use this measure, rather than raw dollar figures, is straightforward — a billion dollars in 1990 is not the same as a billion dollars today, and a billion dollars in a $1.5 trillion economy is not the same as a billion dollars in a $2.6 trillion one. Measuring both receipts and payments against the size of the whole economy strips out that distortion and answers a cleaner question: is government, relative to the country it serves, getting bigger or smaller?
In the years just before COVID-19, Commonwealth payments ran at somewhere in the order of 24 to 25 per cent of GDP (approx.), with receipts running close behind, near enough to balance to produce budgets that were close to balanced or in modest deficit through most of the 2010s (Budget Paper No. 1, historical fiscal data, various editions, approx.). Treasury's own long-run reference point — commonly described in budget commentary as the 30-year average — has put payments at somewhere around 24.8 per cent of GDP (approx.) over that longer run (Budget Paper No. 1, Statement 3 and Statement 6, various editions, 2023–2025, approx.). That average is the benchmark this piece measures the post-COVID years against, because it is the benchmark the government's own budget papers use.
What happened to the two lines: budget by budget
The COVID-19 emergency drove payments to a level with no peacetime precedent. In 2020-21, at the height of JobKeeper and other emergency income support, Commonwealth payments rose to somewhere in the order of 31 to 32 per cent of GDP (approx.) — a level not approached outside wartime — while receipts, hit by a sharp recession, fell rather than rose (Budget Paper No. 1 and Final Budget Outcome 2020-21, approx.). That gap produced the largest deficit in Commonwealth history, an underlying cash deficit reported at around $134 billion (approx.) that year.
By the time the Albanese government handed down its first budget in October 2022, payments had already fallen a long way from that peak as emergency programs expired, while receipts were rising fast on the back of high commodity prices and a red-hot labour market. The October 2022 Budget forecast payments easing toward somewhere near 25 to 26 per cent of GDP (approx.) for 2022-23, with receipts, boosted by the windfall, running close to or slightly above that level (Budget Paper No. 1, Statement 3, October 2022-23, approx.). The 2022-23 Final Budget Outcome, published in September 2023, confirmed the shape: an underlying cash surplus of $22.1 billion (approx.), around 0.9 per cent of GDP (approx.), meaning receipts had moved measurably above payments for the first time since 2007-08.
The May 2023 Budget carried the same story into a second year, with the 2023-24 Final Budget Outcome, published in September 2024, recording a second consecutive surplus of $15.8 billion (approx.), around 0.6 per cent of GDP (approx.) — receipts still running slightly ahead of payments, though by a narrower margin than the year before, as more of the ongoing revenue upgrade was directed to new spending rather than banked (Budget Paper No. 1, Statement 3, May 2023-24; Final Budget Outcome 2023-24, September 2024, approx.).
The May 2024 Budget marked the turn. It forecast a return to deficit of around $28.3 billion (approx.), about 1.0 per cent of GDP (approx.), for 2024-25 — payments moving back above receipts, driven by new spending commitments, including cost-of-living relief, the aged-care wage decision and further defence funding, landing at the same time as the commodity windfall began to fade (Budget Paper No. 1, Statement 3, May 2024-25, approx.). Crucially, the same budget's payments-to-GDP figure did not fall back to the pre-pandemic run rate; it was forecast to settle at a level budget commentary at the time described as running above the 30-year average across the forward estimates — commonly reported as being in the order of 26 to 27 per cent of GDP (approx.) across the four years to 2027-28, compared with the roughly 24.8 per cent long-run average the government's own papers cite (Budget Paper No. 1, Statement 3 and Statement 6, May 2024-25, approx.).
The March 2025 Budget, brought forward from its usual May slot ahead of that year's election, confirmed the same trajectory. Deficits were projected across the full four-year forward estimates, and the payments-to-GDP line in the medium-term projections — Treasury's decade-long extension of the forward estimates, built on the same broad method the Intergenerational Report uses — did not return to the pre-pandemic average within the published projection window (Budget Paper No. 1, Statement 3 and medium-term projections, March 2025-26, approx.). The December 2023 and December 2024 MYEFO updates, which check the budget's numbers against how the economy has actually moved since the previous budget, confirmed the same shape at each mid-year mark: payments running persistently above the long-run average, even as receipts eased back from their windfall-era peak.
It is worth measuring the whole trajectory against a single fixed point: the Pre-Election Fiscal Outlook (PEFO), the budget update Treasury and the Department of Finance are required by law to publish during every election campaign, independent of the government of the day, so voters have an honest baseline to judge promises against. The April 2022 PEFO, published in the caretaker period before the election that brought this government to office, was itself already shaped by the tail of the pandemic response, but it gives Cheryl — and any reader — a fixed marker that predates every one of the four budgets this piece examines. Reading the October 2022, May 2023, May 2024 and March 2025 budgets against that single starting point, rather than against each other, is what makes the shape of the trajectory hardest to argue with: receipts moved up first and fastest, on the back of conditions nobody in Canberra had engineered, payments came down more slowly and by less, and the gap between the two, which briefly favoured the government in 2022-23 and 2023-24, has since reopened in the other direction while sitting on a higher payments base than the PEFO itself assumed.
Where the extra money goes: payments by function
The payments-to-GDP line is a single number, but Budget Paper No. 1's expenses statement breaks it down by function — health, social security and welfare, defence, education, general public services and so on — and that breakdown is where the "permanently higher" argument gets its clearest support. Reading the functional tables across the four budgets, five categories account for a disproportionate share of the growth above the pre-pandemic base: the NDIS, sitting within social security and welfare; aged care, sitting within health; health spending more broadly, including the Medicare package; defence, building toward the AUKUS submarine program's long-run costs; and public debt interest, which is not a program at all but the mechanical cost of servicing money already borrowed (Budget Paper No. 1, Statement 6, Expenses by function, various editions, 2022–2025, approx.). Each of these five is examined at greater length in companion audits on this site. What matters for this piece is simpler: these are the categories Treasury's own papers project growing faster than the economy as a whole across the forward estimates, and they are also the categories least able to shrink quickly if a future government wanted to bring payments back down toward the 30-year average, because each involves either a legislated entitlement, a signed multi-decade contract, or a wage decision already delivered. A payments line built substantially from commitments of that kind behaves differently to one built from discretionary programs that can be wound back in a single budget.
Real spending growth: the other half of the story
A share-of-GDP measure can move for two different reasons: because payments themselves change, or because GDP — the denominator — changes. A booming economy can make a growing payments bill look smaller as a share of GDP even if the government is spending more in real terms; a weak economy can make the same dollar figure look larger. To separate the two, Budget Paper No. 1's expenses statement also reports real growth in payments — spending growth with inflation stripped out — year by year across the forward estimates.
Read across the four budgets, real payments growth ran at rates above Australia's trend real GDP growth rate in several of the years covered, particularly where new permanent commitments — the NDIS, aged care, defence and the interest bill on accumulated debt — were phasing in at the same time as legacy pandemic programs were phasing out (Budget Paper No. 1, Statement 6, Expenses, various editions, 2022–2025, approx.). Trend real GDP growth across this period ran at an annual rate in the order of 2 to 2.25 per cent (approx.; Budget Paper No. 1, Statement 2, Economic Outlook, various editions); in some individual years, real payments growth ran meaningfully faster than that, before moderating later in each budget's forward estimates as new measures matured (Budget Paper No. 1, Statement 6, various editions, approx.). The interest bill on Commonwealth debt was, on the government's own figures, one of the fastest-growing individual payment lines across the period — a direct, mechanical consequence of the debt built up through the pandemic, growing regardless of what else happens in the economy (Budget Paper No. 1, Statement 6, Interest payments, various editions, approx.).
This is the distinction the receipts-and-payments chart alone cannot make visible on its own: a strong economy in the early post-pandemic recovery flattered the ratio in 2022-23 and 2023-24, holding payments-to-GDP down even as real payments themselves kept growing. As growth normalised in later years, the underlying growth in payments became more visible in the GDP-share figure too — one reason the payments line does not fall back to its pre-pandemic level even once the pandemic-era programs it originally measured have long since ended.
Four budgets tell a two-part story: two years where a windfall made spending look restrained, and two years where the same spending, without the windfall, did not.
The critique: a bigger government, and not entirely by design
The clearest evidence in the government's own documents is what happened once the commodity-and-jobs windfall faded. Deficits returned in 2024-25 and were projected to continue across the full forward estimates published in March 2025, but the payments line did not return with them to where it had sat before the pandemic. Budget Paper No. 1's own figures, read across the May 2024 and March 2025 editions, show payments settling at a level budget commentary at the time put in the order of 26 to 27 per cent of GDP (approx.) across the forward estimates — around two percentage points of GDP (approx.) above the 30-year average the same documents cite, a gap in the order of $50 to $60 billion a year in today's economy (approx.; Budget Paper No. 1, Statement 3 and Statement 6, May 2024-25 and March 2025-26, approx.). That is the empirical basis for describing this as a structurally larger government, not simply a temporarily larger one: the elevated payments share persists well beyond the pandemic programs that first produced it, and well beyond the point where the commodity windfall that funded two years of surplus had faded.
Several of the specific pressures behind that persistence are structural by design, not temporary by accident. The NDIS, legislated in 2013 under a different government, has grown into one of the fastest-growing lines in the entire budget, with its own built-in growth dynamics that do not shrink when the economic cycle turns — examined in full in a companion audit on this site. The aged-care wage decision, health spending and Medicare, and defence spending building toward the AUKUS program's long-run costs, all share the same feature: they are decisions with multi-year, and in some cases multi-decade, funding tails, made in specific budgets but binding on every budget after them. None of that makes the underlying need illegitimate — an ageing population, a rebuilding defence posture and a scheme that has genuinely transformed disability support are real pressures, not accounting artefacts. But it does mean the payments side of the ledger has a ratchet quality other lines of government spending do not: once a program of this kind is legislated and its cohort grows, walking the spending back down is far harder, politically and practically, than the original decision to fund it.
The receipts side of the ledger tells a less flattering version of the same story. The two surplus years leaned heavily on an extraordinary, largely unrepeated run in iron ore and coal prices and a labour market tighter than at almost any point in fifty years — both examined at length in a companion piece on the structural deficit — rather than on a change to the tax system that would keep generating revenue at the same rate once conditions normalised. As that windfall faded, an increasing share of the receipts line's ongoing growth came instead from bracket creep: inflation pushing wage earners into higher marginal tax brackets that Australia's income tax system, unlike some peer countries, does not automatically adjust for inflation. The Parliamentary Budget Office's medium-term projections have highlighted the personal income tax share of total receipts rising over time in the absence of further legislated relief, a dynamic examined in full in a separate audit on this site (Parliamentary Budget Office, medium-term fiscal projections, 2023–2025, approx.). In plain terms: a growing share of the revenue funding an already larger payments base is being supplied not by a deliberate, transparent tax decision but by the quiet erosion of workers' take-home pay relative to the tax-free thresholds and bracket boundaries legislated years earlier.
Put the two sides together and the four-budget record reads less like a single coherent fiscal strategy and more like two different forces working in the same direction without either being fully chosen. On the payments side, a series of individually defensible decisions — a wage rise for some of the lowest-paid essential workers in the economy, support for people with disability, more money for hospitals and defence — added up to a payments base that does not return to its starting point. On the receipts side, the mechanism doing an increasing share of the work to keep the budget in the same broad vicinity of balance is one almost nobody voted for directly: the passive effect of inflation on a tax scale Parliament has not kept pace with. Neither side of that ledger was dishonestly presented — the budget papers disclose both dynamics in their own tables, for any reader prepared to look. But neither was it the subject of the kind of explicit, up-front public debate that a genuine, deliberate decision to permanently enlarge government, funded by a permanently higher personal tax share, would deserve.
A permanently higher payments base also changes how exposed the budget is to a variable this government does not control: the interest rate the Commonwealth pays to borrow. Gross debt built up through the pandemic has to be refinanced in slices as older bonds mature, at whatever rate the market demands at the time, and the Reserve Bank's tightening cycle through 2022 and 2023 — the cash rate rising from 0.10 per cent to above 4 per cent (Reserve Bank of Australia, cash rate target series, 2022–2023) — lifted the cost of that refinancing well above the rates locked in during the record-low-rate years of the pandemic itself. Budget Paper No. 1's own figures show interest payments as one of the fastest-growing individual lines in the expenses statement across this period, and unlike almost every other payment category, the interest bill has no discretionary component a future budget can simply choose not to fund (Budget Paper No. 1, Statement 6, Interest payments, various editions, approx.). A larger permanent payments base, sitting on a larger permanent stock of debt, is more sensitive to the next rate cycle than a smaller one would be — a risk that compounds quietly in the background of every one of the four budgets this piece examines, independent of anything else in the payments or receipts tables.
The payments line came down from its pandemic peak. It did not come all the way back — and the receipts line meant to match it leans increasingly on bracket creep, not on any tax change anyone had to defend in public.
What the government got right
Three things are worth crediting plainly. First, the scale of restraint in the first budget was real: banking close to all of an unexpected, multi-billion-dollar revenue upgrade rather than matching it with new permanent spending is not the easy or the popular choice, and the government made it. Second, the two surpluses were not manufactured by cutting services in a downturn — they landed alongside unemployment near a fifty-year low and alongside genuine, means-tested cost-of-living relief, which shows that short-term fiscal repair and support for households under pressure were not treated as mutually exclusive in this period. Third, net debt came in lower across this period than the pre-election forecasts had projected, a real and durable saving in the interest the Commonwealth has had to pay ever since, regardless of what happened to the deficit later.
It is also fair to note that Australia was not unusual in running a larger government after the pandemic than before it — nearly every advanced economy's payments-to-GDP settled above its own pre-pandemic average through the mid-2020s, reflecting a genuinely global shift in what citizens expect governments to fund after a once-in-a-century health and economic shock (OECD and IMF fiscal comparisons, various, approx.). And several of the specific pressures pushing payments higher — an ageing population's health and aged-care costs, a disability scheme that has changed hundreds of thousands of lives, a deteriorating strategic environment driving defence spending — predate this government and would confront any government holding office through the same years, of either political colour. The Intergenerational Report the Treasury published in 2023 makes that point in its own right: the pressures pushing payments upward were already built into Australia's demographic and program settings well before the 2022 election, and no government elected in that period could have avoided confronting them (Australian Government, Intergenerational Report, Treasury, 2023).
What the government cannot fairly claim credit for is having made the resulting size of government the subject of an explicit choice, put to the country and defended on its own terms. Budget documents disclose the payments-to-GDP trajectory in full, and any reader who does what Cheryl does each year — turning to the same two lines in Budget Paper No. 1 — can see it. But disclosure in a budget table is not the same as a government arguing, in a budget speech or an election campaign, that Australia should permanently run a larger public sector funded by a rising personal income tax share. That argument has not, on the public record, been made directly. The trajectory has instead emerged from a series of individually reasonable decisions and a passive tax mechanism, adding up to an outcome nobody explicitly proposed and nobody has yet had to defend as a stated destination.
Credit for transparency itself is also due, and it should not be understated. Every figure in this piece — the payments and receipts shares, the 30-year average, the functional breakdown, the medium-term projections — comes from documents the government itself tabled in Parliament and published in full, not from a leak, a freedom-of-information request, or an outside estimate built to embarrass it. Budget Paper No. 1's parameter-and-policy reconciliation tables, its expenses-by-function breakdown, and its medium-term projections are more detailed than the equivalent disclosures many comparable governments provide, and the independent Parliamentary Budget Office exists, and is funded, precisely to check the government's own numbers against reality. A government that wanted to hide a permanently higher payments base had the tools to make that harder to see than it is. This one did not use them. The honest reading of the four-budget record is that the country has been told, in its own primary documents, exactly what has happened to the size of government — the open question is not whether the information was available, but whether anyone in Canberra has yet been asked to answer for what it shows.
The risk to the future
A government that has quietly gotten bigger, and stayed that way
The risk this topic anchors is not that spending rose during a pandemic — every advanced economy's did, and Treasury's own risk statements always flagged that the pandemic-era peak would come down. The risk is what the four-budget record shows happening after that peak passed: payments settling at a share of the economy that budget documents themselves describe as running above the long-run average, without a matching, explicit decision about how that permanently larger government gets paid for once the one-off commodity windfall that funded two good years is gone. On the current trajectory shown in the government's own medium-term projections, an increasing share of the answer is bracket creep — a rising personal income tax burden borne disproportionately by wage earners, arrived at without a specific vote on it — rather than a deliberate choice between higher visible taxes, sustained spending restraint, or a larger, permanent stock of public debt.
That matters for the next generation in two concrete ways. First, a payments base that structurally outgrows the economy leaves less fiscal room for the next genuine emergency — the next pandemic, the next global shock, the next war-driven energy crisis — because a government already running near or above its own long-run spending average has spent the flexibility a lower baseline would have preserved. Second, if bracket creep keeps doing the quiet work of paying for that base, today's and tomorrow's wage earners carry a rising share of the cost of decisions — about disability support, aged care, defence and debt already accumulated — made without their tax burden being the explicit subject of the debate that produced those decisions. Neither outcome is a prediction of crisis. Both are a description of a country with less room to move than it had in 2019, carrying a bigger, more expensive government that nobody in Canberra has yet stood up and explicitly asked the country to choose.