The Commonwealth has handed down roughly a hundred budgets since the modern system settled in the 1920s. In six of them, the government of the day deliberately strengthened the nation's finances at a cost to itself. Not six terms. Not six governments. Six budgets, in three short bursts, each burst over within two or three years.
That is the real record of fiscal discipline in Australia. Everything else is either a windfall that was banked, an asset that was sold, or a surplus that fell out of a growing economy without anyone having to say no. The myths around particular treasurers exist because the genuine article is so rare that any approximation gets inflated into legend.
The test
A responsible budget here means one that passes four tests. It measurably improves the structural fiscal position. It does so through deliberate decisions, not cyclical revenue. The amount is material, and the government pays a political price for it.
Applying the test means a screen, not a vibe. From 1970-71, Treasury's historical series allows every year to be checked: payments and receipts as a share of GDP, and the underlying cash balance, which excludes asset sales. Any year in which payments fell by a point of GDP or more, or the balance improved by a point or more, was examined to see whether the change came from spending decisions or from revenue. Before 1970 the series is not comparable, so the Depression, war and post-war budgets were assessed on the historical record of what each one did and what it cost.
A headline surplus is not enough, because revenue can grow faster than spending without anyone deciding anything; that does not make such surpluses worthless, it just means they are not evidence of discipline. Asset sales are not enough, because they move debt without moving net worth, and political pain is not enough on its own, because an unpopular budget can also be a bad one.
The honour roll
1951-52 — Fadden (Menzies). 10% income tax levy, higher company and sales tax, to choke wool-boom inflation of 20%. Large surplus budgeted. Passed; reversed August 1952.
1986-87 — Keating (Hawke). Real cuts to outlays after the "banana republic" warning; deficit halved. Payments 27.4 → 27.0% of GDP. Receipts 25.4 → 26.2%. Balance -$2.4bn. Passed.
1987-88 — Keating (Hawke). Second round through the Expenditure Review Committee; first surplus in a generation. Payments 27.0 → 25.3%. Receipts 26.2 → 25.8%. Balance +$1.5bn. Passed.
1988-89 — Keating (Hawke). Third round, May Statement and budget; largest one-year fall in payments in the series. Payments 25.3 → 23.2%. Receipts 25.8 → 24.7%. Balance +$5.4bn. Passed.
1996-97 — Costello (Howard). Cuts to universities, the public service, labour market programs and the states; deficit halved. Payments 25.6 → 25.1%. Receipts 23.5 → 24.0%. Balance -$6.1bn. Passed.
1997-98 — Costello (Howard). Second round; budget balanced with receipts flat. Payments 25.1 → 23.9%. Receipts 24.0 → 23.9%. Balance +$0.1bn. Passed.
2014-15 — Hockey (Abbott). GP co-payment, fee deregulation, pension indexation, dole waiting period, deficit levy. Most measures blocked. Attempted.
The receipts column is the evidence that these were decisions. In 1987-88 and 1988-89 revenue fell as a share of GDP while the surplus grew; the improvement was all on the spending side. In 1997-98 revenue was flat and payments fell 1.2 points. Only 1986-87 and 1996-97 had help from revenue, and in both the deficit was halved while payments were cut in real terms, which is why they stay on the list.
The century outside the table is not blank, just different. The 1920s were borrowed prosperity, with the Bruce-Page government funding development on London loans. The 1931 Premiers' Plan, a 20% cut in adjustable spending agreed with the states under pressure from the Bank of England's Niemeyer mission, was the harshest consolidation in the country's history and destroyed the Scullin government; it is excluded here because the alternative was default, so there was no choice to make. The war budgets raised taxes to levels never seen since, for a war. Chifley's post-war surpluses of 1946 to 1949 were real, but earned on a wartime tax base and a rationed spending base he inherited rather than chose.
The screen also surfaces near-misses after 1970. Howard's own 1978-79 budget as Treasurer cut payments by a point and imposed a tax surcharge, but receipts fell with the economy, the deficit did not move, and the surcharge was abandoned in 1980. Costello's 1998-99 and 1999-00 budgets continued the fall in payments, but by then the surpluses owed more to a strong economy and one-off revenue timing than to new cuts. The 1.5-point fall in payments in 2010-11 was the GFC stimulus switching itself off. The Chalmers surpluses of 2022-23 and 2023-24, the first in fifteen years, were built on iron ore prices and bracket creep; banking most of the upgrade was a decision, the surplus itself was not.
So: no qualifying budget between 1952 and 1986, thirty-four years. None between 1998 and 2014, spanning the longest boom in the country's history. None since. Each of the three runs followed a crisis and ended the moment the pressure lifted. Fadden was reversed when wool collapsed and recession bit. Keating's run ended with the 1990 election, the interest-rate squeeze and the recession. Costello's ended with the GST, which had to be bought through the Senate with compensation and tax cuts, and then the boom.
The Costello case
The Howard years are the test case for the myth, because they are remembered as eleven years of discipline and contained two. Net debt went from $95.8 billion in June 1996 to minus $44.8 billion by June 2008. The budget papers show where that came from: about $23 billion of surpluses before the mining boom, about $74 billion of surpluses during it, and about $72 billion of asset sales, Telstra above all. Roughly two-fifths family silver, two-fifths boom, one-sixth hard work. Net financial worth, which counts public-service superannuation, was still minus $14.7 billion in 2007-08 with net debt gone.
The hard work was real. The 1996-97 and 1997-98 budgets cut payments from 25.6% to 23.9% of GDP and took an $11.1 billion deficit to balance with no boom to help. After that, payments kept drifting down as a share of GDP only because nominal GDP was growing at close to 8% a year. The Parliamentary Budget Office's structural estimates, which strip out the terms-of-trade windfall and the cycle, show the budget moving from a structural surplus of 1¾ to 3¼% of GDP in 2002-03 to roughly zero by 2007-08. Treasury's 2008 analysis found that of $334 billion in revenue upgrades between the 2004-05 budget and the 2007 election, $314 billion went out again in tax cuts and new spending. Banking the Future Fund's $69 billion took restraint, since every dollar in it was a dollar some minister wanted to spend, but keeping part of a windfall is a different thing from cutting, and the two should not be sold under one label.
Why so few
The easy answer is that voters punish repair and reward sweeteners. True, but it lets everyone else off.
Governments bring it on themselves by promising what the cycle cannot deliver: Fraser's 1977 "fistful of dollars", reversed after the election; Keating's 1993 L-A-W tax cuts; the 2007 bidding war in which both sides matched each other's cuts; the 2019 and 2022 matching of stage-three tax cuts. Oppositions make it worse by opposing repair they would attempt in office, as Labor did with the 2014 measures and the Coalition did with Swan's 2012 savings. The media treat every budget as a contest of winners and losers on the night, which rewards handouts and punishes anything that has to be explained. The Senate, since 1984, has made it nearly impossible to pass measures that take money off anyone, so the only repair that survives is the kind nobody notices.
The result is a ratchet. Windfalls become permanent tax cuts and permanent payments; downturns add stimulus that is never fully withdrawn; and the structural position erodes between crises. The 2014 budget showed the trap from the other side. Abbott and Hockey tried genuine structural repair, the first attempt in sixteen years, and it was annihilated: the co-payment dropped, fee deregulation blocked, the pension and dole changes defeated, and Abbott gone within eighteen months. The attempt deserves credit. The design, which loaded the pain onto the young and the poor while keeping a generous parental leave scheme, deserved to fail. Every Treasurer since has drawn the first lesson and ignored the second.
What it would take
A country that manages six responsible budgets in a hundred years does not have a fiscal management tradition. It has, three times, been frightened into behaving, and the behaviour lasted as long as the fear.
The structural deficits of the last fifteen years are not a fall from the Australian norm; they are the norm. The Howard-era surpluses were the departure, and most of them were the economy's doing. When a Treasurer says the books are hard to fix, the honest reply is that nobody has fixed them by choice since 1997, and nobody before that since 1988.
The central question stands: will a government deliberately strengthen the structural position when doing so could cost it votes? The historical answer is no, unless the markets or inflation force the issue. Sustainable fiscal management needs more than a crisis response. It needs repair done in good years, when there is room to do it fairly; it needs oppositions willing to pass savings they would make themselves; and it needs a public that has stopped mistaking a windfall for a Treasurer. Until the answer is yes, the next pay-down will come, as the last one did, from China, iron ore and the family silver, and we will keep calling it good management.
