Australia's Company-Tax Bill More Than Doubled in a Decade. The Big Four Banks Paid Almost Exactly the Same.

Each December the Australian Taxation Office publishes a spreadsheet that makes an uncomfortable list public: the name of every large company operating in Australia, its total income, and the company tax it paid. The annual headline it generates is familiar — a third of big companies paid no tax — and is usually answered, correctly, with the reminder that revenue is not profit and losses are real.

Read the eleven years of that spreadsheet together and a different pattern appears, one that has nothing to do with the companies paying nothing. It is about the companies that used to pay the most. A decade ago Australia’s four major banks were the anchor of the large-company tax base. They are not any more — and it is not because the base shrank around them. It grew, fast, while their contribution stood still.

In 2013-14 the big four banks — Commonwealth Bank, Westpac, NAB and ANZ — paid a combined $9.53 billion in company tax. In 2023-24 they paid $9.99 billion. Over the same eleven years the total company tax recorded across the whole transparency population went from $39.9 billion to $96.1 billion, more than doubling. The banks added about five per cent; the base as a whole added a hundred and forty.

The banks’ bill stopped growing — then shrank#

The flat line is not a rounding illusion. Consumer prices rose about 31% between mid-2014 and mid-2024. Holding the banks’ 2013-14 payment to that same inflation would have put their 2023-24 bill near $12.5 billion. They paid $10.0 billion. In real, inflation-adjusted terms, the four banks’ company tax is roughly a fifth lower than it was a decade ago.

And the decade was not a smooth plateau. Bank company tax climbed to a peak of $11.8 billion in 2017-18, then fell for four straight years to a trough of $8.4 billion in 2021-22 — the stretch covering the financial services Royal Commission, a long run of record-low interest rates, and the pandemic — before recovering to roughly where it had started. Across the four, the 2023-24 split was Commonwealth Bank $3.43 billion, Westpac $2.64 billion, NAB $2.24 billion and ANZ $1.67 billion.

Because the total climbed while the banks’ bill did not, the banks’ share of the base collapsed. In 2013-14 the four paid 23.9% of all the company tax in the transparency data — close to a quarter of the whole from four companies. By 2023-24 they paid 10.4%.

What replaced them swings with the commodity price#

The tax the banks did not add came, overwhelmingly, from the ground. Screening the same data for the major mining and oil-and-gas companies — a name-based approximation, described in the method note — traces the mirror image of the bank line: from about a quarter of the base in 2013-14 up to roughly two-fifths by the early 2020s.

The two lines cross, then trade places — a steady payer swapped for a volatile one. The bank line stays inside a narrow band and drifts down. The resources line lurches: 11% at the 2015-16 price trough, above 40% through the iron ore and gas boom of the early 2020s. The base did not only change hands. It became far more sensitive to a price set in Singapore and Rotterdam than to anything decided in an Australian branch.

The top of the 2023-24 table shows the new order plainly. Of the ten largest company taxpayers, seven were miners or gas producers, led by Rio Tinto ($6.25 billion) and BHP ($6.01 billion). The first bank, Commonwealth Bank, sits fifth.

Why this matters beyond the banks#

Company tax is one of the Commonwealth’s three biggest revenue lines, and the entities in this data pay about two-thirds of all of it, on the ATO’s own reckoning. When the most reliable slab of that base — banks, whose earnings move slowly — stops growing, and the fastest-growing slab is the one that doubles and halves with commodity prices, the whole line inherits that volatility.

That is visible in the budget. Treasury’s receipt forecasts publish explicit assumptions for the iron ore, coal and LNG price, and a sensitivity analysis for what a different price would do to the bottom line, precisely because so much now rides on it. The run of upgraded company-tax receipts that produced the 2022-23 and 2023-24 surpluses was driven in large part by resource-sector profits on high prices — the same prices that, on the way down, take the receipts with them. A tax base leaning on banks is dull and dependable. A tax base leaning on miners is neither.

What could explain this instead#

The banks’ profits genuinely stalled — this is not avoidance. The flat tax line most likely reflects flat pre-tax profit, not any erosion of what the banks owe. The 2018-19 Royal Commission brought large, deductible customer-remediation provisions; years of record-low interest rates compressed net interest margins; and 2020 brought pandemic loan-loss provisions. Bank earnings were pressured for much of the decade, and pressured earnings pay less tax. Nothing here says the banks paid less than the law required.

Some bank profit — and its tax — simply moved to other companies. Across 2016-2021 the big four sold off their wealth-management, insurance and advice arms. Those businesses kept earning and kept paying tax; they just did it under new owners rather than inside the banking parent. Part of the “missing” bank tax did not vanish from the economy — it changed the name on the return, and may still sit elsewhere in the same dataset.

This is entity data, not group data. The transparency report lists single legal entities and warns against reading them as economic groups. Banks restructure — ANZ moved to a holding-company structure in 2022-23, which is why its name changes in the files — and intra-group changes can move where tax is booked without changing the group’s total. The four-entity series is a consistent measure of those four entities, not a full audit of the banking groups.

The miners’ rise is partly a moment in the cycle. The resource share hit 11% at the 2015-16 price trough and 42% at the 2022-23 peak. Some of the “handover” is simply where the commodity cycle sat at the end of the window. If prices fall, the resource share will fall with them — which is the point about volatility, not a rebuttal of it, but it does mean the current split should not be read as a permanent settlement.

The annual story about this dataset is the companies that pay nothing. The longer story is quieter and harder to fix: the part of the corporate tax base that a government could once forecast in its sleep has stopped growing, and the part that took its place is the part nobody can forecast at all.