Australian education has achieved something rather impressive.

After two decades of curriculum reform, pedagogical innovation, wellbeing initiatives, inclusion strategies, digital transformation, and enough educational acronyms to incapacitate a small bureaucracy, Australian 15-year-olds are performing substantially worse in reading and mathematics than they did at the beginning of the century.

The OECD’s Programme for International Student Assessment, better known as PISA, makes uncomfortable reading. Australia’s average reading score fell from 528 in 2000 to 498 in 2022. Mathematics fell from 524 in 2003 to 487. Science, first measured as a major PISA domain in 2006, fell from 527 to 507.

Australia remains above the OECD average in all three domains, which is certainly preferable to the alternative. But international league tables can distract from the more uncomfortable comparison: Australia against itself.

The OECD calculated that Australian 15-year-olds in mathematics and reading in 2022 performed more than 25 PISA points below their counterparts in the early 2000s. In practical terms, the OECD says this is roughly equivalent to today’s 15-year-olds performing at the level expected of 14-year-olds 20 years earlier.

None of this should necessarily be confused with failure, of course. Failure is a rather unfashionable concept in modern education.

Perhaps the results simply require a differentiated definition of success.

The uncomfortable question is how one of the wealthiest countries in the world, after decades of educational reform, managed to go backwards.

The easiest answer is to blame teachers. It is also the laziest.

Teachers did not design the national curriculum. They did not create departmental behaviour frameworks, reporting requirements or the extraordinary administrative machinery surrounding modern schooling. They did not decide that almost every social problem discovered by government required a corresponding program to be delivered in schools.

Teachers inherited the lot, generally accompanied by an email explaining that implementation begins Monday.

Blaming teachers has the additional inconvenience of requiring us to ignore the fact that Australia does not have enough of them. In PISA 2022, 61 per cent of Australian students attended schools whose principals reported that a lack of teaching staff hindered their capacity to provide instruction. In 2018, it was 17 per cent.

Apparently, the profession responded to being asked to do more with less by eventually developing the rather inconsiderate habit of running out of people.

This is not merely an historical argument. At the time of writing, Victorian public-school teachers are voting on the latest pay and conditions offer from the Carroll government. The ballot closes at 10am on Monday, August 17.

The negotiations have already outlasted the previous premier.

AEU members rejected the government’s previous offer in July, shortly before Jacinta Allan’s departure from the premiership. More than 50,000 members participated, with 51.18 per cent voting against it. Ben Carroll inherited the dispute along with the Premier’s office. His government is now hoping the revised offer will finally settle it.

The offer contains significant increases. The Victorian government says salaries will rise by at least 28.3 per cent over four years, including an increase of 12.75 per cent by October this year. Required meeting hours would also be halved from 80 to 40 annually.

The Carroll government has declared that this will make Victorian teachers the best paid in Australia.

There is only one small problem.

Queensland.

Queensland teachers are themselves in arbitration before the Queensland Industrial Relations Commission after negotiations failed to produce an agreement. Their previous agreement expired last year, and, for most Queensland teachers and school leaders, there has been no general salary increase since July 2024.

The Queensland Teachers’ Union is seeking an interim increase followed by further increases through the eventual determination. Its published modelling makes the interstate contest rather amusing. For the experienced teacher classifications it compares, the QTU calculates that Queensland would sit at $135,671 while Victoria reaches $133,456 in October 2026, if the QTU succeeds in its claim.

Apparently even the title of Australia’s best paid teacher now requires a footnote, a date, and an industrial lawyer.

The point is not that teachers should be engaged in an interstate salary competition. The point is what these disputes reveal. Australia has spent decades expanding what teachers are responsible for, while teachers are now bargaining with governments over remarkably fundamental requirements for doing the job: competitive salaries, manageable workloads, sufficient time and conditions capable of keeping people in the profession.

The modern teacher must, of course, teach. But teaching is merely the beginning.

They must differentiate. Document. Record. Report. Intervene. Adjust. Monitor wellbeing. Communicate with families. Collect data. Implement individual plans. Complete professional development. Manage behaviour. Record the management of behaviour. Attend meetings about the behaviour they have already managed and recorded.

Australia’s education departments have rarely encountered an objective they were unwilling to add to a teacher’s workload. Providing the time necessary to accomplish all those objectives has apparently proved more complicated.

Somewhere along the way, Australia also changed what it expected schools to be. Teaching children to read, write, calculate, reason and acquire knowledge was apparently insufficiently ambitious. Schools are now expected to concern themselves with wellbeing, resilience, inclusion, digital citizenship, social development, emotional regulation and an expanding collection of worthy social objectives.

Most of these things are perfectly defensible. The problem begins when they cease to support education and begin competing with it.

There has also been a subtler philosophical change. We have become increasingly uncomfortable with children being uncomfortable.

Failure can damage confidence. Consequences can damage relationships. Difficult work can disengage students. Firm discipline may not be sufficiently restorative. Deadlines require flexibility. Expectations require differentiation. Assessment requires adjustment.

Each proposition can sound entirely reasonable in isolation. Stack enough of them together, however, and something rather strange happens.

The standard becomes negotiable.

Compassion is indispensable in education. Any decent teacher understands that students arrive in classrooms with different abilities, circumstances and challenges. Some require considerable additional support. But supporting a student to reach a standard and lowering the standard until the student can reach it are not the same thing.

We appear increasingly reluctant to acknowledge the distinction.

This has produced one of the great contradictions of modern schooling. The student has been placed increasingly at the centre of education while the teacher has become increasingly responsible for everything the student does.

A student doesn’t complete an assessment. Was the task appropriately differentiated? Were sufficient scaffolds provided? Were adjustments considered? Was the student engaged? Were parents contacted? Were interventions documented? Were wellbeing concerns considered? Was sufficient feedback provided?

There remains, buried somewhere near the bottom of the flowchart, another possibility.

The student chose not to do it.

That is not an argument for abandoning children who struggle. Nor is it an argument for returning to some mythical golden age in which every child sat silently in perfect rows reciting multiplication tables. It is recognition that responsibility matters.

Children learn remarkably quickly from the systems adults construct around them. If deadlines perpetually move, deadlines become suggestions. If poor behaviour reliably produces discussion without meaningful consequence, students notice. If refusing work results primarily in additional work for the teacher, students notice that too.

And teachers certainly notice.

PISA provides an uncomfortable scoreboard here as well. In 2022, 33 per cent of Australian students reported that students did not listen to what their teacher said in most or all mathematics lessons. A quarter said they could not work well in most or all lessons. Forty per cent reported becoming distracted by their own digital devices, while 37 per cent were distracted by other students using theirs.

Perhaps we should consider the radical possibility that the conditions inside a classroom matter.

No pedagogical framework, however fashionable, can overcome a room in which teaching cannot reliably occur.

The greatest irony, however, may be what all this means for equity.

A child from a prosperous and highly educated family has considerable protection against a mediocre education. There may be books at home. There may be tutoring. There may be educated parents capable of identifying gaps and filling them. There is probably money available when the system falls short.

A disadvantaged child may have none of those things. For that student, school may be the only institution capable of providing the rigorous education necessary to substantially expand their opportunities.

This is where the progressive instinct to make education increasingly accommodating encounters an uncomfortable contradiction.

If equity comes to mean expecting less from disadvantaged children, tolerating classrooms where learning is routinely disrupted, or allowing students to progress without mastering fundamental knowledge, it is difficult to see who has been liberated.

Certainly not the child who reaches adulthood unable to read proficiently. He will receive remarkably little comfort from knowing that his Year Eight classroom was an affirming space.

None of this means wellbeing should disappear from schools. It does not mean students with disabilities should lose appropriate adjustments. It does not mean teachers should become indifferent to the circumstances of the children they teach. And it certainly does not mean every educational reform of the past twenty years has been misguided.

PISA can demonstrate that Australia’s academic performance has declined over the long term. It cannot tell us, by itself, why.

But after 20 years of deterioration, surely we are permitted to question some of the assumptions under which the system has been operating.

That includes asking whether the relentless expansion of what schools are expected to do has diluted what only schools can reliably provide. It includes asking whether compassion has sometimes been confused with reduced expectations. It includes asking whether accountability has shifted so overwhelmingly towards teachers and schools that students have been relieved of too much responsibility for their own choices.

And it includes asking why governments struggling to recruit and retain teachers continue finding new things for teachers to do.

Australia’s current industrial disputes should therefore concern people who have never attended a union meeting and have no intention of starting. They are another symptom of a system whose ambitions have become detached from its foundations.

Governments want better academic outcomes. They want greater inclusion. They want improved wellbeing. They want individualised education. They want extensive accountability. They want more data. They want teachers to address increasingly complex behavioural and social needs.

And naturally, they would prefer all of this to occur without excessive staffing, workload, or salary costs.

One begins to understand why negotiations have become difficult.

Perhaps after 20 years of educational innovation, Australia might experiment with something genuinely radical.

Let teachers teach. Give them the time and authority to maintain orderly classrooms. Expect students to learn. Allow consequences to mean something. Permit failure occasionally to mean failure. Pay teachers competitively and provide working conditions that make remaining in the profession attractive.

And when a child cannot reach the bar, rather than congratulating ourselves for compassion as we lower it, provide the support necessary for that child to climb.

It sounds terribly old-fashioned.

It might even work.

For 20 years Australia has asked schools to do everything, teachers to be everyone, and students to be responsible for progressively less.

PISA has been sending us the bill.