The highest-leverage decision in youth justice is not reported nationally
A police officer decides to caution or to charge. The Productivity Commission's 2026 report records that data is not available for reporting against the diversionary services indicator — so the decision that determines everything downstream has no national number.
Steps this would change
Rationale
The remand process has nine steps. The one that decides the outcome is the second, it takes zero days, it involves no application and no fee, and it is made by an individual officer.
A caution ends the matter. A charge starts a court process which, three-quarters of the time when bail is refused, means custody before it means a finding.
There is no national rate. The Productivity Commission's Report on Government Services 2026 records that data is not yet available for reporting against the diversionary services indicator. The most important discretion in the system is the one nobody can chart.
What is known comes from research, and it is not reassuring. In New South Wales, Aboriginal and Torres Strait Islander young people are taken to court in 64 per cent of cases against 48, and cautioned in 14 per cent against 28. Studies of New South Wales, South Australia and Western Australia found Indigenous young people less likely to be diverted in all three, after controlling for age, sex, offence type and prior history. That last clause is what makes the finding hard to explain away.
Meanwhile Closing the Gap Target 11 is not on track. The target is a 30 per cent reduction in the detention rate for Aboriginal and Torres Strait Islander young people by 2031. Nationally in 2024-25 the rate was 25.7 per 10,000, below the 2018-19 baseline of 28.3 but well above the 20.5 recorded in 2020-21. New South Wales, Queensland and the Northern Territory are each worsening against it.
What should be published, per jurisdiction, per year, at a minimum:
- The number of young people proceeded against, split by warning, caution, conference and charge.
- The same, split by Aboriginal and Torres Strait Islander status, offence category and prior contact — because a rate difference that survives those controls is a different finding from one that does not.
- Remand episodes, with the proportion in which accommodation was recorded as a factor. See A bed shortage is being recorded as a judicial decision about risk.
Zero days saved, and that is the point. This is the eighth proposal on this site that saves no time and makes something visible. The pattern by now is established: the reforms that worked were preceded by a published number, and the domains with no published number are the ones where nothing has moved.
Who acts: police ministers and state police forces, with the Productivity Commission as the reporting vehicle that already exists.
What is being traded away
Publishing a disparity does not remove it, and can be used to argue the opposite. A published caution rate gap will be read by some as evidence of police bias and by others as evidence that offending differs between cohorts. Both readings will be made from the same table, and the controlled studies that distinguish between them are harder to commission than the table.
Measuring a discretion changes it, sometimes toward the number rather than the judgement. An officer whose caution rate is reported may caution to the target. That is a real risk in both directions: cautioning someone who needed more, or charging to avoid appearing lenient. Every published performance measure in this database carries some version of this, and it is the strongest argument the people who hold the discretion make.
The data collection is not free and the burden falls on the front line. Recording the basis of a street-level decision is administrative work for police, and this site has objected elsewhere to reporting obligations that consume the capacity they are meant to improve.
And a national number will flatten eight very different systems. The age of criminal responsibility is 10 in most jurisdictions and 14 in the Australian Capital Territory. Diversion schemes differ in name, threshold and legal effect. A single national caution rate would compare quantities that are not the same quantity, which is an argument for publishing per jurisdiction and against publishing a headline.
Precedent
Every other step in this database that got faster got measured first. The Commonwealth publishes visa processing percentiles, aged care assessment timeframes and bulk billing rates by region; none of those were published until someone decided they should be.
Sources
- Report on Government Services 2026, chapter 17 (Youth justice services): data not available for reporting against the diversionary services indicator
- Australian Institute of Criminology: in NSW, Indigenous young people are more likely to be taken to court (64% compared with 48%) and less likely to be cautioned (14% compared with 28%)
- Indigenous Justice Clearinghouse: Indigenous young people were less likely to be diverted in NSW, SA and WA, after controlling for age, sex, offence type and prior history
- Productivity Commission Closing the Gap dashboard, Outcome 11: target is a 30% reduction by 2031; the rate in 2024-25 was 25.7 per 10,000 against a 2018-19 baseline of 28.3 and 20.5 in 2020-21; NSW, Queensland and the NT are worsening
- Age of criminal responsibility is 10 in the NT, Queensland, WA, SA and Tasmania; the ACT has raised it to 14