Government responses · both chambers · as at 2 September 2026
260
government responses that closed a committee report without answering a single recommendation.
A report leaves the register when a response is tabled. It does not follow that the report was answered. Of the 623 government responses tabled since mid-2022, 260 disposed of the committee’s recommendations with a repeated template sentence — that given the passage of time a substantive response is no longer appropriate — and took a position on none of them. Nothing accepted, nothing rejected, nothing qualified.
That sentence is used 2,266 times across those 260 documents — counting every form of it. A stricter count, of the “notes this recommendation” form alone, gives 2,223, and that is the one the floor below is built from. How many individual recommendations the sentence disposed of is a harder question, and the answer below is deliberately a floor rather than a guess.
The slowest response on the Senate’s record
The Senate select committee’s report A certain maritime incident was tabled on 23 October 2002. The government responded on 1 April 2026 — 8,561 days later, or 23 years, 5 months. It is the longest gap between a report and a response anywhere in the Senate’s register.
The response runs to 18 entries: 17 for the committee’s own recommendations and one under the document’s own final heading, “Minority/dissenting recommendations”. Against every one of them it records that the government notes the recommendation and that, given the passage of time, a substantive response is no longer appropriate — the same sentence, 18 times. Nothing is accepted, rejected or qualified. Read it.
How many recommendations that is, and why the number is a floor
The count of reports is exact. The count of recommendations is not, and this site will not print one as though it were. The authoritative number of recommendations a committee made is in the committee’s report, and for this corpus those reports cannot be reached: the Tabled Documents system’s own report-to-response links reach one of the 260 closures, and a title search of that system finds none of the others, because the reports they closed were tabled before its index begins.
What each response can prove about itself is another matter. Where a closure sets out the recommendations numbered one to N with no gaps, quotes each one, and answers each with the template sentence exactly once, the arithmetic is visible on the page. 81 of the 260 closures do that, and they hold 886 recommendations.
So: at least 886 recommendations were disposed of by a sentence that took a position on none of them. At least, because the other 179 closures are excluded rather than estimated:
| Not counted, because | Documents |
|---|---|
| no recommendation labels in the text | 72 |
| the notes sentences and the recommendation labels disagree | 63 |
| a single label, which a table layout can fake | 22 |
| labels do not run without gaps | 22 |
The largest of those groups is the one that matters most: those documents close a report without naming a single recommendation, often in two or three sentences. Whatever they disposed of is not in the 886. The two-recommendation minimum exists because a response laid out as a table carries no “Recommendation N” heading at all, and one document that met every other test with a single recommendation turned out to have at least five. Every closure, its numbers and the reason it is or is not counted are in recommendation_counts.csv in the public dataset.
It is not routine practice. It comes in bursts.
Form-letter closures are not spread evenly across the years. There were none at all in the first two years of the corpus, and then two clearing events.
| Year tabled | Responses | Form-letter closures | Share |
|---|---|---|---|
| 2022 | 14 | 0 | — |
| 2023 | 55 | 0 | — |
| 2024 | 295 | 205 | 69.5% |
| 2025 | 54 | 4 | 7.4% |
| 2026 | 205 | 51 | 24.9% |
In 2024, 69 per cent of all government responses tabled were form-letter closures. A backlog cleared this way falls just as fast on a chart as one cleared by answering.
Four ways to answer without answering
The count above uses one test, and it is a narrow one: a repeated template sentence. That is the most conspicuous way of closing a report without answering it, and it is not the only way. Three more forms are named below. None is detected automatically, each is illustrated by a document anyone can download, and every one of them would be counted as a substantive response by the number at the top of this page.
The passage-of-time closurecounted above
The response notes each recommendation and declines to address it because the report is old.
A Certain Maritime Incident — 17 recommendations and a heading for the dissenting ones, the same sentence 18 times, 23 years after the report was tabled.
Superseded by eventsnot counted
The recommendations are noted as a group because the thing they concerned has been withdrawn or replaced.
Nature Positive (Environment Protection Australia) Bill 2024 — 40 recommendations — five of the committee’s and 35 from dissenting reports — noted together; the bills were withdrawn in February 2025.
Answered by substitutionnot counted
The recommendations are noted as a group, and the response describes a different package of measures instead.
‘You win some, you lose more’: online gambling — 31 recommendations noted as a group; not one addressed individually.
Enumerated without a positionnot counted
Each section lists the recommendations it addresses, then describes existing policy without accepting, rejecting or qualifying any of them.
Senate Select Committee on Adopting Artificial Intelligence — All 13 final report and all 5 interim recommendations mapped to a section; a position stated on none.
So 260 is a floor. The honest figure for “answered nothing” is higher, and reaching it means asking, for every recommendation a committee made, whether the response stated a position on it. The first part of that work is published: wherever a response quotes nothing from the report it answers, the search index reads the report and shows each recommendation with the fact that the response did not address it individually. The figure itself is not published yet. How this is classified, and why the other three are not.
Corpus: 623 responses tabled since mid-2022, after excluding 35 documents that answer somebody other than a parliamentary committee. Every document, its classification and the exclusion reasons are in the public dataset, licensed CC BY 4.0. If you believe a document is misclassified, write to corrections@toohardbasket.org.au; corrections are listed in the corrections log.