MIL-OSI Export

Source: The Conversation (Au and NZ) – By Maria O’Sullivan, Associate Professor of Law, Member of Deakin Cyber and the Centre for Law as Protection, Deakin University, Deakin University

In the wake of the Bondi terror attack, multiple state governments have passed laws to restrict mass protests. Most notably, the New South Wales government introduced controversial legislation giving police the power to restrict public protests for up to 90 days after a terrorist incident.

This unprecedented restriction was tested when thousands took to the streets to protest against the visit of Israeli President Isaac Herzog.

The protest was followed by widespread allegations of police brutality. These are currently being investigated by the state’s police watchdog.

The organisers of the protest are also challenging the constitutional validity of the laws in the NSW Court of Appeal.

This latest round of law reform follows long-running concerns about whether the right to protest in Australia is under threat. Two years ago, the Human Rights Law Centre found “protest is in peril”. Little has changed since.

So, what do these state laws do, and how might they affect people’s democratic right to protest? How can lawmakers strike the right balance between human rights and the protection of public safety?

States tightening anti-protest laws

The right to protest has been restricted in many of Australia’s states and territories over recent years through increased fines and sentences for acts such as traffic obstruction.

These laws have been further tightened in the past year.

In NSW, the government passed legislation in December 2025 to allow the police to make a “Public Assembly Restriction Declaration”, following a terrorist attack.

As the name suggests, such a declaration means specific areas are restricted places for protest. If the police make this declaration, it grants them additional powers to:

• move people on

• close specific locations

• search people inside a designated area

• and issue orders to prevent disruption or risks to public safety.

A number of these declarations were made by NSW Police during the protests against Isaac Herzog’s visit.

The declarations have now lapsed, so they’re no longer in operation. However, the legislation allowing police to make them remains in force.

What about in other states?

In Queensland, draft legislation is currently before parliament. It will, among other things, allow the Queensland government to ban certain protest slogans.

If passed, it will be an offence to publicly use a prohibited expression in a way that “might reasonably be expected to cause a member of the public to feel menaced, harassed or offended”. This offence carries a maximum penalty of two years in prison.

Under this legislation, the government has announced they intend to ban two phrases: “from the river to the sea” and “globalise the intifada”.

[embedded content]

In Western Australia, the government has introduced a bill allowing police to refuse a protest permit if they believe the protest is likely to promote hate based on factors such as religion, race, disability, gender, sexuality or ethnicity.

While various protest restrictions have existed for some time, the recent use of police “declarations” is concerning. They may lead to disproportionate use of force or discriminatory policing practices.

Some of the new hate speech offences are also problematic as they exclusively restrict protest slogans used by some pro-Palestinian protesters. These laws can therefore be seen as targeting particular political causes.

Protesters pushing back

A number of challenges to government restrictions have been lodged with varying success in recent months.

In Victoria, protesters challenged a “designated area” declaration prior to the 2026 Invasion Day rallies. The Federal Court found in favour of the protesters, dismissing the declaration as unlawful.

The declaration gave police the power to search a person or vehicle without a warrant and to direct a person to leave the designated area if they refused to remove a face covering when requested to do so.

The court found the declaration unlawful for a number of reasons. First, the assistant police commissioner did not apply the correct legal criteria and did not satisfy key requirements under the relevant legislation. This meant there was a legal error in the way the decision was made.

Second, the assistant commissioner failed to give proper consideration to the right to privacy under the Victorian Charter of Human Rights.

In NSW, the Palestine Action Group recently tried to overturn a government declaration that Israeli President Herzog’s visit would qualify as a “major event” under the Major Events Act.

That declaration was significant because it granted police many of the additional powers outlined above.

That challenge, which had to be brought to court at short notice, did not succeed.

What can we do?

While governments in Australia are correct to address the perils of hate speech and consider public safety, there is a real concern that current legislation is weighted too heavily on the side of protest restriction.

One way lawmakers can strike a balance between the right to protest and other considerations is to adopt a human rights charter. This would include a formal process for considering the right to protest (and other rights).

Importantly, this should also enforce an obligation on police to consider the right to protest when making decisions (as occurs in Victoria under its state charter).

There have also been calls for a federal human rights act, as well as the introduction of human rights charters in states and territories that do not have one.

Combined, this would enshrine important legal recognition of our right to protest, which is fundamental to the functioning of a healthy Australian democracy.

– ref. With more restrictive laws across the country, how can we protect the right to protest? – https://theconversation.com/with-more-restrictive-laws-across-the-country-how-can-we-protect-the-right-to-protest-276070

Evening Report: https://eveningreport.nz/2026/02/19/with-more-restrictive-laws-across-the-country-how-can-we-protect-the-right-to-protest-276070/

Source: The Conversation (Au and NZ) – By Neeraj Sharma, Scientia Associate Professor of Chemistry and ARC Future Fellow, UNSW Sydney

You might have heard the common claim that electric cars aren’t really green – that their lithium-ion batteries rely on “blood” minerals such as cobalt, mined in terrible conditions.

The critique had some truth to it. But this claim is no longer accurate. Electric vehicle (EV) manufacturers have been shifting away from cobalt because it’s expensive, toxic and ethically fraught.

What’s replacing it? Cheaper lithium-ion battery chemistries based on lithium iron phosphate (LFP), which avoid cobalt entirely. If you remember high-school chemistry, you’ll remember batteries have an anode and a cathode. The anode is nearly always graphite. But the cathode can be made from many different minerals and compounds.

This means battery makers have a great deal of choice over which minerals to include. There’s huge innovation taking place in batteries, as the market grows and diversifies across vehicles and energy storage. Even cheaper chemistries are emerging based around salt (sodium-ion), while high-performance solid state batteries are coming close to reality.

What happened to cobalt?

For years, cobalt has been a mainstay in cathodes due to its useful properties, including how much energy it can help store.

When the first commercial lithium-ion batteries arrived in the 1990s, the chemistry relied on cobalt (lithium cobalt oxide). Over time, lithium nickel mangananese cobalt (NMC) oxide and lithium nickel cobalt aluminium (NCA) oxide came to dominate the market, as their high energy density made them ideally suited for portable electronics.

As demand for lithium-ion batteries accelerated, sourcing cobalt began to be a problem. Three quarters of mined cobalt comes from one country: the Democratic Republic of Congo (DRC), which has half the world’s reserves. Australia is second, with 20%.

Cobalt is toxic. In the DRC, many people risk their health in small mines under conditions often described as slave-like. Illegal and legal mines can do huge environmental damage.

This and other issues led researchers to begin working on reducing or cutting cobalt out altogether. This led to low-cobalt chemistries, in which most of the cobalt was swapped for nickel, manganese or aluminium. To date, it’s been difficult to remove cobalt entirely, given how much of a boost it gives to battery capacity and stability.

In parallel, US researchers found the mineral olivine – made of lithium, iron and phosphate (LFP) – was a good candidate for battery cathodes. This discovery gave rise to cobalt-free LFP batteries. LFP chemistry is cheap, non-toxic and safe, though slightly less energy-dense.

These batteries have had a meteoric rise. Last year, 50% of all EV batteries and more than 90% of stationary home and grid batteries used this chemistry.

Given world-leading battery makers now rely heavily on this chemistry, it’s likely LFP batteries will dominate the market for EV and stationary storage applications in the near term.

What’s on the horizon?

The global market for lithium-ion batteries has risen sixfold since 2020 and strong growth is projected to continue. EVs are taking more and more market share – especially in developing nations – and huge grid-scale batteries are proving essential in modern power grids.

What’s next?

Many next-generation batteries nearing the market are being developed for specific jobs – such as powering drones – or to outcompete current technology. Here are four new types to watch:

Sodium-ion: The world’s biggest battery maker, CATL, and other manufacturers are exploring an entirely different chemistry – sodium-ion – in a bid to eventually replace lithium-ion batteries as home or grid batteries. Sodium-ion batteries are typically heavier and less energy-dense than lithium-ion, so they wouldn’t work well in vehicles. But the chemistry has real promise for stationary energy storage.

Lithium-sulfur: These batteries rely on lithium and sulfur or sulfur-carbon composites. They can currently store four to five times more energy than traditional lithium-ion batteries, making them particularly useful for drones and other technologies where maximum power is needed. The challenge is giving them longevity, as the reactions in these batteries are harder to reverse. That means these batteries are harder to recharge many times at present. Several Australian companies are active in this space.

Solid state: Until now, lithium-ion batteries have relied on a liquid electrolyte as the medium for ions to shuttle between anode and cathode. Solid state batteries do away with the liquid, making them inherently safer. They could potentially lead to a drastic boost to energy storage. They’re not mainstream yet because it’s still tricky to get them to work at room temperature without using high pressure. If engineers figure this out, an EV using solid state batteries might travel 1,000km on a single charge.

Flow batteries: In the 1980s, Australian engineers at UNSW invented the vanadium redox flow battery. A cross between a conventional battery and a fuel cell, these typically larger batteries can feed power back to the grid for 12 hours or more, much longer than current lithium-ion battery systems.

These batteries are likely to be useful in renewable-heavy power grids. Lithium and sodium-ion batteries could provide shorter bursts of power to the grid, while flow batteries could kick in for longer periods.

So do we still need cobalt?

These developments are promising. But they don’t mean an end to cobalt entirely. Smaller amounts of cobalt will still be in the lithium-ion batteries in portable devices and EVs for the foreseeable future.

What we are likely to see is more recycled cobalt coming into the mix, as governments accelerate recycling of lithium-ion batteries and promote recycled minerals to be used in new batteries. Over time, the role of “blood cobalt” could disappear.

– ref. ‘Blood cobalt’ is disappearing from batteries – and cheaper, cleaner batteries are arriving – https://theconversation.com/blood-cobalt-is-disappearing-from-batteries-and-cheaper-cleaner-batteries-are-arriving-263808

Evening Report: https://eveningreport.nz/2026/02/19/blood-cobalt-is-disappearing-from-batteries-and-cheaper-cleaner-batteries-are-arriving-263808/

Source: The Conversation (Au and NZ) – By Tammy Williams, Lecturer, University of New England

There is a growing number of “serious incidents” in Australian early childhood services, including day cares.

A serious incident is one that seriously compromises the health, safety or wellbeing of a child. There were 160 such incidents per 100 services in 2024–25. This is up from 148 and 139 in the previous two years.

These figures follow explosive revelations of safety issues and abuse in the sector.

In response, there are several new national child safety measures. These include banning personal phones in early childhood services, improving recruitment, and making sure parents can see a service’s compliance history.

But one key area has not received the attention it needs. This is educator-to-child ratios.

How are ratios supposed to work?

Ratios are presented as a simple numerical safeguard to ensure enough qualified educators are present and working directly with children.

These can differ between states and territories, the type of service and the age of the children. However, for children aged under 24 months at daycare, there needs to be one educator for every four children. As children get older, the ratios change. For example, in New South Wales for children over 36 months, it is 1:10.

Tammy Williams’ recent PhD research interviewed 16 early childhood educators about their workplaces, which were a mix of for-profit services. Some were small, standalone services; others were part of large corporate chains.

They reported how ratios operate in practice can be very different from how they look on paper. This raises serious questions about how safe the system really is.

‘Under-the-roof’ ratios

The “under-the-roof” ratio is a common term in early learning services. Some services use this to calculate ratios based on the total number of children and educators across the building, rather than within each child’s regular room.

For example, an educator might have 12 children in their room instead of the regulated 1:10. But the room next door might have just eight, so on average, the ratio is met in theory. In the PhD study, one educator said:

There are times when you have two or three children over in your room, but they fit in another room, which I don’t understand. It puts so much more stress on the people in that room. I don’t even get why that is allowed.

Or educators might still be counted if they are having a break. This can be permitted in certain situations under some state regulations if the break is short and the educator is still in the same service as the children.

But media reports suggest this is overused by some services. Staff may also be counted when they are working in the office, or cleaning elsewhere in the service. Other staff, such as chefs and might also be included in the official count, even though this isn’t permitted.

As another educator noted:

Ratios, half the time, are not right […]

What are the rules?

Despite its widespread use, the term “under the roof” does not appear in the national laws or regulations. They clearly state staff counted in the ratios must be “working directly” with the children, and they must be “educators”.

So there is a mismatch between the law and how it is applied in practice. Regulators are increasing their inspections – including more unannounced visits. But regulators are also understaffed and services can go years between inspections. For planned inspections, services can “put on an act”, as one educator described.

‘There is a lot of stress’

When educators are stretched, this obviously limits their ability to provide safe, high quality education and care. In the PhD study, one educator also described how not having enough staff can lead to stress, which can flow on to the children.

I feel there is a lot of stress and […] they’re going to pick up on that and they’re going to feel it as well. I feel they miss out on things because there’s so much stuff that educators need to do.

If educators are too busy, it can also interfere with children’s relationships with educators and disrupt their daily routines and sense of security.

Despite Australia needing another 21,000 educators, educators constantly report wanting to leave, because of burnout and workplace conditions. As one interviewee explained:

The people who are passionate about being here are very, very passionate and they’re getting tired. They’re looking at moving into retail positions and cleaning positions.

What happens now?

Some flexibility when using ratios helps services adapt to unexpected day-to-day changes such as sick leave. However, the research suggests some services are using loopholes as a standard way of operating, rather than for emergencies. This leaves children at risk, without adequate supervision.

The following ideas are based on recommendations from the Productivity Commission, other studies and our research:

1. Increase staffing ratios to accommodate daily realities. These include child illness, breaks, hygiene and additional educational needs. The Productivity Commission has suggested 1:3 for babies, and we recommend 1:4 for toddlers and 1:8 for ages three to five. There also needs to be a “floater” – an educator who covers breaks and staff shortages.

2. Create funded cleaning and administrative positions. This would improve educators’ status and job satisfacton, allowing them to use their training to educate and care for children.

3. Tighten the rules. Make sure staffing rules reflect the rooms in which children belong, including only those staff actively working with the children.

Policymakers and the community rightly expect services to strengthen safety. But unchanged ratios leave educators responsible for delivering more under the same minimum staffing rules. This can lead to educator burnout and attrition. Educators need real support to ensure they are in turn, supporting children and families.

– ref. Childcare centres may have clear rules for staff numbers, but there are loopholes to get around them – https://theconversation.com/childcare-centres-may-have-clear-rules-for-staff-numbers-but-there-are-loopholes-to-get-around-them-275075

Evening Report: https://eveningreport.nz/2026/02/19/childcare-centres-may-have-clear-rules-for-staff-numbers-but-there-are-loopholes-to-get-around-them-275075/

Source: The Conversation (Au and NZ) – By Fabrizio Carmignani, Professor, Griffith Business School, Griffith University

This plan will decimate hotels across South Australia, wiping out many of the 26,000 jobs it directly creates.

Australian Hotels Association (South Australia) chief executive Ian Horne, quoted in The Guardian, February 21, 2018

… a majority of pub employees (over 26,000 in SA!) will likely lose their jobs.

Letter signed by the McCallum family, owners of The Lonsdale Hotel, February-March, 2018

SA Best leader Nick Xenophon has said that if his party wins the balance of power in this Saturday’s South Australian state election, poker machine reform would be “a key issue in any negotiations” about the formation of the next government.

Among other reforms, Xenophon has proposed a reduction in the number of poker machines in some pubs by 50% over five years, and the introduction of a $1 maximum bet per spin for machines in all venues other than the Adelaide casino.

The South Australian branch of the Australian Hotels Association (AHA SA), led by chief executive Ian Horne, says the SA Best policy would “decimate hotels across South Australia, wiping out many of the 26,000 jobs it directly creates”.

A letter signed and shared by the owners of one Adelaide hotel went further, saying “a majority” of 26,000 South Australian pub employees would “likely lose their jobs”.

Is that right?

Checking the source

In response to a request for sources to support the claim made in the Lonsdale Hotel letter, Keith McCallum referred The Conversation to the AHA SA.

A spokesperson for the AHA SA pointed The Conversation to a February 2018 newsletter from Ferrier Hodgson Adelaide partner David Kidman, and the ‘No Way Nick’ website, authorised by AHA SA chief executive Ian Horne.

The Conversation asked the AHA spokesperson to quantify what the association meant by “many” jobs, but did not receive a response to that question.

---

Verdict

The claim made by Australian Hotels Association of South Australia that proposed poker machine reforms would wipe out “many of the 26,000 jobs” in the South Australian hotel industry appears to be grossly exaggerated.

The Australian Hotels Association did not provide modelling or evidence to show how “many” jobs might be affected.

The number of gaming related jobs in South Australian hotels in 2015 was around 3,000. In the same year, less than 20% of the South Australian hotel industry’s revenue came from gaming.

The reforms proposed by SA Best aim to reduce the number of poker machines in some hotels, and reduce maximum bet limits, rather than removing the machines entirely.

Based on these factors, the Australian Hotels Association claim greatly overstates potential job losses.

In addition, at least some of the money not spent on poker machines would be spent on other recreational activities.

This means that potential job losses due to poker machine reforms may be partially offset by increases in employment elsewhere in the economy – or even within the same hotels.

---

What changes is SA Best proposing?

Among a suite of reforms, the SA Best party wants to reduce the number of poker machines in pubs with 10 or more machines by 10% each year over the next five years. This reduction wouldn’t apply to not-for-profit community clubs or the Adelaide Casino.

SA Best is also proposing the introduction of a $1 maximum bet per spin and a maximum win of $500 for machines in pubs and and not-for-profit community clubs.

SA Best leader Nick Xenophon said these reforms would reduce the number of poker machines in South Australia from around 12,000 to around 8,000, and reduce potential personal losses on pokies in pubs and community clubs from around $1,200 an hour to around $120 per hour.

The policy includes a poker machine buyback scheme, a “jobs fund” to assist affected employees, and the possibility of compensation for smaller poker machine operators.

Would ‘many of 26,000 jobs’ be wiped out?

First of all, let’s look at how many people work in the hotel industry in South Australia, and how many of those jobs are related to gaming.

This information is not available from the Australian Bureau of Statistics.

However, in January 2016, the South Australian Centre for Economic Studies published a report that examined the economic contribution of the hotel industry in South Australia.

The report was commissioned by the AHA SA, but it adopts a sound statistical approach to measuring employment in the hotel sector.

According to that report, a total of 26,250 staff were employed in hotels in South Australia in 2015. Of those, 3,048 were classified as gaming staff (or 11.6% of total employment).

Of the 26,250 people employed across the industry, the majority were casual staff (rather than permanent or part-time staff).

The SA Best proposal is to reduce poker machine numbers and maximum bets in some venues, as opposed to removing pokies entirely. So it’s clear that not all 3,000 gaming staff would be at risk.

However, the AHA SA is arguing that reduced revenue from pokies would threaten other jobs.

According to the same report, in 2015, 17% of the South Australian hotel industry’s annual revenue came from gaming. Around 80% of revenue came from liquor sales, food and beverage sales and accommodation.

So even in light of reduced gaming revenue, assertions that “many” or “the majority” of 26,000 pub employees would be affected appear to be unsubstantiated.

Jobs may be shifted elsewhere

To understand what might happen if Xenophon’s proposed reforms were introduced, we need to take two factors into account.

On the one hand, if less money is spent on poker machines, then the number of hours requested to service gaming activities decreases. This could result in less demand for labour, and hence a potential reduction in the number of those roles.

On the other hand, money not spent on gaming could be redirected to other recreational activities – like going to cafes, restaurants and cinemas – or to the retail sector. This would mean that new jobs would be created in other parts of the economy.

Spending diverted to food and beverage sales and other forms of entertainment could also see new jobs created within the same venues.

Another report conducted in 2006 by the South Australian Centre for Economic Studies, commissioned by the South Australian Independent Gambling Authority, found that following the introduction of electronic gaming machines in South Australia, employment in hotels did increase.

However, most of this increase came at the expense of other businesses, like cafes and restaurants. This shows that there is a strong substitution effect in employment between gaming activities and other recreational activities.

Having been published in 2006, the exact numbers in the report are dated. But the qualitative argument is unlikely to have changed. This conclusion is also supported by more recent studies.

In summary, while some of the 3,000 gaming-related jobs in the hotel industry may be lost as a result of the proposed poker machine reforms, claims that “many” or “the majority” of 26,000 jobs would be lost are grossly exaggerated, and not supported by available evidence or existing research. – Fabrizio Carmignani

Blind review

I agree with the conclusions of this FactCheck.

The assertions that “a majority” or “many” of the 26,000 jobs in the South Australian hotel industry would be lost if the proposals put forward by SA Best were to be implemented are gross exaggerations.

They might not be quite as gross an exaggeration as the analogous assertions made in Tasmania during that state’s recent election campaign, but they are an exaggeration, nonetheless. – Saul Eslake

---

The Conversation is fact-checking the South Australian election. If you see a ‘fact’ you’d like checked, let us know by sending a note via email, Twitter or Facebook.

The Conversation thanks The University of South Australia for its support.

---

The Conversation’s FactCheck unit was the first fact-checking team in Australia and one of the first worldwide to be accredited by the International Fact-Checking Network, an alliance of fact-checkers hosted at the Poynter Institute in the US. Read more here.

Have you seen a “fact” worth checking? The Conversation’s FactCheck asks academic experts to test claims and see how true they are. We then ask a second academic to review an anonymous copy of the article. You can request a check at checkit@theconversation.edu.au. Please include the statement you would like us to check, the date it was made, and a link or a photo if possible.

– ref. FactCheck: would pokies reform in South Australia wipe out ‘many’ of 26,000 jobs? – https://theconversation.com/factcheck-would-pokies-reform-in-south-australia-wipe-out-many-of-26-000-jobs-93189

Evening Report: https://eveningreport.nz/2026/02/19/factcheck-would-pokies-reform-in-south-australia-wipe-out-many-of-26-000-jobs-93189/

Source: The Conversation (Au and NZ) – By Adrian Beaumont, Election Analyst (Psephologist) at The Conversation; and Honorary Associate, School of Mathematics and Statistics, The University of Melbourne

One Nation is just two points behind Labor in a Victorian state DemosAU poll, with the election in November. A federal Resolve poll has One Nation and the Coalition tied at 23% each with Labor well ahead.

The Victorian state election is in late November. A DemosAU and Premier National poll, conducted February 1–10 from a sample of 1,274, gave the Coalition 29% of the primary vote (down eight since an October DemosAU poll), Labor 23% (down three), One Nation 21% (not previously asked for), the Greens 15% (steady) and all Others 12% (down ten).

The Coalition led Labor by 53–47 after preferences, a two-point gain for the Coalition. Even if Labor fell below One Nation on primary votes, Greens preferences would boost them back ahead of One Nation.

On this poll’s results, the Coalition would be likely to win a majority of the 88 lower house seats against Labor, but One Nation would win some seats that would otherwise go to the Coalition. The result could be a Coalition government dependent on One Nation support.

Labor Premier Jacinta Allan was at net -37 favourable, with 53% giving her a negative rating and 16% positive. Liberal leader Jess Wilson was at net +3 (27% positive, 24% negative). Wilson led Allan as preferred premier by 40–31 (former Liberal leader Brad Battin had led by 40–32 in October).

Crime was rated the most important issue by 30%, followed by cost of living on 29% and housing affordability on 12%.

By the November election, Labor will have governed Victoria for the last 12 years and 23 of the last 27. An “it’s time” factor may explain some of Labor’s bad polling.

Upper house voting intentions were 28% Coalition (down two since October), 20% One Nation (up nine), 19% Labor (down two), 14% Greens (steady), 5% Legalise Cannabis (up three) and 4% Animal Justice (up one).

All 40 upper house seats will be up for election in November using eight five-member electorates by proportional representation with preferences. The Coalition and One Nation would be favoured to win a combined majority.

Also in Victoria, a byelection will be needed in Nepean after Liberal MP Sam Groth resigned. At the 2022 state election, Groth defeated Labor in Nepean by 56.4–43.6, from primary votes of 48.1% Groth, 32.6% Labor and 8.8% Greens.

Federal Liberal leadership change and Resolve poll

Last Friday Angus Taylor was elected new federal Liberal leader, defeating former leader Sussan Ley in a spill of Liberal MPs and senators by 34–17. Ley has said she will resign from parliament, setting up a byelection in her seat of Farrer.

At the 2025 federal election, Ley defeated independent Michelle Milthorpe in Farrer by 56.2–43.8. The Liberal vs Labor two-party vote in Farrer was 62.9–37.1 to Ley. Primary votes were 43.4% Ley, 20.0% Milthorpe, 15.1% Labor, 6.6% One Nation and 4.9% Greens. The Nationals are likely to contest the byelection, so the main contenders are One Nation, Milthorpe, the Liberals and the Nationals.

A national Resolve poll for Nine newspapers was conducted February 8–14 (in the days before and after the Liberal spill) from a sample of 1,800. It has been broken into a post-spill and pre-spill sample. The post-spill sample included pre-spill polling that asked how respondents would vote if Taylor replaced Ley.

The post-spill sample gave Labor 32% of the primary vote (up two since the January Resolve poll), the Coalition 23% (down five), One Nation 23% (up five), the Greens 11% (up one), independents 6% (down one) and others 5% (down two). In the pre-spill sample, One Nation had led the Coalition by 25–20 with other parties’ support similar.

No two-party estimate was given, but applying 2025 election flows to the post-spill sample gives Labor nearly a 54–46 lead over the Coalition, a three-point gain for Labor since January.

Anthony Albanese’s net approval improved two points since January to -20, with 55% rating him poor and 35% good. In her final poll as Liberal leader, Ley’s net approval slumped 15 points to -23. Albanese led Ley as preferred PM by 38–22 (33–29 in January).

Pauline Hanson’s net likeability was +7, Liberal Andrew Hastie’s was +4, Taylor’s was +3, Nationals leader David Littleproud’s was -8 and Labor Energy Minister Chris Bowen’s was -8. On issues, 45% rated cost of living most important, with 10% for immigration.

The Liberals led Labor by 24–23 on keeping the cost of living low (29–26 in January). On economic management, there was a 26–26 tie (31–26 to the Liberals previously).

Pre-spill Morgan poll has Labor gaining

A national Morgan poll, conducted February 9–13 (before the Liberal spill) from a sample of 1,216, gave Labor 30.5% of the primary vote (up two since the February 2–8 Morgan poll), One Nation 25% (up 0.5), the Coalition 20% (down 2.5), the Greens 13% (down 0.5) and all Others 11.5% (up 0.5).

Labor led the Coalition by a blowout 58.5–41.5 on respondent preferences, a five-point gain for Labor. On 2025 election flows, Labor led by 55–45, a two-point gain for Labor. No Labor vs One Nation two-party estimate was reported.

Tasmanian DemosAU poll has Liberals down

A Tasmanian state DemosAU poll, conducted January 27 to February 12 from a sample of 1,071, gave the Liberals 35% of the primary vote (down six since an October DemosAU poll), Labor 23% (down one), independents 17% (up three), the Greens 15% (steady), the Shooters 4% (up two) and others 6% (up two). One Nation was not asked for as it is not yet a registered party in Tasmania.

Tasmania uses a proportional system so a two-party estimate is not applicable. Liberal Premier Jeremy Rockliff’s net approval slid three points to +2, while Labor leader Josh Willie’s net approval was down two points to -7. Rockliff led Willie as preferred premier by 43–32 (46–34 previously).

Other politicians listed were Liberal Treasurer Eric Abetz (-21 net approval), former Labor leader Dean Winter (-23) and Greens leader Rosalie Woodruff (-21). Winter’s net approval jumped ten points since October.

Upcoming UK byelection and US special elections

I wrote for The Poll Bludger last Thursday about a February 26 United Kingdom parliamentary byelection in a safe Labour seat and three United States federal special elections for the House of Representatives that will occur by August. Democrats are attempting to gerrymander Virginia’s 11 House districts for a 10–1 Democratic split.

– ref. The Coalition leads in Victorian DemosAU poll, with One Nation posting 21% support – https://theconversation.com/the-coalition-leads-in-victorian-demosau-poll-with-one-nation-posting-21-support-275791

Evening Report: https://eveningreport.nz/2026/02/19/the-coalition-leads-in-victorian-demosau-poll-with-one-nation-posting-21-support-275791/

Source: The Conversation (Au and NZ) – By Rob Manwaring, Associate Professor, Politics and Public Policy, Flinders University

South Australians head to the polls on March 21, with Premier Peter Malinauskas’s Labor Party on track to win by a landslide.

But while a predicted landslide in a state election would not normally garner deep national attention, the SA election will be closely watched this time as a microcosm of the changing dynamics of Australian politics.

The headline two-party preferred figure of 61–39% in favour of Labor is masking the broader splintering of the right side of politics. The Liberals are haemorrhaging votes to One Nation, which leads them on the primary vote by 20–19%.

Liberal woes

While the federal Liberals are suffering from leadership and coalition instability, the SA branch has its own systemic problems. The state Liberals have been a party of almost perpetual opposition, only forming government when Labor is plagued by scandal and longevity in office. The Liberals have only won five elections out of 17 since 1965.

SA does not have a coalition, as the Liberal and Country parties merged in 1932. There has been only sporadic Nationals representation in parliament, with the most recent Nationals MP, Karlene Maywald, controversially serving in the Rann Labor Cabinet from 2004–10.

Geography is a major contributor to Liberal electoral woes, with much of their vote traditionally being concentrated in rural and regional seats. One Nation is running candidates in all lower house seats, and would be most hopeful of picking up some of the more regional Liberal seats, such Hammond. Six of the Liberals’ 13 current seats might be under threat, along with two or three of the crossbench seats such as Mount Gambier.

More pointedly, the party has also long been bedevilled by factional infighting and the regular defection of rural members to the crossbench. These rural defectors have demonstrated an ability to “dig in” across usually safe Liberal seats. Rural and regional MPs disproportionately dominate the Liberal caucus, which skews its ability to appeal to more metropolitan seats.

The Liberals are suffering from leadership churn. Liberal leader Ashton Hurn, who was a media advisor to former Liberal Premier Steven Marshall, took over from Vincent Tarzia in December 2025. Tarzia had only served as leader since August 2024, having taken over from David Speirs, who was convicted for supplying a controlled substance.

Under this tumult, Labor consolidated its lead by picking up two crucial byelection wins in Dunstan and Black. The Liberals currently hold a nominal 13 of the 47 seats in the lower House – a record low. Current polling indicates they will lose more seats at the March election, and there is even a chance they will be left without a single lower house MP.

One Nation and the far right in SA

One Nation is a late entrant to SA politics, and has only elected one member, Sarah Game, to the Legislative Council in 2022. One Nation has historically been organisationally weak in the state – it even failed to register its candidates in time for the 2018 state election. Game subsequently resigned from the party in late 2025, following a preselection dispute with Hanson – a common occurrence in One Nation’s history.

Hanson recently announced former senator Cory Bernardi would head the party’s upper house ticket in 2026.

Bernardi served as a Liberal senator for SA in 2006–17, before forming the short-lived Australian Conservatives party between 2017–19. Sharing a similar Christian Conservative values base, the party merged with the pre-existing Family First party – a morality-focused conservative party with an enduring presence in SA politics.

Bernardi is a polarising figure with lower popularity in the state than Hanson. So there is a question about whether his conservative credentials will add to One Nation’s electoral support.

[embedded content]

One Nation’s prospects

The election campaign to date has been relatively uncontroversial, despite lingering issues of ambulance ramping, the algal bloom and the cancellation of the Adelaide Writer’s Week being potential weaknesses for the government. Malinauskas and Labor have been emphasising their “building” credentials, pointing to major infrastructure projects such as the North-South Corridor, and promising significant urban expansion.

Campaign activity has tended to be focused in Liberal-held seats. These range from urban marginals such as Morialta, Unley and Colton to usually more secure regional areas such as Ngadjuri, Hammond and Kavel, which are being encroached on by Adelaide’s urban sprawl.

The Liberals are struggling to offer a policy narrative. For example, there is confusion over their signature stamp duty policy, which has changed in the switch from Tarzia to Hurn.

The electoral conditions seem ideal for a One Nation surge. However, there are reasons to temper expectations. One Nation tends to poll most strongly in rural seats. Several rural electorates in SA already have independent incumbents, or high-profile independent candidates who have been in the field for months. This may complicate One Nation’s path to victory in these areas.

Crisis on the centre-right

The Malinauskas government is poised for an emphatic win on election night. Yet, the foundations of the results are grounded less in Labor’s policy record than in the deepening crisis of the centre-right in Australia.

In common with countries such as the United Kingdom, France, Sweden and Germany, there is a growing fragmentation of the right. A recent academic study of state politics shows how these dynamics play out in very specific and localised ways in Australia.

At the heart of this is an ideological splintering, with a more assertive conservative politics increasingly at odds with the more socially liberal but economically classical traditions. The SA result may offer insight into how deep this fracture is.

– ref. Labor is set for a big win in the South Australian election. But will One Nation cause ructions? – https://theconversation.com/labor-is-set-for-a-big-win-in-the-south-australian-election-but-will-one-nation-cause-ructions-275671

Evening Report: https://eveningreport.nz/2026/02/19/labor-is-set-for-a-big-win-in-the-south-australian-election-but-will-one-nation-cause-ructions-275671/

Source: The Conversation (Au and NZ) – By Kelsie Boulton, Senior Research Fellow in Child Neurodevelopment, Brain and Mind Centre, University of Sydney

When it comes to autism, few questions spark as much debate as how best to support autistic people with the greatest needs.

This prompted The Lancet medical journal to commission a group of international experts to propose a new category of “profound autism”.

This category describes autistic people who have little or no language (spoken, written, signed or via a communication device), who have an IQ of less than 50, and who require 24-hour supervision and support.

It would only apply to children aged eight and over, when their cognitive and communication abilities are considered more stable.

In our new study, we considered how the category could impact autism assessments. We found 24% of autistic children met, or were at risk of meeting, the criteria for profound autism.

Why the debate?

The category is intended to help governments and service providers plan and deliver supports, so autistic people with the highest needs aren’t overlooked. It also aims to re-balance their under-representation in mainstream autism research.

This new category may be helpful for advocating for a greater level of support, research and evidence for this group.

But some have raised concerns that autistic people who don’t fit into this category could be perceived as less in need and excluded from services and funding supports.

Others argue the category doesn’t sufficiently emphasise autistic people’s strengths and capabilities, and places too much emphasis on the challenges that are experienced.

What did we do?

We conducted the first Australian study to examine how the “profound autism” category might apply to children attending publicly funded diagnostic services for developmental conditions.

Drawing on the Australian Child Neurodevelopment Registry, we examined data from 513 autistic children assessed between 2019 and 2024. We asked:

• how many children met the criteria for profound autism?
• were there behavioural features that set this group apart?

Because we focused on children at the time of diagnosis, most (91%) were aged under eight years. We described these children as being “at risk of profound autism”.

What did we find?

Around 24% of autistic children in our study met, or were at risk of meeting, the criteria for profound autism. This is similar to the proportion of children internationally.

Almost half (49.6%) showed behaviours that were a safety risk, such as attempting to run away from carers, compared with one-third (31.2%) of other autistic children.

These challenges weren’t limited to children who met criteria for profound autism. Around one in five autistic children (22.5%) engaged in self-injury, and more than one-third (38.2%) showed aggression toward others.

So, while the category identified many children with very high needs, other children who didn’t meet these criteria also had significant needs.

Importantly, we found the definition of “profound autism” doesn’t always line up with the official diagnostic levels which determine the level of support and NDIS funding children receive.

In our study, 8% of children at risk of profound autism were classified as level 2, rather than level 3 (the highest level of support). Meanwhile, 17% of children classified as level 3 did not meet criteria for profound autism.

Our concern

We looked at children when they first received an autism diagnosis. Children were aged 18 months to 16 years, with more than 90% under the age of eight years.

This aligns with our earlier research, showing the average age of diagnosis in public settings is 6.6 years.

From a practical perspective, our biggest concern about the profound autism category is the age threshold of eight years.

Because most children are already assessed before age eight, introducing this category into assessment services would mean many families would need repeat assessments, placing additional strain on already stretched developmental services.

Second, modifications will be needed if this criteria is going to be used to inform funding decisions as it didn’t map perfectly onto level 3 support criteria.

On balance, however, our results suggest the profound autism category may provide a clear, measurable way to describe the needs of autistic people with the highest support requirements.

Every autistic child has individual strengths and needs. The term “profound autism” would need to be promoted with inclusive and supportive language, so as to not replace or diminish individual needs, but to help clinicians tailor supports and obtain additional resources when needed.

Including the category in future clinical guidelines, such as the national guideline for the assessment and diagnosis of autism, could help ensure governments, disability services and clinicians plan and deliver supports.

What can you do in the meantime?

If you’re concerned your child requires substantial support, here are some practical steps you can take to ensure their needs are recognised and addressed:

Explain your concerns

Not all clinicians have experience working with children with high support needs. Be as clear as possible about behaviours that affect your child’s safety or daily life, including self-injury, aggression or attempts to run away. These details, while difficult to share, help give a clearer picture of your child’s support needs.

It can also be a challenge to find and access clinicians with appropriate expertise. Another potential benefit of having a defined category is that it can better help families navigate care.

Ask about support for the whole family

Our studies show that many caregivers want more support for themselves but don’t always ask. Talk with clinicians about supports for yourself as well, including respite, or family support groups.

Reach out

Coming together with other carers and families can reduce your own isolation and normalise many of the unique challenges you face. Connecting with like-minded people can provide a supportive, empathetic and empowering community.

Plan for safety

For children with high support needs, prioritise safety planning with your child’s care team. This can include strategies to reduce risks, as well as planning how best to support your child’s interactions with health, education and disability services over time.

---

Read more: Parents of autistic children are stressed. Here’s what they want you to know

---

– ref. A new diagnosis of ‘profound autism’ is on the cards. Here’s what could change – https://theconversation.com/a-new-diagnosis-of-profound-autism-is-on-the-cards-heres-what-could-change-271930

Evening Report: https://eveningreport.nz/2026/02/19/a-new-diagnosis-of-profound-autism-is-on-the-cards-heres-what-could-change-271930/

Source: The Conversation (Au and NZ) – By Rosanna Scutella, Senior Research Fellow, Social and Global Studies Centre, RMIT University

The Conversation fact-checks claims made on Q&A, broadcast Mondays on the ABC at 9.35pm. Thank you to everyone who sent us quotes for checking via Twitter using the hashtags #FactCheck and #QandA, on Facebook or by email.

---

[embedded content]

What’s incredible when you look at those numbers is about 30% – it’s hard to tell often – about 30% of those homeless people have a job.

– Rebecca Huntley, social researcher and author, speaking on Q&A, April 23, 2018

Inequality, class and social mobility in Australia were key issues discussed on a recent episode of Q&A.

Social researcher and author Rebecca Huntley noted an uptick in the idea of “the undeserving poor” in Australia – particularly where homeless people are concerned.

Huntley noted the perception held by some that homeless Australians are simply “not working hard enough”.

Challenging that narrative, Huntley said “about 30% of those homeless people have a job”.

Is that right?

Checking the source

In response to The Conversation’s request for sources, Huntley provided data from the Census of Population and Housing: Estimating homelessness, 2016 report, published by the Australian Bureau of Statistics in March 2018.

Huntley also pointed to the article: “The rise of homelessness and hunger”, written by Mike Seccombe and published in The Saturday Paper, and the website of “Everybody’s Home – A National Housing Campaign”.

Huntley added:

The definition of homelessness in the Census is probably broader than community perceptions about homelessness – that is, that all homeless people are sleeping rough on the streets.

People who are couch surfing or living in their car or living in overcrowded accommodation may well fit into this definition. They may also be working in the gig economy or getting work here and there (the double whammy of insecure work and insecure housing is quite terrible).

What the Census 2016 data show is that there are people with post compulsory education, with various levels of work and hours worked across all categories of people living in insecure housing arrangements.

---

Verdict

Based on the best available data, Rebecca Huntley’s statement that “about 30% of … homeless people have a job” is correct.

According to Census 2016 data, about 30% of people who were recorded as being homeless on Census night (using the Australian Bureau of Statistics definition of homeless) were also recorded as being in the work force.

---

What does it mean to be ‘homeless’?

When we talk about “homelessness”, many of us would think about people “sleeping rough” on the street. This is arguably the most severe and literal form of homelessness. But the state of being homeless is more complex than that.

Under the Australian Bureau of Statistics (ABS) definition, a person can be considered homeless if their current living arrangement:

• is in a dwelling that is inadequate
• has no tenure, or if their initial tenure is short and not extendable, or
• does not allow them to have control of, and access to space for social relations.

The ABS presents its estimates of homelessness using these groupings:

• People living in improvised dwellings, tents or sleeping out
• People in supported accommodation for the homeless
• People staying temporarily with other households
• People living in boarding houses
• People in other temporary lodgings, and
• People living in “severely” crowded dwellings.

On the night of the 2016 Census, more than 116,000 people were counted as being homeless. This includes both children and adults. The estimates of the employment rate include only those age 15 and over.

This may be a conservative count, because some groups of people may be underenumerated (under counted) in the Census.

For example, the ABS notes that Aboriginal and Torres Strait Islander peoples’ are “more likely to be both underenumerated and over represented in the homeless population”, and that:

So called rough sleepers and people staying in supported accommodation for the homeless are also at risk of being underenumerated in the Census.

What constitutes ‘a job’?

In the Census data, people are counted as being employed if they are of working age (age 15 and over) and:

• employed and working full-time
• employed and working part-time, and/or
• employed but away from work.

However, not all people age 15 and over who were experiencing homelessness were counted in the Census labour force statistics. For some people, no information was recorded.

Known employment rates for homeless people

If we calculate the known employment rate for homeless people (using the ABS definition of homelessness outlined above), we find that around 30% are employed, as Rebecca Huntley said on Q&A.

But the employment rate among homeless people could be higher.

That’s because we don’t have employment information for all homeless people. In the Census statistics, there are large numbers of people for whom information on employment status is missing, or not stated.

Overall, we don’t have records of the employment status of about 18% of the total homeless population.

Also, many people experiencing homelessness could be in situations where they wouldn’t be expected to work. For example, full-time students or the elderly.

This makes 30% likely to be the lower bound.

If we assume that the employment rate of those with missing information is the same as those with recorded information, the employment rate would increase to 36%. If we also excluded full-time students and the elderly from these statistics, the rate would be even higher.

---

---

Employment rate for people ‘sleeping rough’

“Sleeping rough”, or sleeping on the street, is arguably the most severe form of homelessness.

People sleeping rough are the group with the highest proportion of missing information on labour force status. The known employment rate for people sleeping rough is 10%.

If about half of the people with missing information were employed, the rate would go up to 30%. My assumption for this group is that most of those people with missing information are not employed.

So for those sleeping rough, the employment rate is probably closer to 10-15%.

The employment rate for people in supported accommodation is also likely to be around 10-15%. These two groups are those usually considered when a more literal definition of homelessness is used.

But as outlined in this FactCheck, the state of being “homeless” is more complex and wide ranging than that.

‘Journeys Home’ survey

Another useful data set on homelessness and employment is the Melbourne Institute’s Journeys Home survey, of which I was the Deputy Director.

This longitudinal survey, which began in 2011 and concluded in 2014, included 1,682 people in Australia flagged by Centrelink as either “homeless” or “at-risk of homelessness”.

The survey also included a group of income support recipients who were not flagged as homeless, but who had characteristics similar to those who had been homeless.

The overall rate of employment among all respondents was 27%. Of those who were homeless, 19% were employed.

In our study, however, we did not include those in overcrowded accommodation as being homeless. (These people are identified as being homeless in the Census).

This highlights the importance of the definition of “homelessness” used when considering the characteristics of the homeless population.

It’s also important to remember that just because someone isn’t employed doesn’t mean they don’t want to be employed, or aren’t seeking employment. Being homeless is a significant barrier to gaining – and retaining – a job. – Rosanna Scutella

---

Blind review

I agree with the verdict of this FactCheck that the overall rate of employment among people experiencing or being at-risk of homelessness is in the vicinity of 30%.

I would add that findings from my research using the Journeys Home data reveal that homelessness is more strongly associated with difficulty in retaining employment than with finding employment. – Neha Swami

---

The Conversation’s FactCheck unit was the first fact-checking team in Australia and one of the first worldwide to be accredited by the International Fact-Checking Network, an alliance of fact-checkers hosted at the Poynter Institute in the US. Read more here.

Have you seen a “fact” worth checking? The Conversation’s FactCheck asks academic experts to test claims and see how true they are. We then ask a second academic to review an anonymous copy of the article. You can request a check at checkit@theconversation.edu.au. Please include the statement you would like us to check, the date it was made, and a link if possible.

– ref. FactCheck Q&A: do ‘about 30% of homeless people have a job’? – https://theconversation.com/factcheck-qanda-do-about-30-of-homeless-people-have-a-job-95514

Evening Report: https://eveningreport.nz/2026/02/19/factcheck-qanda-do-about-30-of-homeless-people-have-a-job-95514/