Holding together, just: Latest ANU Poll finds Australians' democratic support intact but fragile as economic pressures mount
Democracy Data Update 2
In recent months, global economic disruption and geopolitical instability have driven rising costs of living and growing financial stress for Australian households. These pressures are testing democratic resilience, or the capacity of our institutions, communities, and civic infrastructure to absorb shocks and adapt without losing democratic support.
As part of the Network’s effort to monitor democratic resilience, track how are Australians’ trust and attitudes towards democracy holds up under pressure, and where to focus to strengthen systems if fault lines emerge.
To help explore this question, this is the second in our Democracy Data Update series to help make sense of the latest monitoring and analysis that matter for Australia’s democratic resilience over time.
In a recent post for the NumberCruncher Substack, Nick Biddle summarises the latest ANUpoll wave conducted between 11 and 26 March 2026 with 3,662 adult Australians.
The data and analysis suggests that Australians have already begun to price in an anticipated deterioration: wellbeing has fallen, financial stress has risen, and subjective employment risk has reached pandemic-level highs.
With such a large drop in subjective wellbeing, we might expect that Australians will have also worsened in their support for the system. To a certain extent they have, but as outlined below, Australia’s democracy remains resilient with clear strengths, but just.
Summary of findings: The evidence, taken as a whole, does not suggest Australia is on the verge of democratic breakdown. Australians remain broadly supportive of democratic institutions, and the stability of satisfaction with democracy in the face of a substantial deterioration in wellbeing is, by one reading, a reassuring sign of democratic resilience. The question for the coming months is not whether Australia’s democratic consensus is strong on average. It clearly is. The question is whether it is strong enough, and evenly enough distributed, to absorb what may be a sustained period of economic and geopolitical strain.
Confidence the country’s direction is declining- it has to rebound
Australians’ confidence in the country’s direction has declined to lowest levels seen in recent waves of the ANUPoll.
The NumberCruncher post focused on individual measures of wellbeing. In the full discussion paper, authors Nick Biddle and Matt Gray looked at a few related indicators including satisfaction with the direction of the country and confidence in key institutions. These questions from the survey capture views of life in Australia beyond one’s own individual wellbeing and financial position.
The paper found that for the first time in the very long history of ANUpoll (going back to 2008) Australians are more likely to say that they are not very or not at all satisfied with the direction of the country (54.0 per cent) than they are to say that they are very or fairly satisfied (46.0 per cent).
The figure below looks at changes since just prior to the pandemic.
Confidence in the Federal Government has also been tracked in the ANUpoll since prior to the COVID-19 pandemic. Unlike the periods during and after the Black Summer Bushfires in January 2020 where confidence rebounded, and similar rebounds early 2022, the current trends continue to decline.
Democratic satisfaction is holding- but only just
ANUpoll has recently added questions to track satisfaction with democracy in Australia as a related, but distinct, measure to confidence. Confidence measures in specific institutions don’t capture attitudes and view of the system as a whole.
The core tracking question on democracy, asked in every ANUpoll since January 2023, is ‘On the whole, are you very satisfied, fairly satisfied, not very satisfied or not at all satisfied with the way democracy works in Australia?’ Along with the recently published APSC Satisfaction with Democracy Pulse report, this the only other nationally representative Australian survey which has tracked satisfaction with democracy across this full post-Voice, post-election, post-tariff-shock window. Other surveys include versions of this variable, including the McKinnon Poll, but do not have the same frequency as ANUPoll and APSC survey.
Satisfaction with democracy in the ANUPoll was highest in the first wave of data collection, with 77.0 per cent of respondents reporting that they were satisfied or very satisfied in January 2023. Similar to APSC findings, there has been no clear trend, instead shows high variability when measured frequently.
Satisfaction declined immediately following the failed 2023 Aboriginal and Torres Strait Islander Voice to Parliament referendum, falling to 65.9 per cent, and remaining relatively low in the lead up to the 2025 federal election. This was followed by a short-term post-election increase, with 73.3 per cent of respondents reporting being satisfied or very satisfied in May 2025.
Despite declines in wellbeing, increases in financial stress and economic anxiety, and falling confidence in the Federal Government, the recent surveys waves show there was very little change in overall satisfaction with democracy between December 2025 and March 2026. While satisfaction did edge down slightly over this period, the change was not statistically significant.
Younger and older Australians continue to report higher levels of satisfaction with democracy than those in the middle of the age distribution, a pattern that has been consistent across multiple ANUpolls. When we analyse the factors associated with satisfaction with democracy using an ordered probit model, age, education, and location emerge as the key predictors.
What has changed for both ANUPoll and APSC, is the proportion of respondents reporting they are very dissatisfied. The clear increasingly trend that flags a concern for democratic resilience and wider trust in the system. A forthcoming paper from ARDN colleagues will explore this trend and the cohort in more detail and to differentiate those dissatisfied but wanting to fix democracy versus those who want a different system.
Despite this increase of people very dissatisfied, there is strong evidence that support for democracy has not declined over the past ten months. In a previous paper (Biddle and Gray 2026), the authors drew a clear distinction between satisfaction with the way democracy is functioning in practice and the more stable underlying support for democratic principles.
The results of this survey indicate that the overwhelming majority of Australians continue to support fundamental democratic principles, including the independence of the rule of law and strong protections for media and political freedom.
What is particularly notable is that much of the increase in support has occurred among demographic groups that were previously less supportive of democracy.
Wellbeing and attitudes towards democracy are connected- that is the challenge
Although satisfaction with democracy is holding steady and is even slightly higher than some of the observations in the ANUpoll time series, a key question is whether this represents a reasonably firm baseline or this is a high point before accelerating decline.
No one really knows what will happen geopolitically, nor what this will mean for economic circumstances, but how we collectively and individually navigate this uncertainty and support each other does matter. Already, a number of financial institutions are predicting a severe recession in Australia which will only increase these wellbeing pressures.
The previous ANUPoll analysis and paper explores how closely related these pressures are to monitored attitudes towards democracy. this by using a regression approach.
The data shows that cross-sectionally there is a very strong relationship between the tracked national and individual wellbeing measures and democratic satisfaction.
Expanding this earlier analysis (Discussion Paper One- Income inequality and democratic resilience – Impacts and policy choices) the figure below summarises similar findings by simply comparing satisfaction with democracy across multiple binary classifications of the wellbeing of the Australian population.
If you are satisfied with the direction of the country, then you are 44.5 percentage points more likely to be satisfied with democracy. If you are confident in the Federal government, you are 43.4 percentage points more likely to be satisfied with democracy. Financial stress and life satisfaction also have a strong association.
It is important to emphasise that these findings should not be interpreted as causal pathways.
It may be that variation in the wellbeing measures in the table lead to differences in satisfaction with democracy. Equally, it may be that individuals’ views about democracy shape their broader attitudes, including life satisfaction and confidence in institutions. Setting this caveat aside, however, the analysis in the more detailed paper controls for a large set of demographic, socioeconomic, and geographic variables, as well as a full range of other attitudinal variables included in the model. As a result, the associations identified are likely to be strong and robust, regardless of the direction of causality.
A fragile floor on a strong foundation
There is real risk of a wellbeing drop turning into an anti-democratic surge
There are three potential reasons why the drop in individual and national wellbeing has not translated into immediate significant declines in satisfaction with democracy.
The first is that there is something wrong with the measures. This can’t be ruled out, but it is unlikely due to the consistency in findings both across multiple wellbeing measures and across multiple democracy measures.
The second explanation is that the cross-sectional link is false, and the observed relationship is due to some other variables correlated with both. Again, this can’t be disproven, but given the magnitude of the relationship, and the robustness when a number of controls are included, one wouldn’t want to rely on it.
The third explanation then is that wellbeing measures impact on democratic attitudes, but with a lag. If this explanation is true, however, then sustained declines in wellbeing, in the current period of international democratic pressure, are very risky for the support the population has for our democracy. Betting on the last of these explanations not being true is a very risky proposition.
The framing of the ANU paper as Holding Together, Just is deliberate by the authors.
The paper’s finding that satisfaction with democracy is positively and strongly associated with institutional trust, financial security, and life satisfaction does not inevitably mean that declines in those measures will translate into democratic disaffection. Democracies have weathered worse.
But the same analysis identifies where there are potential fault lines.
Support is weakest among those experiencing financial stress, those with low institutional trust, and those whose own wellbeing is low. What this means is that if the lagged explanation is correct, if wellbeing declines eventually flow through into democratic disaffection, it is not aggregate support that will crack first, but support among those groups whose margin is already thin.
These findings from the ANUpoll sit alongside a growing body of regular tracking data on democratic attitudes in Australia. An earlier post on this Substack summarised newly released data from the Australian Public Service Commission, which has been running quarterly pulse surveys on democratic satisfaction since June 2023. The APSC data tells a broadly consistent story: overall satisfaction with democracy has remained above 54 per cent across all ten waves, but the groups reporting the lowest satisfaction (women, people in regional areas, and those without post-secondary education) are precisely the groups where margin is thinnest.
The APSC data also shows that trust in apolitical institutions (the AEC, emergency services, courts) remains considerably higher than trust in political ones, a pattern consistent with the ANUpoll’s own Federal Government confidence findings. Together, these two data sources point in the same direction: the democratic consensus is holding, but it is doing so unevenly, and the floor is thinner in some places than the aggregate numbers suggest.
For practitioners and policymakers, this analysis has a specific implication.
The fault lines are not evenly distributed: they are concentrated among those under financial stress, with lower institutional trust, and with lower education. How we collectively support each other during these times of additional financial stress, matters most for those who are already experiencing hardship.
Further, there are opportunities to better understand what strengthens democratic attitudes and trust in the institutions to navigate these challenges. While confidence for the future is at its lowest levels, literature has consistently suggested that the conditions around these public narratives matter. Broad awareness campaigns and one-off engagement initiatives are unlikely to reassure these groups.
What the evidence points toward is the opportunity for sustained, embedded, participatory mechanisms that create better feedback loops for groups to meaningful voice their needs in the decisions that affect them. This needs to be done before democratic disaffection becomes entrenched.
The question for the coming months is not whether Australia’s democratic consensus is strong on average. It clearly is. The question is whether it is strong enough, and evenly enough distributed, to absorb what may be a sustained period of economic and geopolitical strain.
 | A guest post by
|