DemosAU Federal Polling Results - May 2026

Capital Brief/DemosAU federal voting intention poll, conducted 15–20 May 2026.

Primary vote: One Nation 28%, Labor 26%, Coalition 23%, Greens 13%, Other 10%.

Against the 2025 federal election result that is a swing of 21.6 points to One Nation, 8.8 points away from the Coalition and 8.6 points away from Labor, with the Greens up 0.8 and others down 5.0.

Seat projection: Labor 65–74, One Nation 47–58, Coalition 16–28, Greens 1–5, Others 2–6.

Preferred Prime Minister: Anthony Albanese (Labor) 34%, Pauline Hanson (One Nation) 27%, Angus Taylor (Liberal) 23%, don't know 16%.

Leader net ratings: Pauline Hanson +3%, Angus Taylor +1%, Anthony Albanese -20%.

Hypothetical — without Pauline Hanson: asked how they would vote if Hanson announced her retirement and Barnaby Joyce led One Nation, respondents put Labor on 28%, the Coalition 27%, One Nation 21%, the Greens 14% and others 10% — One Nation seven points lower than on the live question. Seat projection: Labor 74–83, Coalition 36–52, One Nation 15–24.

Biggest issue facing Australia: cost of living 39%, housing affordability 14%, immigration 12%, the economy 10%. The question was open-ended and its categories were generated by AI; responses can fall into more than one category, so the totals do not sum to 100%.

Budget: 43% call the 2026 Federal Budget generally bad against 23% generally good. Ending negative gearing for existing homes purchased after 12 May 2026 draws 34% approval to 29% disapproval; scrapping the 50% capital gains tax discount 28% to 29%; a minimum 30% tax rate on family trusts 27% to 34%.

Sample: 1,502 Australians aged 18+ eligible to vote. Effective sample after weighting: 754. Effective margin of error: ±3.6%.

Method: Online, with the sample drawn from internet panels. Rim weighting on age by gender, education, personal income, residential tenure, language, region, 2025 federal vote and left-right self-identification. Federal voting intention was the first substantive question after demographics. Initial undecided voters were asked a leaner question and the remaining 7% of undecided respondents were excluded from the totals. No two-party preferred figure was calculated for this wave.

Full question text, response categories and randomisation are set out in the report.