Contests and polls

How the ACT senate has become contested territory

Public servants, lobbyists, academics and senior military officers, living in Canberra, believe it is where all the nation’s important decisions are made. The reality is that for the nation’s most important decision, the federal election, Canberra hardly counts. Its three House of Representative electorates are seen as safe Labor territory, and its two senate seats have consistently gone one to Labor, one to Liberal.

But there is now a nationally important contest for that second ACT Senate seat.

With two senate seats, the vote threshold for a quota is 33.3 percent (=100/(2+1)). In the 2019 election Labor’s Katy Gallagher easily won a quota with 39.4 percent of the vote, while the Liberal’s Zed Seselja, with 32.4 percent, just fell short, but exhausted votes from a right-leaning independent and the UAP got him over the line after 25 rounds of counting and distribution of preferences. The Greens also polled well with 17.7 percent of the senate vote. (Antony Green has a comprehensive explanation of the 2019 senate vote.)

This time Seselja faces a tougher challenge from two prominent independents standing on generally similar platforms – action on climate change, a strong federal integrity commission, and more humane treatment of refugees. In a remarkably civilized manner they appear together as guests on an ABC Breakfast segment Independents hope for historical win in ACT Senate. (13 minutes)

A Redbridge poll in early April, reported on William Bowe’s Poll Bludger site, shows that in the ACT support for all three parties – Labor, Liberals and Greens – has fallen, while there is strong support for these new independents. That poll showed Labor’s Gallagher on 32.7 percent (just short of a quota), Liberal’s Seselja on 22.7 percent, the Greens’ Tjanara Goreng Goreng on 12.8 percent, independent David Pocock on 9.9 percent and independent Kim Rubenstein on 5.3 percent.

Since then Pocock, already well known in the ACT as a rugby union football player, has lifted his profile considerably. The far-right Advance Australia group has entered the fray, supporting Seselja by mounting savage attacks on Pocock.

While Labor’s Gallagher will safely occupy one seat, the second position could go in any direction – a status-quo with the Liberal’s Seselja scraping over the line, a second quota for Labor, a quota for the Greens, a quota for one of the independents. The ACT, with its highly-educated population, is difficult territory for the Coalition, and public servants and others know from close experience about the Coalition’s incompetence, waste and corruption. Public servants, who still constitute a significant proportion of Canberra’s population, generally vote for “left” leaning parties, and in this election, because Morrison has treated the public service with such contempt, and has taken politicization to new levels, that sentiment will be particularly strong this time.

Because the next parliament will probably once again see a small number of independent and minor-party senators with a legislative balance of power, the ACT contest is an important one, particularly in light of the strong polling by far-right parties vying for senate positions. Yet most national media, who tend to see the election in a traditional winner-take-all two-party contest in the House of Representatives, seem to find the Senate too hard to fit into its reporting.   

National polls: no evidence of movement

William Bowe’s Poll Bludger has results from the latest polls – Newspoll, Ipsos and Roy Morgan. Ignoring the margin of error, a Labor optimist may note that Labor’s Newspoll primary vote is up this week, but a Coalition optimist could make the same observation about the Coalition’s support in the same Newspoll.

In fact when one digs into all three polls, the only conclusion that can be drawn is that there is no evidence of anything that happened in the polling period (generally up to Anzac Day), resulting in any movement in parties’ primary support.

The polls’ two-party-preferred estimates all point to a Labor win, but there is a wide range of estimates, from a 6 point lead up to a 10 point lead. Such a range is unsurprising in view of the accumulated sampling errors and the number of assumptions in going from primary support to TPP estimates. It is notable that polls that calculate TPP figures based on respondents’ choices of preference distributions give Labor slightly higher leads than polls that assume preferences will be distributed the same way as they were in 2019. While the same parties dominate as in 2019, the field of independents this time is different. It would be a brave person who could assert how their preferences will flow.

Poll Bludger reports on a number of state-by-state estimates and some estimates for seats in contention. The only conclusion one can draw is confirmation of what one learns in first-year statistics: small samples give inconsistent results.

But Bowe has dug up some polling gems. One is his own analysis of minor party support, focussing on One Nation and the UAP – Minor revolution: independents’ day is coming, but what will it mean? – published in Crikey (may be paywalled). The two parties’ support bases are different, he points out. One Nation supporters are older, while UAP has captured the support of many younger people, including anti-vaxxers. Countering our usual stereotype of Victoria as the home of left-progressive liberalism, he shows that UAP’s support base is particularly strong in Victoria.

Another polling gem Bowe has uncovered is that the Katherine Deves controversy is “playing well” in the suburbs and non-metropolitan regions. We don’t know, however, if it will have any influence on people’s final vote.

Morrison is trying to make sexual identity and “cancel culture” into election issues, but it would be a hard job to wedge Albanese on these issues: he hardly comes across as hyper-woke.