Month: June 2022

Croakey provided extensive coverage of the 2022 election campaign (www.croakey.org/category/elections-and-budgets/). But because they were simply not available, it included very few projections of the result.

Despite this, polling was (as usual) a prominent source of news and discussion generated during the campaign.

The apparent reluctance of  the pundits and pollsters to go public with predictions was presumably the result of their multiple failures in the lead-up to the 2019 election.

It was perhaps due to the absence of precise projections from the ‘professional’ psephologists that  Croakey’s Editor-in-Chief, Melissa Sweet, approached some casual observers of federal politics about their expectations of the results. On Sunday 15 May, one week before the election, I emailed my thoughts.

The net bottom-line of my predictions was that the election would result in a House of Representatives comprising 78 from the ALP, 65 from the Coalition and 8 independents including the six already on the cross-benches.

With the results declared we know that the House of Representatives will now comprise 77 from the ALP, 58 from the Coalition and 16 independents.

Although the  net prediction ended up being close to perfect, it was made up of a number of erroneous predictions which in effect cancelled each other out.

In the lead-up to 21 May my overall summary – given to people who sought one – was that the ALP had not done badly enough in the campaign to offset two strong forces: dislike of  Scott Morrison and his ways; and the general feeling that it was time for someone else to have a go after nine years of the Coalition.

In my view what happened in the campaign was virtually irrelevant to the election result. What did matter were three things which were fixed or determined by the time the election was called:

1. an aggregated and generalised view across the voting population of the current government, this view built up over three years and shaped by the media;
2. personal voters’ perceptions of the leaders of the major political groupings – their likeability, the way they present on television, and what one has heard about them in the media and in the pub;
3. local, electorate-by-electorate issues, including changes in local candidates (eg the retirement of a popular member) and any cut-through of particular local issues (such as aircraft noise in Brisbane).

The hardest of these three to allow for is the third. Keen observation can give one a pretty accurate assessment of the public’s view of the government and of the perception of party leaders. On the other hand, accurate predictions of the impact of local issues requires intelligence from 151 different locations.

Like everyone else, I failed to pick the strength of the greening of metropolitan seats and of the rejection of the Liberal Party in Western Australia.

I considered electorates which everyone knew would determine the result. I picked eight of the 12 electorates that shifted from the Coalition to Labor: Bennelong, Boothby, Chisholm, Higgins,  Pearce, Reid, Robertson and Swan. But in this category I missed Hasluck and Tangney, and was wrong on Braddon, Lindsay, Longman and Nicholls, all of which were retained by the Coalition.

My prognostications were based partly on the false premise that most of the seats that changed hands would do so between the two major party groups. On the back of my envelope, as well as being wrong on those four that were in the event retained by the Coalition, I also thought that three seats would go the other way – from the ALP to the Coalition: Corangamite, Dobell and Gilmore. It might also be that I underestimated a gender phenomenon because all three of them were retained for the ALP by women (Libby Coker, Emma McBride and Fiona Phillips).

In terms of the numbers, my greatest failure was in relation to the community independents. Whereas I thought the number on the cross-benches would increase from 6 to 8, as we now know it is increasing from 6 to 16.

No one could have picked the fact that three seats in Brisbane would go to the Greens. In relation to high-profile seats, I thought that both Bridget Archer and Trent Zimmerman (North Sydney) would hold on due to the profile and credibility associated with their crossing the floor on the religious discrimination Bill. Bridget Archer did, Trent Zimmerman did not. Also I picked Josh Freudenberg to retain his seat, on the basis of how he presents and communicates and a view about his capability and leadership prospects.

I was correct about a number of seats which it was thought might change hands but which did not. These included Bass, Dunkley, Flynn, Hunter, Parramatta and Warringah (this last already being held by an Independent). One of the interesting matters for the future is how easy it will be for strong independent candidates to build a wall or moat around their incumbency.

In such an exercise there are always outliers that it is impossible to predict. This time they included Fowler (where the result was determined before the election by an administrative order), Tangney, Griffith, Ryan and Brisbane. In Tangney Special Minister of State Ben Morton was beaten by Sam Lim with a swing of 11.44% to the ALP.

This exercise in election-watching should remind us that it is not the number of seats that change hands that is important but the balance – the net balance – between flows in two main directions. An election in which three seats change hands can have the same aggregate result as one in which 33 seats change hands, if the 33 are distributed 15 one way and 18 the other.

On the matter of the extent to which an election campaign influences the result, there is much more to be said.