Wednesday, November 11, 2020

Groom: Australia's Most Boring By-Election?

Where time permits I aim to do a preview post for any federal by-election, but in the case of Groom (Qld, LNP, 20.5%) I'm not expecting to be hanging on the edge of my seat on November 28, and nor am I expecting any flurry of polls.  The major purpose in writing this guide is to point to some unusual features of this by-election in terms of its lack of competitiveness.  However, it is still an electoral indicator of some kind, and the swing will be watched with some interest in view of events affecting both major parties this week.

Groom History

Groom is mostly (in population terms) the city of Toowoomba, plus surrounding rural areas radiating to the west.  Groom is the successor to the Federation division of Darling Downs, the name being changed when the division was redrawn for the 1984 expansion.  The seat has had only nine incumbents, all of them male, since Federation (one of whom, Sir Littleton Groom, served two disjunct spells in the seat.)  

The seat has invariably been won by conservative MPs, with the slight complication of Sir Littleton Groom serving as an independent briefly in 1929 and 1931-3. He was expelled from the Nationalists after not using his casting vote as Speaker to save them from a no-confidence motion, and lost his seat in 1929, but won it back in 1931 and eventually joined the United Australia Party.  The seat has, however, gone back and forth between the Liberal/proto-Liberal side and the Nationals/Country side of the Coalition, with four changes of ownership in cases where it became vacant.  Three of these involved three-cornered contests.  The last of these came at the Groom by-election 1988, where the seat switched from National hands to Liberal hands after ousted Premier Joh Bjelke-Petersen endorsed the Liberal candidate.  In this way, the ghost of Joh hangs over the question of whether the winner should sit in the Liberal or National party rooms.  

Labor's closest approach to winning Darling Downs was 47.5% 2PP in 1943 (a landslide national win for the party during wartime).  Since then, Labor has got above 40% 2PP just twice, 43.1% in the 1961 "credit squeeze" election which the Menzies government barely survived, and 41.8% in the 2007 Rudd victory, with Labor led by a Queenslander.  

The 2019 result (70.5% to Jon McVeigh) was one of the most lopsided results in the seat's history, exceeded only on a 2PP basis by a 71.3% result in 1996.  There is a myth that Groom was the Coalition's biggest win in 2019 but in fact it was exceeded on both a 2PP and 2CP basis by Maranoa (where One Nation finished second).  

Groom And Lopsided By-Elections

Groom is remarkably uncompetitive on paper, and the by-election has attracted quite a lack of candidate interest, possibly in part because many parties have exhausted resources on the Queensland campaign.  Normally by-elections attract a plague of no-hopers, often from out of town, but Groom has attracted only four candidates.  

Groom is the most lopsided seat based on the results of the previous election to go to a by-election since Scullin (ALP, Vic, 27.6%) in 1986.

Groom is the first federal by-election not contested by the Greens since Warringah, 1994, and breaks a streak of 30 successive by-elections that the party has contested.  It's hardly the worst seat for the party, which polled 8% there in 2019, but they have said they are focusing on the long-away Queensland Senate campaign.  The party is probably still recovering from the Queensland state election campaign.  Incidentally, the Greens' Eden-Monaro result earlier this year was their second worst in swing terms at any by-election, after only the artificial case of Wentworth 2018 (where much of their usual vote went to independent Kerryn Phelps.)

The total field of four is the smallest for a by-election since four candidates contested the 1995 Wentworth by-election.  The last by-election with fewer than four was Gwydir 1989, contested by one National and two right-wing independents.  There was speculation that Labor wouldn't bother contesting Groom either, but this would have actually been very unusual.  North Sydney 2015 was a rare case of an Opposition not contesting a Government vacancy; before that the most recent one I could find was Bradfield 1952.

Campaign

The most interesting thing about the Groom campaign so far was the LNP preselection contest, conducted in the lead-up to the Queensland election.  Media reports claimed that Toowoomba doctor David van Gend was a frontrunner in the seat.  This produced widespread alarm owing largely to van Gend's history of homophobic comments and his frequent and recent support for so-called "gay conversion therapy".  However, van Gend was defeated by votes pooling between other candidates through successive eliminations, especially with concerns that preselecting him would be a disaster for the state LNP campaign.  Right-wing culture warriors had flocked to van Gend's cause, but not all of them, as mining engineer Garth Hamilton received support from LNP Senators Amanda Stoker and Gerard Rennick.  

Hamilton, the eventual winner, is also of the Right, but in a different way.  His views include support for "pulling apart the Environmental Protection and Biodiversity Conservation Act to free up major infrastructure project approvals", support for IR and taxation changes, opposition to Labor's 2050 zero carbon emission target, and wariness regarding China.  He is a frequent author in Spectator Australia, though a recent piece called "Absent Fathers Matter" that criticised the Black Lives Matter movement has mysteriously disappeared.

Labor showed no clear sign of running until nominations closed, at which point it turned out they were in and their candidate was Chris Meibusch, who also contested the seat for Labor in 2007 and 2010.  In 2007 Meibusch recorded the sixth-highest 2PP swing in the country, 10.6% to Labor, but Groom was not a place where the Gillard years were likely to go down well and it nearly all returned to sender in 2010.  The Latham campaign probably didn't go down there well in 2004 either so these swings were probably not candidate related.  Meibusch, a long-term party member, is a local lawyer who is also secretary of Save Mt Lofty, a local conservation group.  

Meibusch contested the mayoralty of Toowoomba Regional Council this year but finished a very distant second to the winner, Paul Antonio, 78-22 after preferences.  Amusingly, under daft local government laws passed by the Palaszczuk Government recently, had Antonio stepped down in the first year of his term, Meibusch would now become the Mayor.  The Government is now in the process of reversing its blunder, with Meibusch's support.  

Two other parties have candidates contesting this by-election: the Liberal Democrats' Craig Farquharson and Sustainable Australia's Sandra Jephcott.

Prospects

The 2019 election results suggest nothing special about the personal vote of outgoing MP Jon McVeigh, who at that time had only been in the seat for one term.  The swing to the Coalition was slightly higher than the average for Queensland, which is consistent with a typical personal vote for a new MP.  

By-elections usually result in a protest swing against the government of the day.  Since Federation, the average 2PP swing for Government seat by-elections has been around 6.1+/-4.9%.  For by-elections since 1983 it has been slightly lower, but only because of eligibility-related by-elections where the sitting member has recontested.  A Labor win in this seat would not only be around 2.9 standard deviations outside the average (an in-theory 1 in 600 chance) but it would exceed the largest 2PP by-election swing so far (20.1% in Wakefield 1938).  But furthermore, by-election swings are correlated with polling, so monster swings against governments tend to occur when those governments are polling badly - which isn't the case at the moment.  Historic patterns suggest there is no realistic statistical chance for Labor to win this by-election and the question is the margin.

On that point, the by-election looms as a test of whether Labor has yet won back any of the Queensland voters it alienated in 2019.  A greater than average swing could be taken as a sign of success on this point. However, many pandemic-era electoral events have already seen incumbent governments outperforming expectations, and on that basis I'd be more inclined to set the bar for Labor to demonstrate progress at a little lower than average.  Thus, provisionally, I'd set the bar as follows:

No swing or swing to Coalition: excellent result for Coalition, questions about Labor leadership and campaign, suggestion that Labor has not addressed causes of 2019 result
Up to 3 points to Labor: good result for Coalition, mediocre for Labor
3-6 points to Labor: solid result for Coalition if at the lower end but really not much to see here
6-9 points to Labor: good result for Labor in the circumstances
9+ points to Labor: outstanding result for Labor, suggestion Labor has successfully addressed causes of 2019 results in Queensland.

Of the four parties running, on a four-party basis the results in the Senate (including below the lines) were LNP 65.3%, ALP 28.4%, Sustainable Australia 3.6% and Liberal Democrats 2.7%.  However that excludes nearly a tenth of Senate formal votes that reached none of these parties, and those voters (often One Nation) might behave a little less predictably.  So the two minor parties may be able to get the 4% needed to get their deposits back and get public funding.  That said there are not a lot of libertarians in Toowoomba, and the selection of Meibusch as Labor candidate should discourage Greens voters from preferring Sustainable Australia.  If Sustainable Australia polls something in the high single figures and the Labor primary is lacklustre then that would raise some questions as well.

At the moment I expect to cover this count live on the night.

1 comment:

  1. No Greens, no PHON, Labor of course wont win, but will do very very well.

    ReplyDelete

The comment system is unreliable. If you cannot submit comments you can email me a comment (via email link in profile) - email must be entitled: Comment for publication, followed by the name of the article you wish to comment on. Comments are accepted in full or not at all. Comments will be published under the name the email is sent from unless an alias is clearly requested and stated. If you submit a comment which is not accepted within a few days you can also email me and I will check if it has been received.