Note: The Australia Poll Average and Seat Projection has been updated this morning after the Galaxy poll. Click on the Australia Tab above for details.
The landslide victory of Liberal National Party in the 2012 state election obscured the arrival of a two and a half party system in Queensland politics. The solid showing of the Katter’s Australian Party heralded the possibility that when LNP support returned to normal levels it would be a two front battle to re-elect its MPs.
The downward trajectory of the support for the LNP in opinion polls over the second half of 2012 and the defection of Ray Hopper to the KAP has now sharpened the public attention to two and half party politics in Queensland. With the former LNP life member and mining billionaire Clive Palmer threatening to form a new party, the spectre of three or possibly four party politics state-wide now looms.
With the parties that are capable of winning seats in Queensland now beyond the two majors, in this post we will assess the new electoral calculus that is emerging in Queensland politics.
South East vs the Rest
In the below table we take the last election and split the results between South East Queensland (SEQ) and the rest of Queensland (RoQ). The definitions for these areas are based on the geographic seat groupings Antony Green uses in his election analysis publications for Queensland elections after the previous redistribution in 2008.
As you can see if we split the state into two separate elections we get two different contests. In SEQ we have a two party system where one party has won a crushing landslide. In RoQ, the LNP still is the dominant party but faces two oppositions with almost equal support.
This geographical divide gives us the “two and half” party system with a two party system in one part of the state and a three party system in the other.
With the possibility of a new well funded party backed by Mr Palmer, this dynamic may change again depending on the nature of the new party and any relationship it may have with the KAP. We now turn to the two broad areas of Queensland and look more closely at the differing electoral contests in each.
Rest of Queensland (RoQ) & Katter’s Australian Party (KAP)
With the rise of the KAP, the electoral contest in RoQ has now become a three way contest. To map out this 3-way contest, we will visualise it with a scatter plot.
In the below chart we have taken every seat currently held by an LNP member in RoQ except Burnett and Maryborough which had independent incumbents which makes these seats difficult to compare on the below chart. Please click to enlarge.
We have plotted each seat with regards to the difference in primary votes between the LNP and KAP on the horizontal axis and the difference in primary votes between the LNP and Labor on the vertical axis.
The graph gives us an ability to look to each seat in two dimensions, in terms of each LNP seat’s vulnerability to capture by Labor and the KAP. As we need to look in two dimensions we need to compare primary votes not the one dimensional TCP count. This acceptable for our purposes as under OPV; the primary vote leader has a mortgage on a victory. However, we should keep in mind that a primary vote lead of 5% of less can be overturned on preferences.
As you can see of the 23 seats in the above graph we have split the seats into six groups based on their vulnerability to either or both of Labor or the KAP with some consideration to geography.
The transition of moving from a two party system to a two and half party system depends on the third party’s ability to win seats and force a hung parliament. The more seats won by a third party and the major party that is shut out of seats won by the third party are factors in determining the dynamics of two and a half party system.
For example if the KAP can only win significant number of natural conservative seats then the LNP ability to form a government in its own right will be restricted to solid to landslide election victories for the major centre-right party.
However if the KAP can also win a significant number of hitherto Labor seats in RoQ, most of which were won by the LNP in 2012, then both major parties’ ability to form government will be impaired and will require a landslide victory by a major in the SEQ for majority government to be achieved.
To address the effect on the LNP or Labor from forming majority government we have coloured the data points by the party that won the seat in 2009 so as to indicate which party is effect by each respective KAP gain. We will now turn to these groups as labelled on the charts.
Group A – 3 seats
These seats are the closest to three-way marginals in Queensland. Labor held- Mulgrave can also be considered in this group. These seats are all North Queensland seats
If the KAP can take these seats this is the next step to becoming a long term broad based third party in RoQ. It will also impair Labor’s recovery efforts as the three seats in this group were all held by Labor at the 2009 election.
Group B – 4 seats
These rural seats all had the KAP finish in the TCP count in 2012 and are all rural by nature. They would be the first seats to fall if there was a large rural swing away from the LNP to the KAP and consequently they are in the front line of any LNP defence against the KAP.
Group C – 3 seats
Similar to the Group B seats, these seats are a degree more safer for the LNP. If a KAP rural rout was on they would fall.
Group D – 6 seats
These seats as a group are based on regional centres or surround them. All bar one seat was held by Labor at the 2009 election. These LNP held seats are slightly more vulnerable to Labor, however it not out of the realms of possibility that they can be won by the KAP. If the KAP can snatch these seats than any future Labor representation outside the SEQ would be sparse.
Group E – 4 Seats
These seats are all in provincial regional centres and recorded relatively low KAP votes in 2012. All bar Toowoomba North were held by the LNP in 2009 and thus are safe for the governing party from either challenger.
Group F – 3 Seats
These rural seats are hyper-safe for the LNP and probably can withstand a rural rout by the KAP.
A best case scenario for the KAP is that they win:
- All LNP seats in Groups A-D above (16 seats);
- Labor held Mulgrave;
- hold the three seats current held by KAP MPs; plus
- Maryborough and Burnett.
These 22 seats can be considered the present KAP best case scenario, given that the party has limited appeal in the SEQ and could not win any seats there in its present format. Twenty-two seats is more than enough to force a hung parliament on a regular basis but is just shy of minimum 23 seats needed to have any chance at forming a minority government.
Of course this all requires a lift in the KAP vote from its present position in the polls of 7-9%. The fact that the KAP can hold this support in the polls given the absence of an election campaign bodes well for the minor party. However to win all of the broader range of seats indentified above (Groups A-D), the KAP would need a state-wide vote of at least 17-18%. Such support for the KAP has not yet been registered in opinion polls before or after the election. The above analysis is designed to map out the possible for the KAP not the probable.
If this best case scenario of 22 KAP seats (Groups A-D) occurs then it becomes difficult for both parties to win majority government. Labor would need to win 43 of the 57 SEQ seats to form majority government, more seats than the ALP won at the 2009 election in SEQ.
The LNP would need 38 of the 57 SEQ seats to form a majority government. Taking into account the recent defections of Carl Judge and Alex Douglas, the LNP would have to lose 13 seats in SEQ to fall below this threshold. Such a loss would occur on a uniform TPP swing of 5.6% to the ALP in SEQ.
If Labor wins the provincial seats Group D as a group the KAP universe is reduced to 16 seats and Labor would only need to win 37 of the 57 SEQ seats to form a majority government only one seat less than won by the party in SEQ in 2009.
If KAP gains are limited to the more marginal A & B Groups (10 seats). The SEQ seat requirements for either major party to win majority government would vary depending on how many seats Labor won back off the LNP in RoQ. The ranges would be 27-33 seats for the LNP, 36-42 seats for the ALP.
South East Queensland (SEQ)
In SEQ the electoral contest for now is still a two party affair. A table of all SEQ seats ordered by geography, party status and LNP TCP margin is set out below.
In all likelihood the SEQ electoral battle ground in 2015 will be the LNP seats held on margins of up to 10% which were formerly Labor safe seats.
However a new possibility has risen on the political horizon in the form of a new party to be formed at the instigation of the mining billionaire and former LNP life member Clive Palmer. The mining billionaire who was also a lifelong political conservative political activist who has fallen out with the Queensland LNP and is now a bitter opponent of premier Campbell Newman. Mr Palmer is now threatening to set up his own party in Queensland and has suggested that it will be known as the “United Australia Party” (UAP).
There are further suggestions the Mr Palmer’s new party will seek a relationship with the KAP. This is just as well as Palmer’s renegade conservative party would share at least some natural supporters with the KAP. If the UAP chose to run a full state wide slate of candidates then they would split the vote in regional seats with the KAP.
The relationship between the UAP and the KAP could range from a coalition to an outright merger. Even if a coalition cannot be agreed between the parties it would in their interests to form a looser arrangement such as an alliance or an electoral pact to avoid contesting the same seats. Such an arrangement for optimum electoral advantage preserve the KAP in RoQ where it performed best and leave SEQ for the UAP.
So how would the infant party do in SEQ? Obviously it is too early to tell with certainty as there are many factors in play. However a few factors deserve some exploration.
Leadership, Ideology and Appeal
It is not certain that Mr Palmer will even lead the new party as he has made some equivocal statements in this regard. While his approval ratings are tepid (25.6% approval vs 33.7% disapproval- Nov ReachTEL), there is no other public figure associated with the party at this time which has any public profile.
A Galaxy poll released early last week indicated that 27% of voters would vote for a Palmer Party. If this was a simple yes/no question instead of a full suite of party voting options, then it does not carry much weight.
Will the new party have any ideological differences with the LNP? At this point the purpose of the party seems to be to oppose the Newman Government. However we can only assume that it will be right of centre.
LNP “oncer” Lifeboat?
Another possibility is that the UAP may attract current LNP MPs in SEQ which hold seats in 0-10% battleground noted above. A common term for these MPs who hold seats that are deep within Labor territory are “oncers”, reflecting to assessment that they cannot win hitherto safe Labor seats in a regular electoral contest.
A jump to the UAP by any of these MPs would allow these MPs to campaign against the Newman Government in their electorates and give the MP’s the possibility of holding on to their seats. However with the exception of the recently unbound MP Carl Judge no such MPs have indicated any desire to throw their lot in with the UAP. However impending political demise can be a powerful incentive.
If this occurred, the likely scenario is that the LNP would put up a candidate but not seriously contest such seat so as to save resources for the seats needed to retain government. However the rump LNP vote in these seats would in all likelihood prevent the re-election of UAP MPs unless an explicate preference recommendation was made.
Sunshine and Gold Coasts
Another possible base of support for the UAP is the Sunshine and Gold Coasts. Mr Palmer is a resident of the Gold Coast and owns the Coolum resort on the Sunshine Coast. Mr Palmer’s links to the community are the strongest in these areas and combined with the fact that all of the Sunshine And Gold Coast seats are safe LNP, none of these seats will be targeted by the ALP In 2015 when they will seek to win back seats on lower margins in Greater Brisbane. The UAP could fill this vacuum in these areas.
The electoral contest in 2015 is still too far away to say with certainty about its outcome. Minor parties do fizzle out and their electoral threat may disappear by 2015.
But given the huge landslide win of the LNP this year, the decimation of Labor and the rise of new parties, the electoral contest 2015 could be a patchwork quilt of novel contests if the KAP and the UAP do not dissipate as political forces in the meantime as minor parties commonly do.
In at least half of RoQ seats the contest could either be a three way match between the majors and the KAP or a LNP/KAP duel.
In SEQ former Labor heartland seats snatched by the LNP in 2012 could see contests between sitting MPs under the UAP umbrella competing against Labor, while on the Sunshine and Gold Coasts the LNP may have to fend off a well resourced UAP challenge.
However the end game for the KAP and for any UAP is to win enough seats to force a hung parliament. From the current position the KAP cannot win enough seats on its own in RoQ to ensure a hung parliament. Either Labor needs to win back some its former heartland seats and/or the UAP needs to win seats.
Given current polling Labor look set to win back a score of its former seats. If the LNP wants to avoid a hung parliament then it needs to hold its second front against the KAP and possibly (if Clive Palmer is so inclined) a third front against the UAP.
However with 2 years to go until the next election, this is plenty of time for Queensland’s electoral calculus to change again.