The publicly available opinion polls are in agreement about the fate of the Labor government. A landslide defeat in the House of Representatives is looming, the magnitude of which might place Julia Gillard alongside Bert Evatt, Arthur Calwell, Gough Whitlam (twice) and Paul Keating as federal Labor’s biggest electoral losers.
Little, however, has been said about the Senate contest. Opposition leader Tony Abbott clearly foresees an outcome that might frustrate his future government and has promised a double dissolution election if this should happen.
The current polling data indicates that the disaster Labor is facing in the lower house will also impact on the Senate, notwithstanding the different electoral systems being used. What’s more, Labor’s poor performance has the potential to seriously affect the Greens as well.
Forecasting Senate outcomes can be difficult given the complexity of the voting system, the potential for a vast number of candidates to nominate, and the lack of information about how preferences will be directed by the various parties. It is also difficult to find opinion poll data broken down by states. Using the polling data below, some preliminary predictions about the Senate can be made.
The table above uses the same polling data used by Fairfax last week to make claims about the state variations that might occur within the national swing as voters make their choices for the House of Representatives. In other words, these polls did not ask voters about their Senate voting intentions.
Psephologist Malcolm Mackerras’ argument that, in normal circumstances, half-Senate election results usually end up seeing the collective left-of-centre (Labor, Greens) and the collective right-of-centre (Liberal, National, DLP, Family First and so on) sharing three seats each is a useful starting point. As the LNP Coalition rarely has serious competition from minor right-of-centre parties, the joint ticket usually wins three seats in each state.
Labor, however, has lost ground to its competitive minor parties: the Australian Democrats for much of the 1980s and 1990s and more recently to the Greens. Therefore, a contemporary “normal” pattern involves Labor winning at least two seats in each state, with Labor and the Greens fighting for the third left-of-centre seat.
As part of this process, the successful left-of-centre minor party has had to rely on a flow of surplus Labor votes to secure a seat. The importance of Labor’s preference direction was demonstrated in 2004 when Family First’s Steve Fielding won a Senate seat for Victoria owing to being preferenced ahead of the Greens on the Victorian Labor Senate ticket.
For Labor to be in a position to assist the Greens - or whoever it prefers - to win a seat, its Senate ticket must poll more than at least 28.8% of the primary vote.
To assist in making predictions, the data in Table 1 is calculated in Table 2 as a quota, with the parties winning a seat for every full quota achieved. The table also expresses the quota as an aggregated left-of-centre compared with right-of-centre result which can assist in indicating how the numbers might fall in terms of a left-right contest.
The first important thing to note from this table is that in three states, the combined left-of-centre does not have enough support for three seats, whereas the right-of-centre has more than enough for four. If nothing else, these data are indicating a significant shift to the right-of-centre in terms of outcome of seats, although it is not clear which of the parties will win these seats.
Both Table 1 and Table 2 show that in three states Labor is polling under the 28.8% threshold needed to secure two seats, let alone provide some surplus to help the Greens.
It should be borne in mind that the above data may be worse than it appears. If the past is any guide, Labor’s primary Senate vote will be less than its national lower house vote (the average difference since 1949 is negative 3%), while the Greens can expect to poll better in the Senate than the House of Representatives (a difference of plus 1%). This won’t help the Greens, for in all likelihood Labor will need Green preferences to secure a second seat in Queensland and NSW.
The situation in South Australia is complicated by the fact that independent senator Nick Xenophon is up for re-election. If Xenophon were to poll in excess of the quota and had surplus to distribute, the Greens’ Sarah Hanson-Young might be returned. Were Xenophon to poll less than 14.4%, however, Hanson-Young would be stranded and her preferences would elect the second Labor candidate.
In Queensland the presence of populist parties such as the Katter Australia Party (KAP) and the Palmer United Party (PUP) are the wildcards. If the primary vote of Labor and the Greens is added together, the collective left-of-centre only has enough support to win 2.7 quotas. Meanwhile, the aggregated LNP and other right-of-centre (KAP, PUP, and all others) results in 4.19 quotas. Queensland will thus return two left (most likely Labor) and four right (most likely three LNP and probably one of the populist right, most likely from the KAP) to the Senate.
The situation for Labor and the Greens is better in Victoria, where Labor will win two seats and the Greens one. No data has been given for Tasmania, where the Greens would nonetheless be confident of picking up one seat, and Labor two seats. The Liberals would have expectations of three seats, and this would be a gain: in 2007, the Liberals won only two seats.
Table 3 sets out tentative predictions for the Senate based on the available polling data. The predicted outcome in Tasmania is purely speculative, however, given the lack of specific polling data.
The overarching theme of this table is that the Senate will shift to the right, and that Tony Abbott could be within a whisker of winning control of the Senate. These figures and the speculation about the outcome in Tasmania would result in the three “others” – existing DLP senator John Madigan, Nick Xenophon (predicted to be re-elected) and a right-of-centre senator from Queensland, possibly from the KAP – holding the balance of power.
The prediction of the Greens’ success in Victoria and Tasmania can be made with confidence, but the situation in Western Australia is critical to everything. According to the Fairfax polls, Labor should win two and the Greens one seat. The result will be very close, however, and the possibility of the Liberal-National joint ticket picking up a seat at the expense of the Greens is a real possibility.
If that happens, the Liberal-National government will take control of the Senate on July 1, 2014, and Tony Abbott will not have to worry about a double dissolution election.