We now know that the Senate elected at the July election comprises 30 Coalition members, 26 from the ALP, 9 Greens, 4 from One Nation, 3 from the Nick Xenophon Team (NXT) and one each from Family First, the Liberal Democrats, Derryn Hinch’s Justice Party and the Jacqui Lambie Network. Thirty-nine votes are required in the Senate to pass legislation, so the government will require either ALP or Green support, or otherwise support from nine of the eleven minor party members. Given that NXT has three Senators and One Nation has four, their support for every Bill opposed by the ALP and the Greens becomes essential. It will be a difficult situation for the government but there is another issue I wish to discuss.
Before the election, the Coalition and the Greens combined to introduce a new voting system for the Senate, the aim being to reduce the number of minor parties or people being elected despite starting with only a handful of first preference votes. It did not work this time, largely due to the election being a double dissolution, but will it achieve its aim in the future when we resume the cycle of half-Senate elections?
Senators are elected for six years but on a rotational basis so that half the Senate faces the electorate every three years. After a double dissolution a decision has to be made as to which of the newly elected Senators will serve a full six-year term and which will serve only three before facing an election. There are two ways of doing this.
The common approach has been to use the order in which Senators from each state are elected and give the first six the six-year term, with those elected from seventh to twelfth to serve three years. In a Senate election those with smaller votes are progressively ‘excluded’ and their preferences distributed and preferences from those who have an ‘excess’ quota are also distributed on a proportional basis: as that process unfolds a clear order of election emerges.
The other way, put in legislation by the Hawke government in 1984, allows the Australian Electoral Commission (AEC) to do a recount of the election as if it was a half-Senate election and those ‘elected’ under that count would be given the six-year term. The underlying idea supporting that approach was that using the ‘first six elected’ may have been fair when we had ‘first past the post’ voting but not when we have a proportional representation system: despite that, even after the 1987 double dissolution this method was not used
Although those two methods are available, the Constitution states only that the Senate itself can decide who will serve six years and who three.
Derryn Hinch had suggested that if he did not get a six-year term he would challenge the decision
in the High Court. I am not a lawyer but I don’t like his chances if he carries out that threat. When the Constitution states the Senate can decide, it would seem that it is not even bound to use either of the methods I have described.
It was reported on 12 August that the Coalition and Labor had agreed
they would use the ‘traditional’ method of the ‘first six elected’ to determine the six-year Senators. That means the Coalition will have 16 six-year Senators, the ALP 13, the Greens 3 (Di Natale, Ludlam and Whish-Wilson), NXT 2 (Xenophon and Griff) and One Nation (Pauline Hanson) and Jacqui Lambie Network (Jacqui Lambie) one each. So that will be the starting point for a new Senate from 1 July 2019.
At a state level that translates as 3 each from the Coalition and the ALP in NSW; in Queensland, 3 Coalition, 2 ALP and Pauline Hanson; in SA, 2 each from the Coalition, ALP and NXT; in Tasmania, 2 each from the Coalition and ALP, and 1 Green and Jacqui Lambie; and in both Victoria and WA, 3 Coalition, 2 ALP and 1 Green.
I will basically ignore the Senators from the NT and ACT, except in discussing total numbers, because they face election every three years and the territories invariably return one Coalition and one ALP Senator, meaning that we just add two to each major party.
What becomes more important is who will face re-election in three years: 12 Coalition Senators, 11 from the ALP, 6 Greens, 3 One Nation, 1 NXT and Derryn Hinch (Derryn Hinch’s Justice Party), Bob Day (Family First) and David Leyonhjelm (Liberal Democrats). From that list you can see why the Coalition and the ALP agreed to the ‘first six elected’ method: their numbers for re-election are close and 13 of the 20 cross-benchers have to face re-election. In that regard, the government may be relatively content with the outcome from the new voting procedure and see the next half-Senate election as an opportunity to reduce the size of the cross-bench.
To be elected a Senator has to achieve a ‘quota’ which is determined by dividing the total number of votes in a state by the number of Senators facing election plus one. In the recent double dissolution that meant the vote was divided by 13, or in percentage terms a quota was about 7.69% of the vote. In a half-Senate election, the number is divided by 7 and so a quota becomes 14.29% of the vote. While preferences are important in determining who is elected, the quota achieved after first preference votes gives a reasonable indication of who will be elected, with those achieving at least 0.4 of a quota having a higher probability of achieving a full quota after preferences.
Given the results in the 2016 election who is likely to win re-election at a half-Senate election and what will that mean for the Senate from 2019?
Firstly, neither David Leyonhjelm, NSW, (Liberal Democrats) nor Bob Day, SA, (Family First), appear likely to be re-elected: their 2016 vote becomes only 0.2 of a quota at a half-Senate election. They would each require something close to a doubling of their vote and that is highly unlikely.
In Victoria, Derryn Hinch may be in with a chance of being re-elected if he can maintain his vote: his half-Senate quota would start at 0.42 rather than the 0.79 at the 2016 election. My guess is that at a half-Senate election Hinch, if he runs again, may even achieve a slightly higher vote, but even with a starting point of 0.4 of a quota I expect that he could be returned.
Jacqui Lambie was elected in Tasmania at the recent election but will not face re-election at the half-Senate election. The 2016 vote translates to 0.58 of a quota at a half-Senate election but because that vote was for Jacqui Lambie herself, it is unlikely to be repeated when she is not running. So even if she runs a Jacqui Lambie Network candidate, I would expect a very reduced vote and it is unlikely a second Network member would join her in the Senate.
Three One Nation Senators will face re-election with the NSW and WA Senators appearing unlikely to win as their starting quota would drop below 0.3. A Queensland win is possible based on the 2016 vote as it would become 0.64 of a quota at the half-Senate election. Hanson herself will not be running so that could reduce the One Nation vote but it may still be enough to secure a second One Nation member in the new Senate (unless there is a large drop in the vote, which is possible based on the past history of One Nation).
Based on the 2016 vote, NXT could pick up two more Senators in SA at a half-Senate election as it would have 1.52 quotas. As with Lambie and Hanson, however, Xenophon himself won’t be running and that may reduce NXT’s vote. Even so, it seems likely that at least one NXT member will be returned and two can’t be entirely ruled out. So NXT will maintain at least three Senators, and possibly increase that to four, in the new Senate from 2019.
One Green Senator from every state will face re-election. They are likely to have at least four and probably five returned. SA is most problematic for them largely because of the magnitude of the NXT vote. If the NXT vote drops, it may become a battle between the Greens and a second NXT candidate for the final Senate position. So that will not change the number on the cross-bench, just the composition.
So, on my estimation, it is possible that 9 of the 13 cross-benchers facing a half-Senate election could be returned, meaning there would still be 16 rather than the current 20. Counting the six-year Senators, the new cross-bench could comprise 8 (possibly 9) Greens, 4 (possibly 3) NXT, 2 One Nation, Jacqui Lambie and Derryn Hinch.
How the Coalition and the ALP fare at a half-Senate election depends very much on who wins government in the HoR because, obviously, the winning party would probably see an increased vote compared to the 2016 election. An increased vote for either party may also have some influence on the results for the minor parties, with Hinch in Victoria, the One Nation candidate in Queensland and possibly the second NXT candidate in SA being most at risk.
At best, the Coalition could win three seats in each state, although if NXT continues its success in SA it may only be two there. Even with an increased vote, and irrespective of the NXT vote, I suggest that 17 Senators is its very best
outcome, for a total of 35 Senators (including the two Territory Senators) in the Senate from 2019, still four short of a majority.
The ALP’s best result, with an increased vote, appears to be 16 Senators for a new total of 31 (also including its two Territory Senators) so, as in the past, it would be reliant on the support of the Greens to pass legislation.
Those ‘best’ results for the Coalition and the ALP include the scenario that Hinch, One Nation and NXT do not win the extra seats I mentioned. If they do win, then the Coalition could expect to win 14 seats and the ALP 13, in which case the 2019 Senate would be 32 Coalition Senators, 28 from the ALP, 8 Greens, 4 NXT, 2 One Nation and Derryn Hinch and Jacqui Lambie.
In that scenario the Coalition would require ALP or Green support, or all but one of the minor party Senators to pass legislation, and the ALP would require the Greens and at least three of the other Senators, which most likely would mean gaining the support of NXT.
There are of course many permutations. Will there be a stronger vote for the Coalition or the ALP in 2019 and, if so, will it be strong enough to reduce the vote for minor parties? Will there be a resurgence in the Greens’ vote? Will the One Nation vote collapse as it has done in the past? Can NXT maintain the very high vote it achieved in SA in 2016? The answer to those questions can change what happens at the next half-Senate election and perhaps re-write my scenarios.
But unless the answer to those questions is ‘yes’, ‘no’, ‘yes’ and ‘no’, then it appears likely that the government’s Senate voting changes will not achieve its intention of significantly reducing the size of the Senate cross bench. It may have to wait until 2022, even 2025, and try, try again.