My previous post dealt with the report of the Joint Standing Committee on Electoral Matters’ inquiry into representation of the territories, which recommended the Northern Territory be crudely guaranteed a second House of Representatives seat while removing the more sophisticated statistical fiddle that helped preserve it when the issue last arose in 2003. As Antony Green noted, this proposal raised the strong possibility that the Australian Capital Territory might lose its recently acquired third seat the next time the determinations are made during the next parliamentary term. However, the federal government has sprung into action with new legislation that promises to preserve both territories’ seats by following Antony’s advice rather than the committee’s.
This is to be done by having the territories’ seat entitlements calculated through the harmonic rather than arithmetic mean, at least so far as their first three seats are concerned (beyond which the issue is likely to remain academic). The principle behind the harmonic mean can best be explained by using a simplified version of the Northern Territory case as an example. The basic problem is that the territory has around 150,000 voters, whereas the average House of Representatives seat has around 100,000 (population rather than voter enrolment is actually used, but the near accuracy of these nicely round figures means I will continue with them for purposes of illustration). Using the conventional arithmetic mean, this places the territory right at the cut-off point between a one-seat and two-seat entitlement. Two seats prevailed when the local economy had the wind in its sails during the late mining and resources boom, but in the more straitened circumstances of the present it only makes it to one.
Using the harmonic mean, the point at which rounding occurs is based not on the mid-point between the two quotas, but the point at which electorates’ populations differ least from the national average. Were the Northern Territory to lose its second seat, the remaining seat with its enrolment of around 150,000 would have 50,000 voters more than the national average. But if its second seat is retained, the two would have around 75,000 each, differing from the national average by only 25,000. The harmonic mean is all about minimising this difference, which in the present example would mean only one-and-a-third quotas would be needed for a second seat, or around 133,333 voters. For the Australian Capital Territory, which similarly stands on the precipice of two quotas and three, the third seat would be retained with 2.4 quotas (240,000 voters in the present example) rather than 2.5. The differences between the arithmetic and harmonic mean tipping points continue to reduce with each additional seat. By Antony Green’s reckoning, the ACT would have fallen below the arithmetic mean benchmark at 2.4796 quotas without the aforesaid statistical fiddle, which the committee had proposed to abolish without the remedial action of using the harmonic mean.
It is perhaps not surprising that the federal government has determined to save the second Northern Territory seat, notwithstanding that both seats are held by Labor: both are winnable for the Country Liberal Party, particularly the Darwin-based seat of Solomon, and an overstuffed single electorate for the Northern Territory would essentially amount to an act of malapportionment to the disadvantage of the territory’s substantial indigenous population. However, there is no such impetus in the Australian Capital Territory, where the Liberals only win House of Representatives seats under extraordinary circumstances (the most recent being the Canberra by-election of 1995), and the removal of a seat could be rationalised, if not justified, with recourse to public service bashing. At a time when mainstream conservatism in the United States is taking to the foundations of democracy with an axe, our own government’s defiance of self-interest to preserve Labor-held seats is worth acknowledging and celebrating.
Elsewhere: in the only bit of polling news to relate right now, JWS Research has released its latest True Issues survey of issue salience, as it does around three times a year. When respondents were asked to nominate the country’s three most important issues without prompting, 42% offered a response within the “hospitals, health care and ageing” category, which is down five from July but well up on the 24% recorded in the pre-COVID days of February. Results are otherwise very similar to the July survey, with economy and finances steady in second place at 32% after shooting up from 18% in February. A plunge in concern for the economy and climate change, down from 26% to 16% last time, has only slightly corrected to 19%, remaining well behind third-placed employment and wages on 32%, up two from July and eleven from February. The poll was conducted online between November 20 and 22 from a sample of 1035.