Against Provincialism

The bourgeoisie has subjected the country to the rule of the towns. It has created enormous cities, has greatly increased the urban population as compared with the rural, and has thus rescued a considerable part of the population from the idiocy of rural life.

—Karl Marx and Friedrich Engels[1]

I think that Mr de Tocqueville had a lot to say about it in his book on democracy in the United States when he talked about the `tyranny of the majority'—Australian government decision making is done by the big cities to suit the big cities.

—The Hon. Robert Katter Jnr., MP[2]

When the first census was conducted in 1911, approximately sixty per cent of Australians lived in the regions and forty per cent in cities[3]; today the inverse is true[4]. The rise and fall of regional Australia was the result of Gold Rush overexpansion which receded as the cities began to produce economies of scale. As the historian W.K. Hancock wrote,

There is no particular reason why the Australians should reproach themselves for the emptiness of these vast areas. They have gained very little in the past from their premature and expensive attempts to make an imposing show of settlement. It is sound policy to take the best country first.[5]

Those who inhabit the land of the second best typically lament the urban sprawl. A common argument is that people in regional Australia should be afforded special political rights to redress the perceived subversion of their interests by a metropolitan élite. The most obvious mechanism for achieving this is the Australian Senate, which to an extent already acts as a safeguard against elective dictatorship[6] and, since the introduction of proportional representation in 1948[7], has consistently produced more diverse party representation than the lower house[8]. The aim would be to restate this dual function of the Senate in terms of polis rather than party.

Representation is a difficult concept to define[9], and often entails objectives that are mutually exclusive[10]. One such tradeoff appears to be between party and regional representation. The Senate is in effect a parties’ house[11]; most of its Senators serve on the criterion of loyalty to a highly centralised and urban apparat[12]. An elector generally cannot vote for a senator who will exclusively represent their region—like, say, a Member of Parliament represents their constituency—because in order to win, senators and their parties need to garner support from across the state. Making the Senate like the House of Representatives only with bigger constituencies would relax this requirement, as well as give senators a more decentralised mandate. In 1999, Labor-turned-Independent Senator for Queensland Malcolm Colston introduced a bill which would have divided each state into six wards[13], similar to the pre-2003 Victorian Legislative Council. The Colston proposal is the only one to reach the floor of Parliament, and even then sans debate. Privately governments have considered similar proposals, though more out of a partisan desire to purge a recalcitrant crossbench than to better represent the country-minded[14].

State politicians have more often invoked the folklore of the frontier when justifying the practice of malapportionment—or gerrymandering say their adversaries—in South Australia and Queensland[15]. The Western Australian upper house, in spite of a High Court challenge[16] and some limited reform, remains heavily malapportioned in favour of the Wheatbelt region and seats in the Northern Territory Legislative Assembly have noticeably fewer constituents when a high proportion of them live in remote Indigenous communities. A latent anti-urban or pro-rural bias might also explain why the Australian Capital Territory, Australia’s most densely populated state/territory, has as many senators as the Northern Territory with twice as few electors[17].

The Constitution creates malapportionment to an extent by requiring each original state to have a minimum of five seats in the House of Representatives[18] and equal representation in the Senate[19]. The result is that, at the last Senate election, a Tasmanian’s vote was worth more than thirteen times that of a New South Welshman’s[20]. Section 7 of the Constitution theoretically allows for these disparities to be replicated on an intrastate level.

The Senate shall be composed of senators for each State, directly chosen by the people of the State, voting, until the Parliament otherwise provides, as one electorate.[21]

The ethos of section 7 managed to survive the many challenges to it at the Constitutional Convention. Some delegates wanted states explicitly divided into electorates[22]. Some didn’t support malapportionment but concurred with all five colonial governments’ recommendation that it should be left to the states[23]. Others believed doing so would negate the federation principle[24]. The result is a clause designed by committee. The method for electing senators did not fundamentally change from the 1891 draft bill[25], but the addition of “until the Parliament otherwise provides” by the Framers at the Sydney session in 1897 leaves open the possibility of malapportionment without themselves having to commit to it[26]. One wonders then why there is a further provision for malapportionment specific to Queensland.

But until the Parliament of the Commonwealth otherwise provides, the Parliament of the State of Queensland, if that State be an Original State, may make laws dividing the State into divisions and determining the number of senators to be chosen for each division, and in the absence of such provision the State shall be one electorate.

This work of supererogation was added not at the convention but subsequently by the Premiers at their so-called ‘secret meeting’[27]. It was a consolation to Queensland Premier James Dickson[28], who faced resistance from Central and North Queensland separatist movements and almost lost government over passing the legislation to enable federation[29]. If New South Wales Premier George Reid was more provincial minded perhaps he would have accounted for a potential state of New England as well. But the topic of regional representation was and is essentially an argument over Queensland—the only state whose topography could realistically allow decentralisation.

The Commonwealth left the specifics of voting at federal elections to the states until the passage of the Commonwealth Electoral Act 1918[30]. Electoral laws are now a prerogative of the federal government with very few limits. The High Court has so far given the Parliament relative carte blanche in determining electoral laws, a judgement upheld mutatis mutandis in 2016 when then Senator Robert Day challenged the Turnbull Government’s Senate voting reforms[31]. While blatant malapportionment has only been practised by state governments, there is little constitutionally preventing a federal government from changing the electoral laws in the future to favour one region over another.

The Australian Constitution, in spite of several recommendations[32] and two referenda attempts[33], does not mandate ‘one-vote, one-value’. An amendment to the Commonwealth Electoral Act 1918[34] could conceivably make the value of a Senate elector’s vote weighted according to their level of remoteness, a metric easily knowable and already in use by several health and welfare agencies. Other methods would be less conspiciously undemocratic but will in effect produce the same outcome: an individual voting in a region will have twice as much franchise as someone voting in a city. While vote weighting would be the most efficient method and circumvent messy partitions, it may contravene the Commonwealth’s external affairs obligation to guarantee “universal and equal suffrage” (emphasis mine) as per Article 25 of the International Covenant on Civil and Political Rights[35].

There are two ways of altering regional representation through section 7: divide states up into regions or deliberately weigh votes unequally. However this can only be asserted prima facie because malapportionment of this scale would be without precedent and likely attract a constitutional challenge. In the recent case relating to political donations, McCloy v New South Wales[36], a majority of judges found an implied right of “equality of opportunity to participate in the exercise of political sovereignty”[37]. Applying the McCloy principle here promiscuously, deliberate vote weighting would seem to be unconstitutional. The implications for regional malapportionment, an indirect form of vote weighting, are less clear. The High Court upheld malapportionment in both explicit challenges to it, McKinlay v Commonwealth[38] in 1975 and McGinty v Western Australia[39] in 1996. Though the reasoning behind each judgement was different, both focused on a perhaps false antithesis between the English tradition, with its rotten boroughs, and the American tradition with its highly contested redistricting commissions[40].

The High Court upheld parliamentary sovereignty in McKinlay, though four judges acknowledged obiter that the Court would theoretically have to intervene if malapportionment ever led to a conspicuously undemocratic or partisan outcome, i.e. gerrymandering[41]. The dissenting Murphy J, echoing the United States Supreme Court, said that leaving such matters to the parliament was a self-defeating tautology.

The argument was advanced that no guarantee of electoral equality should be read into the Constitution because it was left exclusively to the Parliament to do what was fair. If Parliament failed, the remedy was with the people. The hollowness of this argument has been made apparent by the history of maldistribution in the United States and in Australia. If the legislature fails to ensure fairness, and there is no constitutional right enforceable in the courts, where is the remedy?[42]

Though still upholding McKinlay, the High Court moved closer to this American view twenty years later in McGinty. It is even possible that, had the activist Mason CJ still been serving on the bench, the Court would have ruled an implied right to one-vote, one-value[43]. In 2007, the High Court repudiated parliamentary sovereignty as grounds for denying franchise in Roach v Electoral Commissioner[44]. The validity of any change is therefore predicated on the type of judicial philosophy favoured by the High Court at the time. Speaking on the Colston bill in 1999, Malcolm Mackerras alleged that there was a “50-50” split among his friends at the bar as to its constitutionality[45]. Certainly a Dixonian black-letter reading of section 7 would imply that malapportionment is consistent with the Constitution.

Malapportionment already exists in federal elections. The question is whether further malapportionment is constitutionally desirable and, most saliently, consistent with the doctrine of federalism. Federalism is an old Calvinist idea about protecting religious minorities from domineering states[46]. One wonders what comparable identity could be forged out in the bush—a hotchpotch of farmers, graziers, townfolk, miners, hippies and those who were there ab origine, or “from the beginning”. The sentiment is also somewhat condescending in assuming that these people vote based on locality rather than interests like normal people[47].

There is a further question of whether regional Australians actually have different political interests or if it is merely a recrudescence of what Freud called ‘the narcissism of small difference’ with a touch of self-pity. The existing system of statewide representation at least prevents the Senate from degenerating into such facile politics of identity, and the chauvinism and demagoguery that comes from it. Admittedly this is an endorsement faute de mieux. My own view is that, since all the essential political debates are either national or international, voters should expect and elect senators to serve at-large. Regional malapportionment would be a further deviation from this ideal, as well as deprive the Senate of much of its raison d'être. Samuel Griffith outlined a similar criticism in his Notes on Australian Federation.

Election by separate constituencies is not consistent with the notion of representation of the State as a whole. […] The purpose for which a Second Chamber is established in a Federation is not to echo the voice of the First. It is an independent representative body with equal powers. Those who would make it a mere echo are open to the charge of either not understanding or not desiring a real Federation in which the rights of the individual States will be conserved. [48]

Regional malapportionment would not ipso facto render the upper house a vestigial ‘Senate of Representatives’, as Oily Sam seems to suggest here. The hope is that the change would provoke fission within the major parties, leading to splinter groups like the North Queensland Labor Party that once existed at a state level[49]. This would be a necessary condition for achieving greater regional diversity given that, with each region only electing two or four or six senators, the quotas would be much higher; it would likely result in a chamber comprised entirely of major party senators barring a few ‘favourite son’ independents. But if parties did not reform their appointment processes accordingly then it would not improve regional representation at all and, by making it easier for governments to win majorities, could undermine the Senate as a house of review.

Granted, the sheer size of Queensland and Western Australia, and the remoteness of Tasmania and the Northern Territory, makes it practically impossible for many senators to represent their entire state. An argument for malapportionment at the Constitutional Convention was the supposition that only magnates and barons could afford to campaign across an entire state[50]. Today the party system handles this, although imperfectly, by having senators and their offices based in particular regions. This leads to a degree of unfairness: at federal elections National Party senators campaign across a ‘wombat trail’ spanning thousands of kilometres while their Coalition partners stay mostly confined to the capital cities. This is a variation of Geoffrey Blainey’s ‘tyranny of distance’, and will persist no matter the proportion of Senate candidates electioneering in the regions. But if forced to consider the vulgar relationship between money and politics one could see it more cynically: if regional electors had their voting powers doubled, presumably politicians would double the subsidies and handouts given to regional Australia, thus perpetuating the inefficiencies and mendicancies that led to an inferiority complex.

Above all, a malapportioned Senate would be illiberal because it denies people full political franchise on the basis of where they live. Absent from many of the arguments is a recognition that demographic differences exist because people have the right to live wherever they want, and increasingly it is the case that they choose the fouled dwellings of metropolis over the unmolested plains of Arcadia. Forcing people to ‘stay on the land’ was the favoured practice of Stalin, Mao and Pol Pot[51]. We should not harbour similar desires in Australia, nor reward stasis by giving its hosts disproportionate representation in the Parliament.

Bibliography