Murphy v Electoral Commissioner
[2016] HCA 36
•5 September 2016
HIGH COURT OF AUSTRALIA
FRENCH CJ,
KIEFEL, BELL, GAGELER, KEANE, NETTLE AND GORDON JJ
ANTHONY JOHN MURPHY & ANOR PLAINTIFFS
AND
ELECTORAL COMMISSIONER & ANOR DEFENDANTS
Murphy v Electoral Commissioner
[2016] HCA 36
Date of Order: 12 May 2016
Date of Publication of Reasons: 5 September 2016M247/2015
ORDER
The questions stated by the parties in the amended special case dated 1 April 2016, as amended by the addendum to the amended special case dated 11 May 2016, and referred for consideration by the Full Court be answered as follows:
Question 1
Do one or both of the first plaintiff and the second plaintiff have standing to seek the relief sought in paragraphs 1, 2, 3 and/or 4 of the Further Amended Application for an Order to Show Cause?
Answer
The second plaintiff has standing and it is otherwise unnecessary to answer the question with respect to the first plaintiff.
Question 2
Are any or all of sections 94A(4), 95(4), 96(4), 102(4), 103A(5), 103B(5) and 118(5) of the Commonwealth Electoral Act 1918 (Cth) contrary to ss 7 and 24 of the Constitution and therefore invalid?
Answer
No.
Question 3
If the answer to Question 2 in relation to a section is yes, do sections 152(1)(a) and 155 of the Act have the same or substantially the same operation or effect as the impugned provisions or any of them and, if so, are sections 152(1)(a) and 155 invalid and of no effect?
Answer
The question does not arise.
Question 4
If the answer to Question 2 or Question 3 in relation to a section is yes, is that section, or are those sections, severable from the rest of the Act?
Answer
The question does not arise.
Question 5
What if any relief should be granted?
Answer
None.
Question 6
Who should pay the costs of the special case?
Answer
The first plaintiff.
Representation
R Merkel QC, B K Lim and C J Tran for the plaintiffs (instructed by King & Wood Mallesons)
Submitting appearance for the first defendant
J T Gleeson SC, Solicitor-General of the Commonwealth and N J Owens with K E Foley for the second defendant (instructed by Australian Government Solicitor)
Intervener
C D Bleby SC with D F O'Leary for the Attorney-General for the State of South Australia, intervening (instructed by Crown Solicitor (SA))
Notice: This copy of the Court's Reasons for Judgment is subject to formal revision prior to publication in the Commonwealth Law Reports.
CATCHWORDS
Murphy v Electoral Commissioner
Constitutional law (Cth) – Legislative power – Franchise – Power of Parliament to regulate exercise of entitlement to enrol to vote – Provisions of Commonwealth Electoral Act 1918 (Cth) precluding consideration of claims for enrolment or transfer of enrolment and amendment of Electoral Rolls during "suspension period" from 8pm on day of closing of Electoral Rolls until close of polling for election – Whether burden on constitutional mandate that Parliament be "directly chosen by the people" – Whether burden justified by substantial reason – Relevance of Roach v Electoral Commissioner (2007) 233 CLR 162; [2007] HCA 43 and Rowe v Electoral Commissioner (2010) 243 CLR 1; [2010] HCA 46.
Words and phrases – "adequacy in its balance", "burden", "constitutional mandate of popular choice", "directly chosen by the people", "franchise", "necessity", "obvious and compelling alternative", "reasonably appropriate and adapted", "structured proportionality", "substantial reason", "suitability".
Constitution, ss 7, 10, 24, 30, 51(xxxvi).
Commonwealth Electoral Act 1918 (Cth), ss 94A(4), 95(4), 96(4), 101, 102(4), 103A(5), 103B(5), 118(5).
FRENCH CJ AND BELL J.
Introduction
The plaintiffs in this special case, which was referred to the Full Court by Nettle J on 1 April 2016, sought relief including a declaration that a number of provisions of the Commonwealth Electoral Act 1918 (Cth) ("the Act") are invalid. The impugned provisions prevent the Electoral Commissioner from placing the name of a person on an Electoral Roll for a Division until after the close of the poll for an election for the Division, where the person's claim for enrolment or for transfer to the Roll from the Roll of another Division has been received after 8 pm on the day of the closing of the Rolls. The period from the closing of the Rolls to the close of the poll for an election is referred to in these reasons as the "suspension period". The impugned provisions also prevent the Electoral Commissioner, during the suspension period, from exercising statutory powers, without claim or notice, to update or transfer enrolments where there has been a change of address and to enrol unenrolled persons. They preclude removal, during that period, of an elector's name from a Divisional Roll where there has been an objection to that person's enrolment under Pt IX of the Act.
At the time of the hearing, the first plaintiff was enrolled on the Electoral Roll for the Division of Wills in the State of Victoria. Initially, his standing was challenged by the second defendant, the Commonwealth. The second plaintiff, who was on the Electoral Roll for the Division of Newcastle, intended to nominate herself as a candidate for election to the House of Representatives in that Division. She was an independent candidate for election to the House of Representatives in the Division in 2013. She was joined as a party, with the consent of the Commonwealth, at the hearing of the special case. She had not been endorsed by a registered political party. In order that she could contest the election, her nomination had to be signed by 100 persons who were entitled to vote at the election in the Division of Newcastle[1]. Her standing was not disputed and, on that basis, the challenge to the standing of the first plaintiff was not pursued.
[1]Act, s 166.
The plaintiffs said that the suspension of the processing of claims, transfers, objections and amendments to the Roll initiated by the Electoral Commissioner precluded people otherwise eligible to enrol and vote at a particular federal election from doing so and produced an inaccurate and distorted Electoral Roll. The provisions were said, on that account, to have an effect adverse to the constitutional requirement for election of members of the Parliament by the choice of the people. That effect being brought about for no substantial reason and being disproportionate to any legitimate end served by the impugned provisions, they were said to be invalid.
In addition to declaratory relief, the plaintiffs sought a writ of prohibition pursuant to s 75(v) of the Constitution prohibiting the Electoral Commissioner from "giving effect to, or taking any steps in reliance upon" the impugned provisions. That claim involved a remedial double negative. It required the Electoral Commissioner to act as though the statutory obligations imposed on him, otherwise limited by the impugned provisions, were extended up to and including polling day.
The plaintiffs' case did not involve a challenge, as in Roach v Electoral Commissioner[2], to a disqualifying provision targeting a particular class of persons such as sentenced prisoners. Nor was it, as in Rowe v Electoral Commissioner[3], a challenge to a law removing an existing opportunity to enrol up to seven days after the issue of the writs for an election. In this case the plaintiffs said, in effect, that Parliament had not gone far enough when it left in place long‑standing time limits, essentially dating back to the 1930s, on the processing of new and transferred enrolments, objections and amendments to the Rolls.
[2](2007) 233 CLR 162; [2007] HCA 43.
[3](2010) 243 CLR 1; [2010] HCA 46.
The questions in the special case are set out at the end of these reasons. At the conclusion of the hearing the Court answered those questions adversely to the plaintiffs, holding that none of the impugned provisions were invalid and that the first plaintiff should pay the costs of the special case. Our reasons for joining in those orders follow. It is necessary first to outline the legislative framework in which the impugned provisions appear.
The general legislative framework
The Act establishes the Australian Electoral Commission[4] and the office of the Electoral Commissioner as its Chief Executive Officer and one of its members[5]. Part IV of the Act provides that each State and Territory shall be distributed into Electoral Divisions[6] and that one Member of the House of Representatives shall be chosen for each Electoral Division[7]. Part VI provides for a Roll of electors for each State and Territory[8]. Each such Roll is made up of Rolls for the Electoral Divisions within the State or Territory which are to contain the names, addresses and prescribed particulars of each elector for the Division[9]. Part VII of the Act sets out the qualifications and disqualifications for enrolment and for voting. Section 93 sets out the conditions upon which persons "shall be entitled to enrolment." Australian citizens who have attained 18 years of age are so entitled[10]. Section 100 allows, but does not require, a person who has turned 16 but is under 18 to make a claim for enrolment. The term "Elector" is defined in the Act as "any person whose name appears on a Roll as an elector."[11] An elector whose name is on the Roll for a Division is "entitled to vote at elections of Members of the Senate for the State that includes that Division and at elections of Members of the House of Representatives for that Division."[12]
[4]Act, s 6.
[5]Act, ss 6 and 18.
[6]Act, s 56. Under s 55A, Pt IV applies to the Northern Territory as if it was a State if the Electoral Commissioner determines that the number of Members of the House of Representatives to be chosen in the Northern Territory at a general election is 2 or greater. On 13 November 2014, the Acting Electoral Commissioner determined that 2 Members of the House of Representatives were to be chosen in the Northern Territory at a general election.
[7]Act, s 57. There is provision for subdivisions of Divisions in the Act which are mentioned in a number of the sections summarised in these reasons. No Division is currently divided into subdivisions, and hence s 4(4) of the Act has the effect that references to subdivisions should be read as references to Divisions.
[8]Act, s 81.
[9]Act, ss 82(4) and 83(1). The addresses of eligible overseas electors and itinerant electors are not required: s 83(2).
[10]Act, s 93(1)(a) and (b)(i). Non-citizens who would have been British subjects within the meaning of the relevant citizenship law had it continued in force and whose names were, immediately before 26 January 1984, on a Roll are also entitled: Act, s 93(1)(b)(ii).
[11]Act, s 4(1).
[12]Act, s 93(2).
Part VIII of the Act deals with mechanisms for enrolment. The principal means by which a qualified person can be included on the Roll for a Division is by making a claim for enrolment[13]. The claim is to be processed by the Electoral Commissioner as provided by the Act[14]. Electors can apply to transfer from the Roll of one Division to the Roll of another[15]. Section 101 provides for compulsory enrolment for persons entitled to be enrolled for a Division other than by virtue of ss 94, 94A, 95, 96 or 100. Each person entitled to be enrolled for any Division whether by way of enrolment or transfer of enrolment, except residents of Norfolk Island, is required to "forthwith fill in and sign a claim and send or deliver the claim to the Electoral Commissioner." The Electoral Commissioner can enrol or transfer the enrolment of persons who have not made a claim for enrolment or transfer[16]. The Electoral Commissioner can also correct any Roll[17]. Alterations to the Roll to correct mistakes and to remove the names of deceased electors may be made at any time[18]. The removal of the name of a person from a Roll who is not entitled to enrol and had made a claim for enrolment containing a false statement may be effected at any time between the date of issue of the writ for an election for the relevant Division and before the close of polling for that election[19]. The Electoral Commissioner may enter the name of an elector who is not enrolled, and who has made a declaration vote, on the Roll for the Division in which the elector was living at the time of voting at a preliminary scrutiny of declaration votes[20] if the vote is in order and the omission of the elector's name from the Roll for the Division was due to an error made by an officer or to a mistake of fact[21].
[13]Act, ss 98, 99 and 101.
[14]Act, s 102.
[15]Act, s 99(2).
[16]Act, ss 103A and 103B.
[17]Act, s 105.
[18]Act, s 105(3).
[19]Act, s 106.
[20]Preliminary scrutinies of declaration votes are continuously held from the last Monday before the close of the poll for the election until a time not earlier than 13 days after the close of the poll: Act, s 266.
[21]Act, s 105(4) and Sched 3.
Part XIII provides for writs to be issued by the Governor‑General for elections to the House of Representatives, pursuant to s 32 of the Constitution, and for the election of Senators for the Territories[22]. The State Governors issue the writs for the election of Senators for each State, pursuant to s 12 of the Constitution[23]. The date fixed for the close of the Rolls is the seventh day after the date of the relevant writ[24]. The date fixed for the nomination of candidates must be not less than 10 days nor more than 27 days after the date of the relevant writ[25]. Polling day must be not less than 23 days nor more than 31 days after the date of nomination[26].
[22]Act, ss 151, 152 and 154.
[23]Act, ss 152 and 153. By convention, each Governor adopts the suggestion of the Governor-General, upon the advice of the Prime Minister, that the writ proclaim the same deadlines to apply as those in the writs issued for elections to the House of Representatives and for elections of Senators for the Territories: Odgers' Australian Senate Practice, 12th ed (2008) at 94-95.
[24]Act, s 155.
[25]Act, s 156.
[26]Act, s 157.
Part XVI of the Act, entitled "The polling", concerns arrangements to be made for the polling day. Section 208(1) requires the Electoral Commissioner to prepare a list of voters for each Division and to certify the list. A paper copy of the certified list for a Division is to be delivered to the presiding officer at each polling place before the start of voting[27]. A copy must also be delivered to each place at which pre-poll ordinary voting is available for voters enrolled for that Division[28]. The list is created from data extracted from the Australian Electoral Commission's IT system, known as "RMANS", in which each elector's details are stored. The Australian Electoral Commission's practice is to process all enrolment applications received prior to the close of the Rolls within 40 hours of their close. It aims to conclude production of the certified list approximately 48 hours after the close of the Rolls. The delivery of certified lists from printers occurs over a 10 day period with priority delivery to early voting centres. The certified list once prepared is not updated. Any error or omission may be remedied by proclamation specifying the matter dealt with and providing for the course to be followed[29].
[27]Act, s 208(3).
[28]Act, s 208(4).
[29]Act, s 285(1).
There is an electronic version of the Electoral Rolls known as a "Notebook Roll" maintained by the Australian Electoral Commission to manage any additional applications for enrolment or changes received before the suspension period, but not processed before the printing of the certified lists, including the addition of names from claims under s 101. Other changes may include the removal of names of deceased electors, the reinstatement of eligible electors previously removed by mistake and changes as a result of the detection of enrolments pursuant to claims including a false statement[30]. The Notebook Roll operates as an administrative mechanism to assist in processing updates to the Roll in order, ultimately, to facilitate the scrutiny of declaration votes. In addition, under s 208A of the Act, inserted in 2010[31], the Electoral Commissioner may arrange for the preparation of an "approved list of voters for a Division". This is a list in electronic form which contains the same information as the certified list for the Division most recently prepared[32]. The certified and approved lists for each Division are used to determine the entitlement of a person to cast an ordinary vote on polling day.
[30]Act, ss 105 and 106.
[31]Electoral and Referendum Amendment (Modernisation and Other Measures) Act 2010 (Cth), Sched 4 item 8.
[32]Act, s 4(1) definition of "approved list".
The impugned provisions
Three of the impugned provisions, ss 94A(4), 95(4) and 96(4), are found in Pt VII of the Act. Sections 94A, 95 and 96 respectively provide for applications for enrolment by persons residing outside Australia, by their spouses, de facto partners or children living outside Australia and by itinerant electors. In each case, the impugned subsection provides that if an application under the section to which it relates is received by the Electoral Commissioner after 8 pm on the day of the close of the Rolls for an election to be held in a Division, and the application relates to that Division, the person must not be added to the Roll for that Division until after the close of the poll for that election.
Section 102 sets out the actions to be taken by the Electoral Commissioner on receipt of a claim pursuant to s 101 for enrolment or transfer of enrolment. Section 102(4) provides that if such a claim is received by the Electoral Commissioner during the "suspension period" then "the claim must not be considered until after the end of the suspension period." The suspension period is defined as the period commencing at 8 pm on the day of the close of the Rolls for an election to be held in a Division and ending on the close of the poll for that election. As noted earlier, although the term "suspension period" is only defined in the Act for the purposes of s 102 it is used in these reasons to refer to the same period in which each of the impugned provisions operates.
Section 103A(5) also prevents the Electoral Commissioner from taking action of his or her own motion to update or transfer a person's enrolment during the suspension period. Section 103B(5) imposes the same prohibition with respect to the power of the Electoral Commissioner under s 103B to enrol an unenrolled person without claim or notice from that person.
Section 118 appears in Pt IX of the Act, which deals with objections to the enrolment of a person in a Division. By s 118(5) the Electoral Commissioner must not, during the suspension period, remove an elector's name from the Roll of the Division under the powers conferred on him or her by s 118(3) and (4A).
History of the suspension period
The legislative scheme for enrolment and for a cut-off date for enrolment for a particular election is to be understood by reference to the history and purposes of the Electoral Roll. Some of that history was set out in Rowe[33]. The registration and listing of qualified electors on an Electoral Roll as a condition of the exercise of the right to vote dates back to the enactment of the Representation of the People Act 1832[34]. The purpose of registration was practical and originally directed to dealing with the complicated and diverse qualifications required for a person to become an elector[35].
[33](2010) 243 CLR 1 at 14-18 [10]-[17] per French CJ.
[34]2 & 3 Will IV c 45, s 26.
[35](2010) 243 CLR 1 at 15 [12].
The electoral laws of the Australian colonies in the late 19th century replicated key features of the British system, including its requirements for listing, enrolment and registration. At the time of Federation those colonial laws conditioned the right to vote in an election upon enrolment on the relevant Electoral Roll[36]. They also provided for closure of the Electoral Rolls to new enrolments or transfers before polling day, albeit there were variations in the cut‑off dates[37].
[36]Parliamentary Electorates and Elections Act 1893 (NSW), s 80; Constitution Act Amendment Act 1890 (Vic), s 241; Electoral Code 1896 (SA), ss 36, 116 and 126; Elections Act 1885 (Q), s 40; Electoral Act 1899 (WA), ss 21, 87 and 104; Electoral Act 1896 (Tas), s 57.
[37]Parliamentary Electorates and Elections Act 1893 (NSW), ss 47‑51; Constitution Act Amendment Act 1890 (Vic), ss 97 and 186; Electoral Code 1896 (SA), ss 51, 52 and 57; Elections Act 1885 (Q), s 40; Electoral Act 1899 (WA), ss 37 and 44; Electoral Act 1896 (Tas), s 57.
Enrolment and voting for federal elections from 1902 were regulated by the Commonwealth Electoral Act 1902 (Cth) and the Commonwealth Franchise Act 1902 (Cth). Section 3(1) of the present Act provided that these Acts and several amending Acts were to be repealed on dates fixed by proclamation, a process which concluded in 1934. Between 1902 and 1983 the Rolls closed on the day the writs for an election were issued. From at least the 1930s an executive practice developed of announcing the election a few days before the Governor-General was asked to dissolve Parliament and issue writs for the election of the Members of the House of Representatives. This administrative practice provided a grace period enabling persons wishing to enrol or transfer enrolment to do so before the issue of the writs.
In 1983 the statutory cut-off point for consideration of claims for enrolment or transfer of enrolments was extended beyond the date of the issue of the writs to the date of close of the Rolls, which was fixed as seven days after the issue of the writs[38]. The objective was "to make it easier for electors to get on the rolls and stay on the rolls"[39]. The Act was further amended in 2006 so that a claim for enrolment received between 8 pm on the date of the issue of the writs for an election and the close of polling for that election could not be considered until after the close of polling for the election. Transfer claims could be considered if they were received before 8 pm on the day of the close of the Rolls, which was fixed as the third working day after the issue of the writs[40]. However, following the decision in Rowe in which those amendments were held to be invalid, the text of the Act was amended to restore the seven day grace period[41].
[38]Commonwealth Electoral Legislation Amendment Act 1983 (Cth).
[39]Australia, House of Representatives, Parliamentary Debates (Hansard), 2 November 1983 at 2216.
[40]Electoral and Referendum Amendment (Electoral Integrity and Other Measures) Act 2006 (Cth), Sched 1 items 41 and 52.
[41]Electoral and Referendum Amendment (Enrolment and Prisoner Voting) Act 2011 (Cth), Sched 1 items 2-8.
Practical operation of the impugned provisions
To provide an indication of the practical impact of the suspension period, the special case included a table of the numbers of claims for new and updated enrolments received during that period for the 2004, 2007, 2010 and 2013 federal elections. They were as follows:
Inter-State transfer Intra-Division transfer Intra-State transfer New enrolment Enrolment update (no change of address) Total 2004 9831 0 72279 60597 21329 164037 2007 6352 0 67469 48235 21579 143636 2010 8644 0 86947 44636 21504 161733 2013 10241 76708 46430 52694 42512 228585 There were no unresolved objections at the beginning of the suspension periods in 2004 and 2007, but there were 22,579 in 2010. That figure was high because the Australian Electoral Commission did not anticipate the calling of the election. The figure at the 2013 election was 2,975.
The figures set out in the special case were said by the plaintiffs to demonstrate the magnitude of the practical "burden" that the impugned provisions imposed upon the constitutional mandate of popular choice. They were said to represent the lower limits of the number of persons effectively disenfranchised by the suspension period for each of the elections to which they related. There were potentially, it was suggested, additional persons who did not make claims because they knew of the suspension period, advice about it having been provided on the Australian Electoral Commission website. In the end, the significance of these figures depended upon the resolution of a fundamental difficulty with the plaintiffs' case arising out of the character of the impugned provisions as long-standing time limits for settlement of the Rolls prior to an election.
The plaintiffs' reliance upon Roach and Rowe
The plaintiffs submitted that:
1.A law which has the practical operation of effecting a legislative disqualification from what otherwise is the popular choice mandated by the Constitution is invalid unless the disqualification is for a substantial reason.
2.Such a law will be for a substantial reason only if it is reasonably appropriate and adapted to serve an end which is consistent or compatible with the constitutionally mandated system of representative government.
Those submissions were based upon the decisions of this Court in Roach and Rowe. The impugned provisions were said to be invalid in light of them.
In Roach, the Court was concerned with the validity of amendments to the Act, enacted in 2004 and 2006, disqualifying sentenced prisoners from voting at federal elections. The 2004 amendment disqualified persons serving sentences of three years or more. The 2006 amendment disqualified any person serving a sentence of imprisonment regardless of duration. The Court held the former disqualification to be valid and the latter disqualification to be invalid. Roach involved a law directly affecting the qualification of a person to be enrolled and to vote. Further, the law which was held invalid in that case had imposed a disqualification upon a class of persons not previously disqualified from voting.
In Rowe, the Court was concerned with the validity of the amendments to the Act, made in 2006, which had the effect of precluding consideration of a claim for enrolment or transfer of enrolment where the claim for enrolment was lodged after the issue of the writs and where the claim for transfer of enrolment was lodged after the close of the Rolls. The amendments removed the pre‑existing seven day grace period allowing enrolment after issue of the writs and significantly abridged the opportunities for electors to transfer their enrolments[42].
[42](2010) 243 CLR 1 at 12 [3] per French CJ.
The impugned laws in this case were similar to the impugned laws in Rowe only to the extent that they both provided for suspension periods during which claims for enrolment, transfers of enrolment, objections and corrections to the Roll could not be considered. The significant difference was that, unlike the present case, Rowe concerned laws which reduced existing opportunities for enrolment or transfer of enrolment prior to an election. The plaintiffs' case depended upon a generalisation of Roach and Rowe and a characterisation of the impugned provisions as imposing a burden upon the realisation of the constitutional mandate of choice by the people.
Some general propositions
There are three key propositions derived from the text of the Constitution which lie at the threshold of the special case:
1.Sections 7 and 24 of the Constitution provide that the members of the Senate and the House of Representatives shall be directly chosen, "by the people of the State" in the case of the Senate and "by the people of the Commonwealth" in the case of the House of Representatives.
2.Sections 8 and 30 of the Constitution, read with s 51(xxxvi), empower the Parliament to make laws providing for the qualification of voters[43]. It was the exercise of that power which was in issue in Roach.
3.Sections 10 and 31 of the Constitution, read with s 51(xxxvi), confer upon the Parliament power to make laws relating to the election of Senators and Members of the House of Representatives[44]. It was the exercise of that power which was in issue in Rowe and is in issue in this case.
[43](2007) 233 CLR 162 at 173 [6] per Gleeson CJ.
[44](2010) 243 CLR 1 at 14 [8] per French CJ.
While the limits of the legislative powers considered in Roach and Rowe were variously expressed by the majority Justices in those cases, they reflected an essentially common approach to the criterion of validity.
In Roach, Gleeson CJ held that the contemporary understanding of the constitutional term "chosen by the people" required nothing less than universal adult suffrage[45]. Parliament's power to define exceptions to that requirement was limited by the negative criterion that[46]:
"disenfranchisement of any group of adult citizens on a basis that does not constitute a substantial reason for exclusion from such participation would not be consistent with choice by the people." (footnote omitted)
Gleeson CJ's use of the term "substantial reason" was a borrowing from Brennan CJ in McGinty v Western Australia[47]. Brennan CJ observed that, unaffected by context, "chosen by the people" admitted of different meanings including "some requirement of a franchise that is held generally by all adults or all adult citizens unless there be substantial reasons for excluding them."[48]
[45](2007) 233 CLR 162 at 174 [7].
[46](2007) 233 CLR 162 at 174 [7].
[47](1996) 186 CLR 140 at 170; [1996] HCA 48. A criterion foreshadowed, as Gageler J points out in his reasons at [84], in Attorney-General (Cth); Ex rel McKinlay v The Commonwealth (1975) 135 CLR 1 at 36, 69; [1975] HCA 53.
[48](1996) 186 CLR 140 at 170.
The word "substantial" appears in the common law and in statutes in areas as diverse as defamation, intellectual property, contract, criminal law, competition law and wills. It directs its readers' attention from form to substance. It generally indicates a narrowing qualification in some otherwise broadly expressed legal criterion and, according to context, may be a cautionary direction about the limits of the judicial power.
Plainly, neither Brennan CJ in McGinty nor Gleeson CJ in Roach meant the term "substantial reason" to be at large. Gleeson CJ said[49]:
"There would need to be some rationale for the exception; the definition of the excluded class or group would need to have a rational connection with the identification of community membership or with the capacity to exercise free choice."
A rational connection between exclusion from the right to vote and community membership might be found in conduct manifesting such a rejection of civic responsibility as to warrant temporary withdrawal of a civic right[50]. Within that general framework, Gleeson CJ held the indiscriminate exclusion of all sentenced prisoners from the right to vote to be invalid. The notion of rational connection thus explained is embedded in the requirement for a "reason" in the term "substantial reason". It precludes arbitrary exclusions or exclusions not founded in any relevant normative or practical objective. The requirement of substantiality means that disqualifying legislation cannot be justified solely on the basis that the exclusions it creates have a formal, logical connection with the purpose it purports to serve.
[49](2007) 233 CLR 162 at 174 [8].
[50](2007) 233 CLR 162 at 175 [8].
The other members of the majority in Roach, Gummow, Kirby and Crennan JJ, like Gleeson CJ, quoted Brennan CJ's reference in McGinty to "substantial reasons" for excluding adults or adult citizens from the franchise[51]. They said that a reason would be "substantial" if it was, in language familiar in Australian public law, "reasonably appropriate and adapted to serve an end which is consistent or compatible with the maintenance of the constitutionally prescribed system of representative government."[52] It did not require a justification of the exclusion as "essential" or "unavoidable". Rather, the criterion of a "substantial reason" imported a notion of proportionality in the sense that[53]:
"[w]hat upon close scrutiny is disproportionate or arbitrary may not answer to the description reasonably appropriate and adapted for an end consistent or compatible with observance of the relevant constitutional restraint upon legislative power."
The terminology "reasonably appropriate and adapted" in this context marks the limits of legislative power and the borderlands of the judicial power. As Gleeson CJ observed in Mulholland v Australian Electoral Commission[54]:
"For a court to describe a law as reasonably appropriate and adapted to a legitimate end is to use a formula which is intended, among other things, to express the limits between legitimate judicial scrutiny, and illegitimate judicial encroachment upon an area of legislative power."
[51](2007) 233 CLR 162 at 198 [83].
[52](2007) 233 CLR 162 at 199 [85].
[53](2007) 233 CLR 162 at 199 [85].
[54](2004) 220 CLR 181 at 197 [33]; [2004] HCA 41.
Gleeson CJ referred to the long Australian history of judicial application of the term "reasonably appropriate and adapted" originally derived from the judgment of Marshall CJ in McCulloch v Maryland[55]. Four Justices of this Court said in McCloy v New South Wales[56] that, as used in Australian law, it describes a class of criteria developed over many years to determine whether legislative or administrative acts are within the constitutional or legislative grant of power under which they purport to have been done[57]. Such criteria have been applied to test the validity of the exercise of purposive powers[58], incidental powers, which must serve the purposes of the substantive powers to which they are incidental[59], and powers whose exercise may limit or restrict the enjoyment of a constitutional guarantee, immunity or freedom, including the implied freedom of political communication[60]. An extensive discussion of the history and application of the criterion appears in the judgment of Kiefel J in Rowe[61].
[55](2004) 220 CLR 181 at 199-200 [39] citing 4 Wheat 316 at 421 (1819).
[56](2015) 89 ALJR 857; 325 ALR 15; [2015] HCA 34 (French CJ, Kiefel, Bell and Keane JJ).
[57](2015) 89 ALJR 857 at 863 [3] per French CJ, Kiefel, Bell and Keane JJ; 325 ALR 15 at 19.
[58]The Commonwealth v Tasmania (The Tasmanian Dam Case) (1983) 158 CLR 1 at 130, 138-139 per Mason J, 172 per Murphy J, 232 per Brennan J, 259-260 per Deane J; [1983] HCA 21; Richardson v Forestry Commission (1988) 164 CLR 261 at 295-296 per Mason CJ and Brennan J, 303 per Wilson J, 311-312 per Deane J, 336 per Toohey J, 344-346 per Gaudron J; [1988] HCA 10; Attorney-General (SA) v Adelaide City Corporation (2013) 249 CLR 1 at 37-40 [55]-[60] per French CJ; [2013] HCA 3.
[59]Davis v The Commonwealth (1988) 166 CLR 79 at 98-100 per Mason CJ, Deane and Gaudron JJ, 101 per Wilson and Dawson JJ agreeing, 117 per Toohey J agreeing; [1988] HCA 63; Nationwide News Pty Ltd v Wills (1992) 177 CLR 1 at 30-31 per Mason CJ, 78 per Deane and Toohey JJ, 101 per McHugh J, 95 per Gaudron J agreeing with Mason CJ and McHugh J; [1992] HCA 46.
[60]Castlemaine Tooheys Ltd v South Australia (1990) 169 CLR 436 at 472 per Mason CJ, Brennan, Deane, Dawson and Toohey JJ; [1990] HCA 1; Lange v Australian Broadcasting Corporation (1997) 189 CLR 520 at 567-568; [1997] HCA 25; Betfair Pty Ltd v Western Australia (2008) 234 CLR 418 at 476-477 [101]‑[103] per Gleeson CJ, Gummow, Kirby, Hayne, Crennan and Kiefel JJ; [2008] HCA 11.
[61](2010) 243 CLR 1 at 133-139 [431]-[455].
A law which excludes a class of adult citizens from an existing right of participation in a federal election and thus reduces the extent to which the election represents a choice of representatives "by the people" might be thought to fall into a category analogous with laws limiting or restricting the enjoyment of a constitutional guarantee, immunity or freedom. Such a law may be tested for validity under the general criterion — is it reasonably appropriate and adapted to serve an end which is consistent or compatible with the maintenance of the constitutionally prescribed system of representative government? The particular aspect of the system of representative government in play in such a case is the mandate of choice by the people. It is true that the criterion does not set out "precise metes and bounds" which chart the generality of that mandate and which Parliament in the exercise of its powers to determine exceptions to or disqualifications from the franchise must respect[62]. Nevertheless it is a criterion of a kind long used in Australian courts in a variety of settings.
[62](2007) 233 CLR 162 at 206 [113] per Hayne J.
In Rowe, the impugned law diminished the opportunities for enrolment and transfer of enrolment which existed prior to its enactment[63]. French CJ characterised it as thereby effecting a "significant detriment in terms of the constitutional mandate", which would have to be weighed "against the legitimate purposes of the Parliament" which it was said to serve[64]. The detriment was "disproportionate to the benefits of a smoother and more efficient electoral system to which the amendments were directed."[65] That approach did not set up a sui generis test of validity. It provided an answer to the question whether the impugned law, notwithstanding its legal effect, was reasonably appropriate and adapted to serve a legitimate purpose of the Parliament consistent with the prescribed system of representative government. It was adopted in the context of an acceptance of the approach of the majority in Roach, including the proposition that a "substantial reason" for an exception to the franchise is needed in order to satisfy the general proportionality criterion[66]. The proportionality criterion was expressly applied by Gummow and Bell JJ. As their Honours pointed out, the Commonwealth in that case accepted that if the legal or judicial operation of a law was to disqualify adult citizens from enrolling and thus from exercising their franchise, the consistency of that law with ss 7 and 24 of the Constitution was to be determined in accordance with the reasoning in Roach[67]. Thus the consideration upon which the Commonwealth relied had to supply "a substantial reason in the sense used in the reasons of the two majority judgments in Roach."[68] Their Honours said that once it was shown that the 2006 amendments in issue in Rowe had the practical operation of effecting a legislative disqualification from participation in the popular choice mandated by the Constitution, the relevant question was that propounded by Gleeson CJ in Roach: "whether ... there has been broken the rational connection necessary to reconcile the disqualification with the constitutional imperative"[69]. Their Honours equated that question, via the term "substantial reason" used by Gleeson CJ, with the question formulated by Gummow, Kirby and Crennan JJ in Roach and expressed in terms of the general proportionality criterion[70].
[63](2010) 243 CLR 1 at 38 [78].
[64](2010) 243 CLR 1 at 38 [78].
[65](2010) 243 CLR 1 at 39 [78].
[66](2010) 243 CLR 1 at 20-21 [23]-[25].
[67](2010) 243 CLR 1 at 56-57 [151].
[68](2010) 243 CLR 1 at 58 [157].
[69](2010) 243 CLR 1 at 59 [161] citing (2007) 233 CLR 162 at 182 [24].
[70](2010) 243 CLR 1 at 59 [161].
Crennan J, the other member of the majority in Rowe, read ss 7 and 24 of the Constitution as mandating a franchise which would result in a democratic representative government, a concept the content of which had evolved to require "a fully inclusive franchise — that is, a franchise free of arbitrary exclusions based on class, gender or race."[71] The impugned provisions in Rowe were not shown to be necessary or appropriate for the protection of the integrity of the Rolls, the purpose advanced in that case by the Commonwealth[72]. Her Honour there used the word "necessary" as not limited to what was essential or unavoidable but encompassing what was "'reasonably appropriate and adapted' to serve a legitimate end."[73]
[71](2010) 243 CLR 1 at 117 [367].
[72](2010) 243 CLR 1 at 120 [384].
[73](2010) 243 CLR 1 at 118 [374].
As appears from the above, in their various ways the reasons of the majority in Rowe fell within the same general rubric of proportionality long applied in Australian public law. The Commonwealth in this case also accepted the applicability of that general rubric, and did not take issue with the result in Rowe.
A structured approach to the application of the general proportionality criterion to a law said to burden the implied freedom of political communication was recently set out in the joint judgment in McCloy. It was invoked by the plaintiffs in support of their case. That approach, foreshadowed in the judgment of Kiefel J in Rowe[74], involved an unpacking of the question whether a law found to burden the implied freedom, and to do so for a legitimate purpose, was "reasonably appropriate and adapted to advance that legitimate object"[75]. The analysis used to answer the proportionality question was undertaken by reference to three considerations drawn from the approach of European and, in particular, German courts[76]:
1.Suitability — whether the law had a rational connection to the purpose of the provision — a criterion which reflects that adopted by Gleeson CJ in Roach.
2.Necessity — whether there was an obvious and compelling alternative, reasonably practicable means of achieving the same purpose with a less restrictive effect on the freedom.
3.Adequacy in its balance — whether the extent of the restriction imposed by the impugned law was outweighed by the importance of the purpose it served.
The adoption of that approach in McCloy did not reflect the birth of some exotic jurisprudential pest destructive of the delicate ecology of Australian public law. It is a mode of analysis applicable to some cases involving the general proportionality criterion, but not necessarily all. For example, as Kiefel J observed in Rowe[77]:
"A test of reasonable necessity, by reference to alternative measures, may not always be available or appropriate having regard to the nature and effect of the legislative measures in question."
In Davis v The Commonwealth[78], Nationwide News Pty Ltd v Wills[79] and Australian Capital Television Pty Ltd v The Commonwealth[80], as her Honour observed, want of proportionality was assessed by reference to a range of factors.
[74](2010) 243 CLR 1 at 140-142 [460]-[466].
[75](2015) 89 ALJR 857 at 862-863 [2]; 325 ALR 15 at 19.
[76](2010) 243 CLR 1 at 140 [460] per Kiefel J.
[77](2010) 243 CLR 1 at 136 [445].
[78](1988) 166 CLR 79.
[79](1992) 177 CLR 1 at 31.
[80](1992) 177 CLR 106; [1992] HCA 45.
The three considerations relevant to proportionality set out in McCloy are capable of application to laws burdening or infringing a constitutional guarantee, immunity or freedom. In the case of the constitutional mandate of choice by the people, they may be relevant depending upon the character of the law said to diminish the extent of the realisation of that mandate. Apart from the suitability requirement, they do not have universal application in determining whether a law, a delegated legislative instrument or an administrative act is a valid exercise of the relevant grant of power, being reasonably appropriate and adapted to serve the purpose of the grant.
The plaintiffs in the present case were concerned with provisions reflecting long-standing limits on the times at which a qualified person could be registered on the Roll. Theirs was not a case about a law reducing the extent of the realisation of the constitutional mandate. It was ultimately a complaint that the legislation did not go far enough in the provision of opportunities for enrolment. The difficulty confronting their case was demonstrated by their attempt to apply the necessity consideration in McCloy by reference to what were said to be obvious and compelling legislative alternatives. One was enrolment up to and including polling day, said to be demonstrated by the electoral systems of three Australian States. Another was a reduction of the suspension period by making it a fixed number of days counted back from polling day. These arguments invited the Court to undertake an hypothetical exercise of improved legislative design by showing how such alternatives could work. In so doing, they invited the Court to depart from the borderlands of the judicial power and enter into the realm of the legislature. The McCloy analysis was inapposite in this case.
Conclusions
The plaintiffs submitted that:
"Any effective burden on the constitutional mandate of popular choice (such as a requirement to enrol in a particular manner or at a particular time, or even to vote in a particular manner or at a particular time) is constitutionally suspect and will be invalid unless justified by reference to a permissible substantial reason. The extent of the burden will inform the extent of justification required, but there should be no narrow approach to what constitutes an effective burden." (footnote omitted)
And further:
"Many burdens may be justifiable, but they must ultimately serve 'the end of making elections as expressive of the popular choice as practical considerations properly permit'." (footnote omitted)
The notion of a "burden", which was central to the plaintiffs' argument, was indicative of the difficulty they had in seeking to generalise from the case of a change to the law adverse to the exercise of the franchise, to the omission of the legislature to maximise opportunities for the exercise of the franchise.
The suspension period is a design feature of the Act, which has existed for a long time and for legitimate reasons. As was submitted for the Commonwealth, by making participation in the electoral process dependent upon membership of the single class of persons defined as "electors" by reference to the content of the Rolls on the date upon which they are closed, the orderly and efficient conduct of elections is advanced. The legislative scheme establishes the Rolls as the means of defining the class of eligible participants in the choice by the "people" to which ss 7 and 24 of the Constitution refer. It defines not only those who can vote on polling day but those who can nominate or be nominated[81] or have their details provided to candidates so that campaign material may be provided[82]. It defines those who can cast a postal vote and a pre-poll vote[83] and have a provisional vote admitted to scrutiny[84].
[81]Act, s 163.
[82]Act, s 90B.
[83]Act, ss 183 and 200A, Sched 2.
[84]Act, s 235.
It may be that in the light of modern technology, with appropriate electronic infrastructure and human and financial resources, a system could be devised which would allow enrolments to occur and alterations to be made to the Rolls up to and including polling day. Proposals for change have been made by the Australian Electoral Commission[85] and by the Joint Standing Committee on Electoral Matters[86], which were referred to in the special case, as were the legislative schemes in New South Wales, Victoria and Queensland. The plaintiffs also invoked the scheme in Queensland in support of a further "alternative" whereby the commencement of the suspension period would be determined by counting a fixed number of days backwards from the date of an election, rather than by counting forwards from the date of issue of the writs. The existence of such possibilities does not support a characterisation of the design limits of the existing Act as a "burden" upon the realisation of the constitutional mandate of popular choice. The impugned provisions do not become invalid because it is possible to identify alternative measures that may extend opportunities for enrolment. That would allow a court to pull the constitutional rug from under a valid legislative scheme upon the court's judgment of the feasibility of alternative arrangements. The plaintiffs' premise that the suspension period reflects a burden on the constitutional mandate of popular choice was not made out. The failure of that premise was fatal to the plaintiffs' attempts to generalise Roach and Rowe in support of their argument.
[85]Australia, Australian Electoral Commission, "Submission to the Joint Standing Committee on Electoral Matters on the Conduct of the 2010 Federal Election", (21 February 2011) at 61-63.
[86]Australia, The Parliament, Joint Standing Committee on Electoral Matters, The 2010 Federal Election: Report on the Conduct of the Election and Related Matters, (June 2011) at 37.
Answers to questions
The answers to the questions in the special case in which we joined were as follows:
Question 1:
Do one or both of the first plaintiff and the second plaintiff have standing to seek the relief sought in paragraphs 1, 2, 3 and/or 4 of the Further Amended Application for an Order to Show Cause?
Answer:
The second plaintiff has standing and it is otherwise unnecessary to answer the question with respect to the first plaintiff.
Question 2:
Are any or all of sections 94A(4), 95(4), 96(4), 102(4), 103A(5), 103B(5) and 118(5) of the Commonwealth Electoral Act 1918 (Cth) contrary to ss 7 and 24 of the Constitution and therefore invalid?
Answer:
No.
Question 3:
If the answer to Question 2 in relation to a section is yes, do sections 152(1)(a) and 155 of the Act have the same or substantially the same operation or effect as the impugned provisions or any of them and, if so, are sections 152(1)(a) and 155 invalid and of no effect?
Answer:
The question does not arise.
Question 4:
If the answer to Question 2 or Question 3 in relation to a section is yes, is that section, or are those sections, severable from the rest of the Act?
Answer:
The question does not arise.
Question 5:
What if any relief should be granted?
Answer:
None.
Question 6:
Who should pay the costs of the special case?
Answer:
The first plaintiff.
KIEFEL J. The plaintiffs challenged the validity of certain provisions of the Commonwealth Electoral Act 1918 (Cth) ("the Electoral Act"). They contended that these provisions are contrary to ss 7 and 24 of the Constitution. At the conclusion of the hearing of this matter the Court answered the questions which had been stated for it, the effect of which was that the provisions were held not to be invalid. I joined in the answers given by the Court. These are my reasons for doing so.
The provisions in question
The provisions sought to be impugned by the plaintiffs fell into two main categories. Sections 94A(4), 95(4), 96(4), 102(4) and 103B(5) provide that a person's name must not be added to the Electoral Roll for a Division during the period between 8:00pm on the day of the close of the Electoral Rolls and the close of the poll for the election ("the suspension period"). Sections 102(4) and 103A(5) provide that a claim for a transfer of enrolment must not be considered until after the end of the suspension period. The commencement of the suspension period is marked out by ss 152(1) and 155. Section 155 provides that the date fixed for the close of the Rolls is the seventh day after the date of the writ for the election, which is provided for in s 152(1).
The practical effect of these provisions is that when a writ for a federal election issues, a person who is not enrolled has seven days within which to do so or they will not be on the Roll and as a consequence not able to vote; and a person who wishes to transfer their enrolment to another Division has seven days within which to do so, otherwise they will not be able to vote in the Division in which they live.
The other provision which the plaintiffs challenged was s 118(5), which provides that a person's name must not be removed from the Electoral Roll during the suspension period. The plaintiffs sought to use this provision in aid of a submission concerning the distortion of the Rolls as a result of the suspension period. It is not directly relevant to questions of validity of the other impugned provisions.
A period of seven days has been provided for by statute for new enrolments, or transfers of enrolment, after the issue of the writs for the election ("the grace period") continuously since 1983, with one exception. The Electoral Act was amended in 2006 so as to remove the grace period for new enrolments and to limit the grace period to only three days for transfers of enrolment. The provisions effecting those changes were held to be invalid by a majority of the Court in Rowe v Electoral Commissioner[87] and the changes were subsequently removed. The position now is that which was maintained prior to Rowe.
[87](2010) 243 CLR 1; [2010] HCA 46.
Standing
The first plaintiff was in fact enrolled to vote at a federal election. The Commonwealth challenged his standing to maintain these proceedings on the basis that he did not seek to have his rights or interests clarified by the orders he sought[88]. The second plaintiff was subsequently joined to the proceedings. She intended to nominate as a candidate for election to the House of Representatives. Section 166 of the Electoral Act requires a candidate in her position to be nominated by 100 persons whose names appear on the Electoral Rolls and are entitled to vote at the election for which the candidate is nominated. This provided the second plaintiff with a sufficient interest in the matter before the Court for standing, as the Commonwealth conceded. It is not necessary therefore to address the position of the first plaintiff.
[88]See Croome v Tasmania (1997) 191 CLR 119 at 127; [1997] HCA 5.
The plaintiffs' argument
The plaintiffs relied upon what they described as a principle established in Roach v Electoral Commissioner[89] and Rowe, that a law which disqualifies a person from exercising the choice mandated by the Constitution is invalid unless the disqualification is for a substantial reason. It will be for a substantial reason only if it is "reasonably appropriate and adapted" to secure an end which is consistent and compatible with the maintenance of the constitutionally prescribed system of representative government. The plaintiffs sought to incorporate into this test of proportionality those tests stated in McCloy v New South Wales[90] as tools of analysis used in that case with respect to legislation which burdens the implied freedom of political communication.
[89](2007) 233 CLR 162; [2007] HCA 43.
[90](2015) 89 ALJR 857; 325 ALR 15; [2015] HCA 34.
The larger part of the plaintiffs' submissions in this matter were concerned with testing for proportionality. Their principal submission in this regard was that there are reasonably practicable alternative means available which would provide for a longer period of enrolment. In the first place, they contended that there is no reason why a person should not be permitted to enrol up to and including polling day. This could be achieved given advances in technology and without the requirement of substantial further resources. The second alternative was to calculate a suspension period back from polling day, rather than forward from the date of the issue of the writs. New South Wales and Victoria permit enrolment up to and including polling day; and in Queensland a person can enrol up to the night before polling day. It is to be inferred that the essential premise of the plaintiffs' argument is that legislation will not be valid unless it ensures that the maximum number of people are able to vote at elections.
Roach and Rowe
The concentration of the plaintiffs' submissions on proportionality testing reflects their assumption that this case is not materially different from the circumstances in Roach and Rowe. That assumption is not correct. Roach was concerned with legislation which disqualified prisoners from voting at federal elections. This case has a closer affinity to Rowe, but only by degree and without the particular aspect of the legislation there in question. As mentioned earlier, the legislative provisions in Rowe removed or substantially limited the grace period which had been in place since 1983, and the basis for its removal was contentious.
It must, however, be acknowledged that Roach and Rowe effected something of a turning in the law. Sections 7 and 24, in their reference to "chosen by the people", had been understood to refer to direct and popular choice. It had been accepted that the Constitution left it to the Parliament to legislate with respect to our system of representative government. The ability of Parliament to legislate necessarily extends to questions such as who is qualified to be elected, who may be the electors and a system for conducting elections. These matters are reflected in the provisions of the Electoral Act.
The majority judgments in Roach, upon which the majority judgments in Rowe were largely founded, recognised a limitation on legislative power with respect to the eligibility of persons to vote. The limitation was founded upon perceptions about the franchise, as being held generally by all adults and, implicitly, entrenched as such in conceptions of representative government[91].
[91]Roach v Electoral Commissioner (2007) 233 CLR 162 at 173-174 [6]-[7] per Gleeson CJ, 198 [80], [83], 199-200 [86], 202 [95] per Gummow, Kirby and Crennan JJ.
The arguments of the parties in Rowe applied the joint judgment in Roach and addressed two questions. The first was whether the provisions in question disentitle, exclude or disqualify any citizen otherwise entitled to vote from voting. The second was whether any such disqualification was for a substantial reason or is disproportionate[92]. It was to those two topics that my reasons, in dissent, in Rowe were directed.
[92]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 4-8.
In Rowe, no disqualification or disenfranchisement such as that dealt with in Roach was in issue. However, the majority in Rowe proceeded upon the basis that a law which permits little time after the writs for the election have issued to enrol or transfer enrolment is also subject to the requirement that it be justified. Such a law was considered effectively to disentitle[93], disqualify[94] or exclude[95] persons otherwise qualified to be enrolled as electors from voting.
[93]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 12 [3] per French CJ, 119 [381], 121 [384] per Crennan J.
[94]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 58 [160], 61 [167] per Gummow and Bell JJ.
[95]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 119 [381] per Crennan J.
In Rowe the justification for the changes to the Electoral Act was said by the Commonwealth to be found largely in a report of the Joint Standing Committee on Electoral Matters with respect to the 2004 federal election[96]. However, the majority in Rowe did not consider that the laws were proportionate. French CJ[97] said that if a law's adverse legal or practical effect upon the exercise of the entitlement to vote is disproportionate to its advancement of the constitutional mandate, it may be invalid. His Honour did not accept some of the practical reasons put forward in the report as to why the earlier cut off had to be made, such as dealing with electoral fraud or encouraging electors to enrol or apply in time[98], and considered that the "heavy price" imposed by the amendments was disproportionate to the benefits of a smoother and more efficient electoral system which the amendments sought to achieve[99]. Gummow and Bell JJ also did not accept that the purpose of preventing systemic fraud, put forward as justification for the legislation, supplied a substantial reason for disqualifying a large number of electors and held that this disqualification went beyond any advantage to be gained to the integrity of the electoral process[100]. Crennan J did not consider the provisions to be either necessary or appropriate for the protection of the integrity of the Rolls[101]. There was no substantial reason for the amendments, in her Honour's view.
[96]Australia, The Parliament, Joint Standing Committee on Electoral Matters, The 2004 Federal Election: Report of the Inquiry into the Conduct of the 2004 Federal Election and Matters Related Thereto, (2005).
[97]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 12 [2].
[98]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 38 [75].
[99]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 38-39 [78].
[100]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 61 [167].
[101]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 120-121 [384].
A further observation may also be made about Roach and Rowe, as relevant to the plaintiffs' argument. Neither of those decisions is authority for the proposition that it is a constitutional imperative that the maximum number of persons entitled to be enrolled and vote have the opportunity to be enrolled and vote.
Applying Roach and Rowe
The Commonwealth did not submit in these proceedings that Roach and Rowe were wrongly decided. This Court was asked to proceed upon the basis that they are to be applied.
The Commonwealth submitted that the first requirement of these decisions was not met. There was no disqualification because persons eligible to enrol were entitled to enrol before and during the grace period; they had a duty to do so; and they had been reminded by the announcement of the election to do so. Whilst I am attracted to such a submission – it reflects the view I took in Rowe[102] – it seems to me to repeat an argument which was implicitly rejected by the majority in Rowe, which considered that any detriment to, or burdening of, the ability to vote had to be justified. This is the topic to which I should now turn.
[102]Rowe v Electoral Commissioner (2010) 243 CLR 1 at 128 [411].
Justification?
The joint reasons in Roach required that there be a substantial reason for provisions which effect a disqualification from the entitlement to vote and that that requirement would be satisfied if the means adopted were not disproportionate to the legitimate end which they sought to achieve[103]. That is to say, a "substantial reason" implies something more than the means being for a legitimate purpose. Even so, to state that there is a substantial reason for a statutory provision, without more, is to state a conclusion in which there inheres a value judgment in respect of which no reasoning is exposed.
[103]Roach v Electoral Commissioner (2007) 233 CLR 162 at 199 [85].
Since the decisions in Roach and Rowe, this Court has confirmed the usefulness of traditional proportionality tests as tools of analysis in determining the limits of legislative power where legislation has the effect of restricting the implied freedom of political communication[104]. Amongst the reasons given for the adoption of that approach were that the question of justification should not turn on a matter of impression and that it should not be stated as a mere conclusion of proportionality, absent reasons[105].
[104]McCloy v New South Wales (2015) 89 ALJR 857; 325 ALR 15.
[105]McCloy v New South Wales (2015) 89 ALJR 857 at 875 [75], [77]; 325 ALR 15 at 35.
The Commonwealth and South Australia, intervening to support the Commonwealth, made the obvious point that this case does not involve legislation affecting the implied freedom. On the other hand, the effect of the reasoning in Roach is that legislative power which burdens the franchise is to be justified. The reasoning applied in the joint judgment in Roach was based upon notions of representative democracy derived from ss 7 and 24 of the Constitution which had been employed in Lange v Australian Broadcasting Corporation[106]. The requirement of "proportionality", which was considered in Roach[107] to inhere in the question whether the disqualification of electors was for a "substantial reason", was said to have an "affinity" to what is called the second question in Lange[108]. And of course it was by reference to the questions posed by Lange that proportionality testing came to be applied, to an extent in Unions NSW v New South Wales[109], and then more fully in McCloy.
[106](1997) 189 CLR 520; [1997] HCA 25.
[107]Roach v Electoral Commissioner (2007) 233 CLR 162 at 199 [85].
[108]Roach v Electoral Commissioner (2007) 233 CLR 162 at 199 [86].
[109](2013) 252 CLR 530; [2013] HCA 58.
The alternative to utilising the tests stated in McCloy in order to conclude whether the provisions here were made for a "substantial reason" would be to ask whether they are "reasonably appropriate and adapted" to their legitimate end. This is a test more commonly utilised with respect to determining whether legislation is within a purposive head of power, but it has also been described and used as a kind of proportionality test for laws which affect constitutionally guaranteed freedoms. However, it does not identify any method by which one is to reason to a conclusion that a law is "reasonably appropriate and adapted". It must necessarily involve value judgments, as any form of proportionality testing must do, but, absent a method, such judgments and the method of reasoning are not required to be exposed by this test.
The aim of any testing for proportionality is to ascertain the rationality and reasonableness of a legislative restriction[110] in a circumstance where it is recognised that there are limits to legislative power. Proportionality analysis does not involve determining policy or fiscal choices, which are the province of the Parliament. Thus the test of whether there are alternative, less restrictive means available for achieving a statutory object, which assumes some importance in this case, requires that the alternative measure be otherwise identical in its effects to the legislative measures which have been chosen. It will not be equal in every respect if it requires not insignificant government funding[111].
[110]McCloy v New South Wales (2015) 89 ALJR 857 at 873 [68]; 325 ALR 15 at 33.
[111]Barak, Proportionality: Constitutional Rights and their Limitations, (2012) at 324-325.
The Roll is at the centre of the Commonwealth electoral system. Electors are defined as those persons whose names appear on the Roll[112]. Enrolment is the single qualification for voting on polling day, or by pre-polling or postal vote. Claims to enrolment are required to be made within a short time of a person becoming eligible for enrolment, on pain of penalty[113]. Claims for enrolment are investigated and determined[114] and objections to enrolment dealt with[115].
[112]Commonwealth Electoral Act 1918 (Cth), s 4(1).
[113]Commonwealth Electoral Act 1918 (Cth), s 101.
[114]Commonwealth Electoral Act 1918 (Cth), s 102.
[115]Commonwealth Electoral Act 1918 (Cth), ss 113-118.
The Electoral Commissioner is obliged to prepare a certified list of persons who are entitled to vote for each Division and to deliver it to the presiding officer at each polling place before the commencement of voting[116]. The Commissioner may also provide an "approved list", which is a list in electronic form containing the same information as the most recent certified list[117]. A "Notebook Roll" is also maintained by the Australian Electoral Commission. It is not required by the Electoral Act. It is used in practice to deal with applications for enrolment or changes to the Roll received before the suspension period but which have not been processed before the certified lists have been printed. It facilitates the scrutiny of provisional votes, which can be made where a person's name or address does not appear on the certified lists but the person declares that they are enrolled[118].
[116]Commonwealth Electoral Act 1918 (Cth), s 208.
[117]Commonwealth Electoral Act 1918 (Cth), s 208A.
[118]Commonwealth Electoral Act 1918 (Cth), s 235.
It is in this scheme that the closure of the Roll is to be considered. The Roll is closed seven days after the writs for election issue and shortly before the first step in the electoral process (nomination) occurs[119].
[119]Commonwealth Electoral Act 1918 (Cth), ss 155, 156.
In the scheme established by the Electoral Act it is evident that there are practical reasons why the Roll is closed to further enrolment or transfer. It enables a number of steps to be undertaken to facilitate the efficient conduct of an election. The closure of the Roll prior to the commencement of the election process ensures there will be few delays in declaring election results, which would occur if persons were entitled to enrol at a point close to, or on, polling day. It achieves accuracy and certainty in the lists which are able to be produced when polling does take place. The provisions for the closure of the Roll bear a rational connection to their purposes[120].
[120]Unions NSW v New South Wales (2013) 252 CLR 530.
It is appropriate to consider alternative statutory schemes in order to determine whether there are alternative, equally practicable means of achieving these purposes without closing the Rolls or closing them at the point presently provided for. This was the approach taken by the Court in Betfair Pty Ltd v Western Australia[121], where Tasmanian legislation on the same topic as the impugned Western Australian legislation showed that the legislative purpose could be achieved without restricting the freedom of interstate trade and commerce guaranteed by s 92 of the Constitution.
[121](2008) 234 CLR 418; [2008] HCA 11.
Until relatively recently, all of the States and Territories followed the Commonwealth model, allowing for the closure of the Rolls some days after the issue of the writs for the election, and this closure had the effect of suspending the processing of claims for new enrolments and transfer of enrolments. Legislation in Queensland and Victoria continues to provide that the Rolls close a specified number of days after the writs are issued[122]. However, in Queensland, a person who is not enrolled may enrol or transfer their enrolment up to 6.00pm on the day before polling day if they are entitled to be, but are not, enrolled. Such persons must make a declaration vote at the polling booth[123]. In Victoria a person who is not enrolled on polling day, but claims to be entitled to be enrolled, may make a provisional vote despite the fact that the Rolls are "closed"[124]. The New South Wales legislation goes further: under it, the Roll does not close and the entitlement to vote does not depend upon enrolment[125]. In each of these States, the Roll is not in any final form at the conclusion of polling and claims to enrolment made in connection with a declaration or provisional vote must be determined after polling day and before the election result is announced.
[122]Electoral Act 2002 (Vic), s 63(3); Electoral Act 1992 (Q), s 84(1)(b).
[123]Electoral Act 1992 (Q), ss 106(1)(d), 115(d).
[124]Electoral Act 2002 (Vic), s 108; Electoral Regulations 2012 (Vic), reg 41.
[125]Parliamentary Electorates and Elections Act 1912 (NSW), s 106(2A).
Reference to these other statutory schemes shows that there is more than one electoral system to choose from. It does not show that these systems are capable of achieving the same objectives that the Electoral Act does. They reflect policy choices of those States' legislatures, which no doubt balance the delays and costs, which must inevitably be associated with keeping the Roll open, against other objectives.
It cannot be concluded that the systems chosen elsewhere would be as efficient as that which operates under the Electoral Act, with the level of certainty which is achieved by the closure of the Rolls and the suspension period. Nor can it be concluded on the facts available that adopting such a system would not require additional resources – for example, by way of staff and computer equipment – and therefore further funding by government. The alternative systems cannot therefore be said to be equally practicable.
Finally, it may be observed that the effect of the suspension period in closing the Rolls for enrolments and transfers is balanced by the certainty and efficiencies which are achieved. In this regard it is appropriate to take into account that many, if not most, of the persons who will be unable to enrol are already in breach of their obligation to do so and that persons who have transferred to, but are not yet enrolled in, another Division retain the ability to vote in their former Division[126].
[126]Commonwealth Electoral Act 1918 (Cth), ss 229, 231, 235.
GAGELER J. Professor Albert Venn Dicey saw the Australian Constitution as rigid, but noted that its rigidity is tempered in three ways. One is the "very wide legislative authority" with which the Commonwealth Parliament is endowed. Another is that the Constitution provides the means for its own alteration. The other is that a large number of provisions of the Constitution remain in force only "until the Parliament otherwise provides": "they can therefore be changed like any other law by an Act of Parliament passed in the ordinary manner; in other words, the constitution is as to many of its provisions flexible"[127].
[127]Dicey, Introduction to the Study of the Law of the Constitution, 6th ed (1902) at 481-482. See now Allison (ed), The Law of the Constitution by A V Dicey, (2013) at 333.
The flexibility to which Professor Dicey drew attention is not unbounded. It is expressly constrained by the terms in which s 51(xxxvi) of the Constitution confers authority on the Parliament to make laws with respect to "matters in respect of which this Constitution makes provision until the Parliament otherwise provides". Like other conferrals of legislative authority, that authority is conferred "subject to this Constitution".
This special case was concerned with three matters falling within the scope of the authority conferred by s 51(xxxvi). One matter, in respect of which provision is made in s 30, is the qualification of electors of members of the House of Representatives. Provision in respect of that matter has a consequential effect through s 8, which makes the qualification of electors of senators the same as the qualification of electors of members of the House of Representatives. The other matters are the method of election of senators, for which provision is made in s 9, and the method of election of members of the House of Representatives, for which provision is made in s 31. Those other matters overlap in practice with the first; they always have.
Legislation enacted by the Parliament under s 51(xxxvi) has always provided that, in order to vote at an election for the Senate or the House of Representatives, a person must be enrolled on the Electoral Roll for an Electoral Division. That was so under the Commonwealth Electoral Act 1902 (Cth) and remained so under the Commonwealth Electoral Act 1918 (Cth) ("the Act").
Enrolment on the Electoral Roll for an Electoral Division under the Act involves the taking of administrative action to enter a person's name on that Roll[128]. Taking that administrative action ordinarily depends on that person making a claim for enrolment or for transfer of enrolment[129]. The person is obliged to make such a claim forthwith on becoming entitled to do so as a consequence of meeting pre-requisites as to age, Australian citizenship, and residence in (or other relevant connection to) a Subdivision of the Electoral Division[130]. Only a person whose name appears on a Roll meets the statutory definition of an "elector"[131] and only a person who is an elector has an entitlement to vote at an election[132]. The result is that "[e]nrolment is not merely evidence of an elector's qualification to vote; enrolment is itself a qualification to vote"[133].
[128]Section 102.
[129]Sections 98(1), 99(1) and (2).
[130]Sections 93 and 101(1).
[131]Section 4(1).
[132]Section 93(2).
[133]Muldowney v Australian Electoral Commission (1993) 178 CLR 34 at 40; [1993] HCA 32.
The scheme of the Act is to make the close of the Rolls on a fixed date a step in the conduct of an election for the Senate or the House of Representatives. Together with dates for nomination, polling and return, a date for the close of the Rolls must be fixed by the writ for an election[134]. The date fixed by the writ for the close of the Rolls must be the seventh day after the issue of the writ[135]. Only persons whose names appear on the Rolls as electors at the close of the Rolls can vote in the election whether on polling[136] or by pre-poll[137] or postal vote[138]. In the conduct of the polling, the Rolls are conclusive evidence of the right of each person enrolled as an elector to be admitted to vote[139]. The Court of Disputed Returns, if required to adjudicate a dispute about the election following the return of the writ, must assume the Roll to be correct and is prevented from inquiring into its correctness[140]: the Court cannot "go behind the Electoral Roll and determine whether a person who is not on the roll is none the less entitled to be on the roll"[141].
[134]Section 152(1).
[135]Section 155.
[136]Part XVI.
[137]Part XVA.
[138]Part XV.
[139]Section 221.
[140]Section 361.
[141]Snowdon v Dondas (1996) 188 CLR 48 at 75; [1996] HCA 27, quoting Re Brennan; Ex parte Muldowney (1993) 67 ALJR 837 at 840; 116 ALR 619 at 623; [1993] HCA 53.
The issue for substantive determination in the special case concerned the validity of the currently legislated time for cutting off enrolment before an election. That time is 8.00pm on the date fixed for the close of the Rolls. The cut-off is implemented by provisions having the effect of suspending all administrative action directed to enrolment from that time until the close of the poll for the election[142] with the result that claims for enrolment or transfer made during that period are not to be processed until after the election. Those provisions, which the plaintiffs impugned, were inserted into the Act in 2011 and 2012[143].
[142]Sections 94A(4), 95(4), 96(4), 102(4), 103A(5), 103B(5) and 118(5).
[143]Electoral and Referendum Amendment (Enrolment and Prisoner Voting) Act 2011 (Cth); Electoral and Referendum Amendment (Maintaining Address) Act 2012 (Cth); Electoral and Referendum Amendment (Protecting Elector Participation) Act 2012 (Cth).
Stripped of distractions of detail and ignoring some vagueness of temporal focus, the plaintiffs' essential argument was that cutting off enrolment before an election at 8.00pm on the date fixed for the close of the Rolls was not within the authority conferred on the Parliament by s 51(xxxvi). That timing of the cut-off was said to be repugnant to the requirements of ss 7 and 24 of the Constitution that senators and members of the House of Representatives be "directly chosen by the people". The repugnance was argued to lie in the cut-off of enrolment at that time resulting in the practical exclusion from voting of some people entitled to enrolment without substantial justification.
Unable to accept that argument, I joined in formally answering the substantive questions posed in the special case to the effect that the provisions imposing the cut-off are not repugnant to ss 7 and 24 of the Constitution. My reasons are as follows.
The major premise of the plaintiffs' argument must be accepted; acceptance is compelled by unchallenged authority. The view that the constitutional phrase "directly chosen by the people" might import "some requirement of a franchise that is held generally by all adults or all adult citizens unless there be substantial reasons for excluding them" was canvassed by Brennan CJ in McGinty v Western Australia[144]. The view had been foreshadowed by three members of the Court in Attorney-General (Cth); Ex rel McKinlay v The Commonwealth[145]. The view was subsequently taken up and acted upon by differently constituted majorities of four members of the Court in Roach v Electoral Commissioner[146] and in Rowe v Electoral Commissioner[147], in each case holding provisions of the Act repugnant to the requirements of ss 7 and 24 and therefore beyond the authority conferred on the Parliament by s 51(xxxvi) of the Constitution.
[144](1996) 186 CLR 140 at 170; [1996] HCA 48.
[145](1975) 135 CLR 1 at 36, 69; [1975] HCA 53.
[146](2007) 233 CLR 162 at 174 [7], 182 [23], 198-199 [83], 199-200 [85]-[86]; [2007] HCA 43.
[147](2010) 243 CLR 1 at 19-21 [23]-[26], 56-62 [150]-[168], 118-121 [372]-[385]; [2010] HCA 46.
Second, the Rolls are available to be used by others. A copy of a Roll for a Division or each State and Territory is available for public inspection[369]. Information in relation to the Rolls and certified lists of voters also must be provided by the Electoral Commission to particular persons and organisations[370]. In particular, "as soon as practicable after the close of the Rolls", candidates in elections for the House of Representatives must receive a copy of the certified list of voters for the Division for which they are seeking election[371].
[369]s 90A of the Act.
[370]s 90B of the Act.
[371]Item 1 in the table in s 90B(1) of the Act.
The system
As the above discussion demonstrates, the electoral system chosen by the Parliament has a detailed, coherent structure. That system includes practical and logical steps directed to the orderly and efficient conduct of elections, which result in senators and members of the House of Representatives being "directly chosen by the people" as required by ss 7 and 24 of the Constitution, leading to the formation of a government.
It is against that background that the plaintiffs' complaints are to be considered.
The plaintiffs' complaints
The plaintiffs complained about the suspension period effected by ss 94A(4), 95(4), 96(4), 102(4), 103A(5), 103B(5) and 118(5) of the Act ("the impugned provisions"). The plaintiffs contended that the text and structure of the Constitution prohibit the suspension period[372] because the impugned provisions prevent persons otherwise entitled to do so from enrolling (or transferring their enrolment) and voting on polling day, thereby effecting a legislative disenfranchisement, disqualification or exclusion of persons otherwise entitled to vote. The plaintiffs also contended that the impugned provisions distort the popular choice mandated by the Constitution.
[372]Lange (1997) 189 CLR 520 at 567.
Limit on Commonwealth legislative power
Whatever choice is made by the legislature about the qualification to enrol, the entitlement to vote and the method of voting, universal adult suffrage may only be subject to a limitation for a substantial reason[373].
[373]Roach (2007) 233 CLR 162 at 174 [7], 198-199 [83]; cf McGinty (1996) 186 CLR 140 at 170 as discussed in Twomey, "Rowe v Electoral Commissioner – Evolution or Creationism?", (2012) 31 University of Queensland Law Journal 181 at 186.
In Roach, the Court dealt with a case where the Parliament, in exercise of its legislative power to determine the qualification of electors[374], had enacted provisions that had a legal effect on the franchise by prohibiting persons serving a term of imprisonment from voting. In Rowe, although the provisions challenged were not directed towards the qualification of electors, they had a substantial practical effect on who participated in the "choice". The provisions challenged in both cases were held invalid. The basis for that invalidity was that the provisions, in the context of the electoral system as a whole, had the effect that the electoral system established by the Act did not provide for senators and members of the House of Representatives to be "directly chosen by the people" as required by ss 7 and 24 of the Constitution[375].
[374]ss 8, 30 and 51(xxxvi) of the Constitution.
[375]See Roach (2007) 233 CLR 162 at 173 [6], 182 [24], 187-188 [46]-[49], 199-200 [85]-[86], 202 [95]; Rowe (2010) 243 CLR 1 at 19 [22], 38-39 [78], 56-57 [150]‑[151], 58-59 [160], 61 [166], 107 [328], 118-119 [375], 119 [381], 121 [384].
In Roach, Gummow, Kirby and Crennan JJ identified that there will be a "substantial" reason for a restriction on the franchise if the reason is "reasonably appropriate and adapted to serve an end which is consistent or compatible with the maintenance of the constitutionally prescribed system of representative government"[376]. Their Honours noted that "[t]he affinity to what is called the second question in Lange will be apparent"[377].
[376](2007) 233 CLR 162 at 199 [85].
[377](2007) 233 CLR 162 at 199 [86] (footnote omitted).
The questions in Lange were directed to whether a law will be invalid for infringing the implied freedom of political communication. More recently, in McCloy v New South Wales[378], a majority of the Court altered the traditional formulation of that test and adopted a framework of "structured" proportionality.
[378](2015) 89 ALJR 857; 325 ALR 15; [2015] HCA 34.
Here, the plaintiffs proceeded on the basis that, because of the "affinity" between the test outlined in Roach and the second question in Lange, the validity of the impugned provisions fell to be determined in accordance with the "structured" proportionality approach of the joint judgment in McCloy. The Commonwealth, while accepting that some form of proportionality testing was appropriate, rejected the suggestion that it should take the form adopted by the joint judgment in McCloy. The Attorney-General for South Australia, who intervened to make submissions only on the issue of the relevant test, supported the Commonwealth's approach.
The concept of proportionality is applied in a variety of areas in Australian jurisprudence[379]. It should not be assumed that, because a particular test for proportionality has been adopted in one particular constitutional context, it can be uncritically transferred into another context, constitutional or otherwise[380], even within the same jurisdiction.
[379]See McCloy (2015) 89 ALJR 857 at 863 [3]; 325 ALR 15 at 19.
[380]See McCloy (2015) 89 ALJR 857 at 874 [72]; 325 ALR 15 at 34. See also Mulholland (2004) 220 CLR 181 at 200 [39]; Roach (2007) 233 CLR 162 at 178‑179 [17]; McCloy (2015) 89 ALJR 857 at 885 [139], 917 [339]; 325 ALR 15 at 49, 91-92.
The "structured" proportionality approach adopted by the joint judgment in McCloy is inappropriate in the constitutional context in this case. That can be demonstrated by considering the "necessity" stage of the McCloy test.
The "necessity" stage of the McCloy test would require a court to inquire as to whether there exists an "obvious and compelling alternative, reasonably practicable means of achieving the same purpose which has a less restrictive effect"[381] on the franchise. If that inquiry is answered positively, then a law will (not may) be invalid[382].
[381]McCloy (2015) 89 ALJR 857 at 863 [2]; 325 ALR 15 at 19.
[382]cf McCloy (2015) 89 ALJR 857 at 885 [135], 899 [222], 915‑916 [328]; 325 ALR 15 at 48, 68, 89.
There are questions about whether such a rigid inquiry (which alone may result in the invalidity of legislation) is appropriate at all in the Australian constitutional context, where the judicial branch of government cannot exercise legislative or executive power.
However, for present purposes, it is enough to consider the immediate constitutional context. There is a critical difference between the implied freedom of political communication considered in McCloy and the issues in this case.
The Parliament is required to enact laws which provide for an electoral system[383]. The consequence of the constitutional framework is that the Parliament is effectively under an obligation to maintain laws of that kind. The legislature must design and maintain a comprehensive system that meets the constitutional mandate. In designing that system, the legislature will be faced with a multitude of options for how the system is to operate and, in doing so, will be required to balance a wide range of matters and values. Those options will be presented at a number of different points as the Parliament addresses many different stages of the electoral process. Each of those different stages must ultimately interact to create a coherent whole. It is not a matter of devising and drafting a single provision or division of an Act. There are countless variations of such a system.
[383]See [261]-[264] above. cf Barak, Proportionality: Constitutional Rights and their Limitations, (2012) at 400.
That positive role given to Parliament, together with the broad scope of the legislative power with respect to elections, distinctly marks out the present constitutional context from any inquiry about the implied freedom of political communication.
And that is also why, in any case, it is not appropriate to apply the "necessity" stage of the McCloy test as rigidly as McCloy would suggest. To do so would create too great a risk of the judicial branch intruding on the legislative function conferred on the Parliament by the Constitution[384]. And at a practical level, the judiciary is not equipped to make definitive judgments about whether there are obvious, compelling and practical alternatives to particular provisions that are part of an entire legislative scheme that the Parliament is required to enact to comply with ss 7 and 24 of the Constitution. That flows from the distinctions that are "to be maintained between powers described as legislative, executive and judicial" by reference to "distinctions generally accepted at the time when the Constitution was framed between classes of powers requiring different 'skills and professional habits' in the authorities entrusted with their exercise"[385].
[384]See Castlemaine Tooheys Ltd v South Australia (1990) 169 CLR 436 at 473; [1990] HCA 1; Unions NSWv New South Wales (2013) 252 CLR 530 at 576-577 [130]-[132]; [2013] HCA 58.
[385]R v Davison (1954) 90 CLR 353 at 381-382; [1954] HCA 46.
The difficulties in assessing an integrated legislative scheme were demonstrated in this case when considering the electoral systems in New South Wales, Queensland and Victoria. While they demonstrate that alternative schemes do exist, upon close examination the difficulties in making effective comparisons become apparent. Those difficulties would only be heightened by considering hypothetical alternatives. Those observations are not to endorse particular labels such as "deference", "margin of appreciation" or "zone of proportionality". Rather, they are observations grounded in, and derived from, fundamental constitutional principle in this country.
None of that denies that comparisons with alternative laws may be instructive[386]. But they are not determinative. That accords with the earlier warning that "care is called for in elevating a 'direct choice' principle to a broad restraint upon legislative development of the federal system of representative government"[387].
[386]See Lange (1997) 189 CLR 520 at 568; Unions NSW (2013) 252 CLR 530 at 556 [44].
[387]Mulholland (2004) 220 CLR 181 at 237 [156]; Roach (2007) 233 CLR 162 at 197 [77]; Day (2016) 90 ALJR 639 at 645 [19]; 331 ALR 386 at 393.
Impugned provisions are valid
Turning to the present case, is part of the universal adult suffrage excluded or the franchise restricted by the impugned provisions, which form part of the electoral system prescribed by the Act? If so, is there a "substantial reason" for that exclusion or restriction?
Terms such as "exclusion", "detriment", "disentitlement", "disenfranchisement" and "disqualification" were used in the different reasons for judgment in Roach and Rowe. Some of these terms are unhelpful. Most are not defined or not defined consistently. To take "disqualification" as an example, it may mask a distinction between legal disqualification in the Roach sense and cases such as the present where persons who are qualified to vote are unable to vote for some other reason[388]. Here, the plaintiffs contended that there was a restriction on, exclusion from, or distortion of the franchise.
[388]See Rowe (2010) 243 CLR 1 at 66 [187].
No restriction on, or exclusion from, the franchise
Is there a relevant restriction on, or exclusion from, the franchise in this case? The answer is "No".
First, there was no legislative diminution of an existing opportunity to enrol, transfer or vote. That is not to endorse the legislative diminution of opportunities to enrol, transfer or vote as being a legitimate criterion to determine the constitutional validity of legislation affecting the franchise. As a matter of constitutional interpretation, treating legislative diminution as a criterion raises complicated issues[389]. It is enough to note that when the grace period was restored by the decision in Rowe and reflected in the enactment of the impugned provisions by Parliament in 2011, that amounted to a restoration of the opportunity to enrol, transfer or vote to the position that first existed in 1983.
[389]See Rowe (2010) 243 CLR 1 at 102 [310]-[311]. See also Twomey, "Rowe v Electoral Commissioner – Evolution or Creationism?", (2012) 31 University of Queensland Law Journal 181 at 183-195.
The plaintiffs' reliance on Rowe to contend that there was a restriction on the franchise in the present case was misplaced. The outcome in Rowe does not dictate the answer to the present case. There was no consensus between the majority in Rowe on this issue. For example, French CJ placed significance on the fact that the impugned provisions in Rowe operated to diminish opportunities to enrol and transfer that had previously existed under the Act[390]. That fact was not determinative for the other members of the majority.
[390](2010) 243 CLR 1 at 38-39 [78].
In numerical terms, the question of whether there is a relevant restriction on, or exclusion from, the franchise might be said to be answered by the facts agreed in the special case. For each of the four federal elections prior to the 2016 election, there has been an increase in the number of provisional votes cast. Such a vote may be cast in certain circumstances where a person claims to vote, but there is a specific reason why their entitlement to vote cannot be verified at the time[391]. In the 2004 election, 180,865 provisional votes were cast, with 112,560 counted for the Senate and 90,512 counted for the House of Representatives. In the 2013 election, 202,246 provisional votes were cast, with 95,619 counted for the Senate and 48,299 counted for the House of Representatives. The principal explanation for the difference between the votes in fact counted for the Senate and the House of Representatives in each election is that the Senate vote was accepted and counted, whereas the House of Representatives vote was rejected, when an enrolment was valid for the State but not for the Division.
[391]s 235 of the Act.
In addition, in each of the four federal elections prior to the 2016 election, people lodged claims for enrolment or updates of enrolment during the suspension period[392]. In 2004, 164,037 people lodged claims: 60,584 were new enrolments and 103,453 were to update enrolment. At the close of the Rolls, approximately 1.2 million people were eligible to be enrolled but were not enrolled. In 2013, 228,585 people lodged claims: 52,692 were new enrolments and 175,893 were to update enrolment. At the close of the Rolls, 1,212,616 people were eligible to be enrolled but were not enrolled. It can be inferred that, in each of these cases, a number of people who were probably qualified to vote, but not enrolled to vote, were ineligible to cast a vote or their vote was not counted for one or both of the Senate and the House of Representatives.
[392]It can be noted that "update" in this context includes any changes to a person's enrolment details, and is not limited to a change of the person's address.
None of these facts taken singularly or collectively provides an answer to the relevant constitutional question. To treat them as doing so would overlook both the nature of the limitation imposed by ss 7 and 24 of the Constitution, and fundamental components of the electoral system provided for by the Act.
The electoral system provided for by the Parliament must result in senators and members of the House of Representatives being "directly chosen by the people". Numbers alone cannot establish that senators or members have not been chosen in accordance with that constitutional mandate.
That proposition may be tested in this way. Many people who are qualified to enrol fail to do so. Those numbers are not insignificant. Others are enrolled, but do not vote[393]. No one suggests that even if a large proportion of the enrolled population decided not to vote at a particular election the result of that election would not "yield" Houses of Parliament constituted by senators and members directly chosen by the "people"[394].
[393]Rowe (2010) 243 CLR 1 at 96 [288].
[394]Rowe (2010) 243 CLR 1 at 64 [182].
These observations expose a fundamental flaw in the plaintiffs' argument. The plaintiffs assumed that the Constitution requires the Parliament to enact legislation to provide for "elections as expressive of the popular choice as practical considerations properly permit"[395]. That assumption cannot be sustained. It finds no support or foundation in the text or structure of the Constitution, or elsewhere[396].
[395]Rowe (2010) 243 CLR 1 at 57 [154].
[396]Rowe (2010) 243 CLR 1 at 73 [210], 75-76 [220].
Further, the plaintiffs' numerical argument proceeded on the basis that polling day has constitutional significance. According to the plaintiffs, polling day is when the people "choose". That proposition does not emerge from the text or structure of the Constitution. Moreover, the contention is inconsistent with the interpretation of the phrase "shall be incapable of being chosen" in s 44 of the Constitution, which has been held to refer to "the process of being chosen"[397]. Section 44 concerns the disqualification of persons from being "chosen" as senators or members. In that way, it is directly linked to ss 7 and 24. There is no reason to think the word "chosen" has a different meaning in the different sections.
[397]Sykes v Cleary (1992) 176 CLR 77 at 99-100; [1992] HCA 60. See also Langer (1996) 186 CLR 302 at 332-333. cf Rowe (2010) 243 CLR 1 at 58-59 [160].
Accordingly, the numbers are not an answer because whether senators and members are "directly chosen by the people" necessarily requires consideration of the electoral system provided for by the Act, which extends to and includes the franchise, qualification to enrol to vote, entitlement to vote, the process for electing senators and members and the exercise of the right of an elector to vote, ultimately leading to the formation of a government.
Moreover, the choice made by the electors is not one made only on the day of the poll. Every person qualified to enrol for a Division, either by way of enrolment or transfer, and whose name is not on the Roll, is obliged to make a claim for enrolment or transfer of enrolment under the Act[398]. A person whose name is not on the Roll within 21 days from the date of entitlement is guilty of an offence[399]. That obligation is not tied to polling day. Moreover, electors can and do vote prior to polling day. Polling day is simply the last possible time a qualified elector can vote.
[398]s 101(1) of the Act.
[399]s 101(4) and (6) of the Act.
Finally, the suspension period does not prevent any group of electors from voting. In contrast to Roach, the impugned provisions are not directed to the qualification of electors. Generally, a person who cannot vote on polling day because of the suspension period could have voted if they had complied with their statutory obligations[400].
[400]Subject to those whose eligibility to vote only arises during the suspension period.
It is simply not possible or appropriate to isolate one aspect of the system (the suspension period) and contend that, considered in isolation from the rest of the system, it results in the exclusion of part of universal adult suffrage or restriction on the franchise and, instead, the system must enable persons to enrol and vote on polling day. There is no relevant restriction on, or exclusion from, the franchise. The elected senators and members can still be said to be directly chosen by the "people".
No distortion
The plaintiffs also contended that because of the requirement that senators are to be chosen "for each State" by the "people of the State" and that members are to be chosen by the "people of the Commonwealth", the impugned provisions effected a distortion of the popular choice. The distortion was said to arise because the franchise must be geographically perfect in the sense that a person can only vote in the Division and State in which they are resident.
That argument wrongly assumed that the Constitution dictates that a person can only be considered to be a person of a State or of the Commonwealth if they have a particular geographic connection. There is no basis for that assumption in the text or structure of the Constitution. Indeed, if the plaintiffs' argument was correct (and it is not), it would render unconstitutional those provisions which permit persons to vote who no longer reside in Australia.
Substantial reason
In any event, even if there was an exclusion from, or restriction on, the franchise (and neither of those was established in this case), the features of the electoral system chosen by the legislature (which have been set out earlier) demonstrate that there is a substantial reason for the impugned provisions.
The suspension period reflected in the impugned provisions must be considered in the context of the coherent electoral system chosen by the legislature, which extends to and includes qualification to enrol to vote, entitlement to vote, the process for electing senators and members and the exercise of the right of an elector to vote, ultimately leading to the formation of a government.
As stated earlier, the system is coherent and structured. That structure is rational, logical, efficient and prompt, and provides for an orderly process for senators and members of the House of Representatives to be "directly chosen by the people" as required by ss 7 and 24 of the Constitution, and the subsequent formation of a government. The formation of a representative government is the purpose for which the system exists.
The impugned provisions form a part of that system and are directed to achieving that end. Not only is the end consistent and compatible with the maintenance of the constitutionally prescribed system of representative government, the end is for the maintenance of that system.
Consideration of whether the impugned provisions are reasonably appropriate and adapted to achieving the identified end cannot be done by isolating one aspect (the suspension period) of the electoral system. The electoral system, which the legislature must design and maintain, has a coherent structure. Isolation of the impugned provisions is artificial and, in this case, distracts attention from consideration of the whole structure.
The Rolls are a central component of that system[401]. Indeed, ss 152(1)(a) and 155 specifically acknowledge the inevitability of the Rolls "closing", and provide for the date on which that is to occur. Each of the impugned provisions simply recognises the logical corollary of the closing of the Rolls – namely, the suspension of changes to the Rolls while they are "closed".
[401]See [271]-[287] above.
The Rolls close (and the suspension period commences) only three days before the earliest date upon which the first step of the process by which senators and members are chosen takes place (the close of nominations under s 156 of the Act), where some candidates require the signatures of 100 "electors" entitled to vote at the election for which the candidate is nominated[402]. And, on the closure of the Rolls, the nominated candidates for the House of Representatives are provided with a list of the electors in their Division – who will directly decide whether they should be elected.
[402]s 166(1)(b)(i) of the Act.
Moreover, as noted earlier, any consideration of alternative measures in this case was of limited utility. The alternatives put forward by the plaintiffs demonstrated that an electoral scheme or system may be designed in a variety of ways. However, none of them could be considered a "compelling" alternative when its implementation would require substantive consequential amendment to the broader legislative scheme and raise questions about the allocation of financial resources.
The preceding analysis explains why, even if there was an exclusion from, or restriction on, the franchise, or a distortion of the popular choice, the impugned provisions are reasonably appropriate and adapted to serve an end. That end was and remains an orderly process for senators and members of the House of Representatives to be "directly chosen by the people" as required by ss 7 and 24 of the Constitution.
Conclusion
The impugned provisions form part of a system. They do not relevantly exclude part of universal adult suffrage, restrict the franchise, or produce a distortion of the popular choice. To the extent that the impugned provisions may be considered to effect such an exclusion, restriction or distortion, there is a substantial reason for them doing so. The impugned provisions are not invalid.
Standing
The question of standing can be dealt with shortly. At the time of the hearing, the second plaintiff was enrolled to vote but intended to seek nomination as a candidate for election to the House of Representatives. She had a sufficient interest to raise the issues based on the requirement that she would have needed the signatures of 100 "electors" to seek nomination as a candidate. It is unnecessary to address the standing of the first plaintiff.
Result
For these reasons, I joined in the answers given to the questions reserved.
Murphy v Electoral Commissioner [2016] HCA 36
DBP16 v Minister for Home Affairs [2020] FCA 781
25
22
1