Chapter 4
Elections for the Senate
The powers and operations of the Senate are inextricably linked with the manner of its election, particularly its direct election by the people of the states by a system of proportional representation. This chapter therefore examines the bases of the system of election as well as describing its salient features.
The constitutional framework
The Constitution provides that “The Senate shall be composed of senators for each State, directly chosen by the people of the State, voting, until the Parliament otherwise provides, as one electorate”. Each Original State had initially six members of the Senate and now has twelve. The Parliament is authorised to increase the number of senators elected by each state subject to the qualification that “equal representation of the several Original States shall be maintained and that no Original State shall have less than six senators”. Senators representing the states are elected for terms of six years, half the Senate retiring at three yearly intervals except in cases of or following simultaneous dissolution of both Houses. A state may not be deprived of its equal representation in the Senate by any alteration of the Constitution without the consent of the electors of the state.
Bases of the constitutional arrangements
The constitutional foundations for composition of the Senate reflect the federal character of the Commonwealth. Arrangements for the Australian Senate correspond with those for the United States Senate in that each state is represented equally irrespective of geographical size or population; and senators are elected for terms of six years. Both Senates are essentially continuing Houses: in Australia half the Senate retires every three years; in the United States, a third of the Senate is elected at each biennial election. A major distinction is, however, that the United States Senate can never be dissolved whereas the Australian Senate may be dissolved in the course of seeking to settle disputes over legislation between the two Houses.
An important innovation in Australia was the requirement that senators should be “directly chosen by the people of the State”. Direct election of United States senators was provided in the constitution by an amendment which took effect in 1913, prior to which they were elected by state legislatures.
The innovatory character of Australia's Senate is also illustrated by contrasting it with the Canadian Senate created by the British North America Act 1867. The provinces are not equally represented in the Canadian Senate; and senators are appointed by the national government, initially for life and now until age 75. Composition on this antiquated basis has deprived the Canadian Senate of the legitimacy deriving from popular choice and has meant, in practice, that the Canadian Senate has not contributed either to enhancing the representivity of the Canadian Parliament (the more desirable because of the first-past-the-post method of election used in the House of Commons) nor to assuaging the pressures of Canada's culturally and geographically diverse federation. Prominent proposals for reform of Canada's Senate in recent decades have included equality of representation for provinces and direct election of senators.
The principle of equal representation of the states is vital to the architecture of Australian federalism. It was a necessary inclusion at the time of federation in order to secure popular support for the new Commonwealth in each state especially the smaller states. It ensures that a legislative majority in the Senate is geographically distributed across the Commonwealth and prevents a parliamentary majority being formed from the representatives of the three largest cities and their environs alone. In contemporary Australia it acknowledges that the states continue to be the basis of activity in the nation whether for political, commercial, cultural or sporting purposes. Many organisations in Australia, at the national level, are constituted on the basis of equal state representation or with some modification thereof; this includes the major political parties. By contrast, very few nation-wide bodies are organised on the principle of the election and composition of the House of Representatives. Indeed, in Australia's national life, a body such as the House of Representatives is, if not an aberration, at least relatively unusual. This demonstrates that in Australia federalism is organic and not simply a nominal or contrived feature of government and politics.
Constitutional provisions governing composition of the Senate thus remain as valid for Australia in the 21st century as they were in securing support for the Commonwealth in the nation-building final decade of the 19th century.
In addition to senators elected by the people of the states, the Constitution also provides, in section 122, that in respect of territories, the Parliament “may allow the representation of such territory in either House of the Parliament to the extent and on the terms which it thinks fit”. Since 1975 the Northern Territory and the Australian Capital Territory have each elected two senators. The particular arrangements for election and terms of territory senators are set out in detail below.
The principles of direct election by the people and equal representation of the states are entrenched in the Constitution and cannot be altered except by means of referendum and with the consent of every state. On the other hand, the principle of choosing senators “by the people of the State, voting ... as one electorate” is susceptible to change by statutory enactment. It is, however, essential to the effectiveness of the Senate as a component of the bicameral Parliament.
Current electoral arrangements and proportional representation
As explained in Chapter 1, the Senate, since proportional representation was introduced in 1948, taking effect from 1949, has been the means of a marked improvement in the representivity of the Parliament. The 1948 electoral settlement for the Senate mitigated the dysfunctions of the single member electorate basis of the House of Representatives by enabling additional, discernible bodies of electoral opinion to be represented in Parliament. The consequence has been that parliamentary government of the Commonwealth is not simply a question of majority rule but one of representation. The Senate, because of the method of composition, is the institution in the Commonwealth which reconciles majority rule, as imperfectly expressed in the House of Representatives, with adequate representation.
Proportional representation applied in each state with the people voting as one electorate has been twice affirmed. In 1977, the people at referendum agreed to an amendment to the Constitution so that in filling a casual vacancy by the parliament of a state (or the state governor as advised by the state executive council), the person chosen will be drawn, where possible, from the party of the senator whose death or resignation has given rise to the vacancy. A senator so chosen completes the term of the senator whose place has been taken and is not required, as was previously the case, to stand for election at the next general election of the House of Representatives or periodical election of the Senate. The previous arrangement had the defect of, on occasions, distorting the representation of a state as expressed in a periodical election. The Constitution thus reinforces a method of electing senators which is itself only embodied in the statute law. The present combination of statute and constitutional law serves to underline and preserve the representative character of the Senate.
If the statute law were amended so as to abandon the principle of state-wide electorates for choosing of senators in favour of Senate electorates, this would not only have the defect of replicating the House of Representatives system, which by itself is an inadequate means of even trying to represent electoral opinion fairly, but would invalidate the special method of filling a casual vacancy now provided for in section 15 of the Constitution. Single member constituencies would probably be unconstitutional, as they would result in only part of the people of a state voting in each periodical Senate election. There are grounds for concluding that anything other than state-wide electorates and proportional representation would be unconstitutional.
The second affirmation of state-wide electorates for the purpose of electing the Senate may be found in the decision of the Commonwealth Parliament, on the basis of a private senator's bill, to remove the authority of the Queensland Parliament to make laws dividing Queensland “into divisions and determining the number of senators to be chosen for each division”.
The irresistible conclusion of any analysis of basic arrangements for election of senators is that, for reasons of principle and practice, these features are essential: direct election by the people; equality of representation of the states; distinctive method of election based on proportional representation as embodied in the 1948 electoral settlement for the Senate; elections in which each state votes as one electorate; and filling of casual vacancies according to section 15 of the Constitution.
Terms of Service – State Senators
Except in cases of simultaneous dissolution, senators representing the states are elected for terms of six years. Terms commence on 1 July following the election. The commencement date was originally 1 January but was altered by referendum in 1906 in an ultimately unsuccessful attempt to avoid the problem of unsynchronised elections for both Houses.
The terms of senators elected following a dissolution of the Senate commence on 1 July preceding the date of the general election. Following a general election for the Senate, senators are divided into two classes. Unless another simultaneous election for both Houses intervenes, those in the first class retire on 30 June two years after the general election; those in the second class retire on 30 June five years after the general election. The method of dividing senators is described below.
Terms of service – Territory Senators
Territory senators' terms commence on the date of their election and end on the day of the next election. They therefore do not have the fixed six year terms commencing on 1 July of the senators elected to represent the states. Their terms are, however, unbroken, which is important in ensuring that the Senate has a full complement of members during an election period. Their elections coincide with general elections for the House of Representatives.
Number of senators
Under the Constitution each original state is represented by a minimum of six senators. This number has been twice increased, in 1948 (taking effect at the 1949 elections) to 10, and in 1983 (taking effect in the election of 1984) to 12. The Senate's size also increased after 1975 following election of two senators each by the Australian Capital Territory and the Northern Territory. The size of the Senate was 36 from 1901 until 1949; 60 from 1950 to 1975; 64 from 1976 to 1984; and 76 since 1985. The places of half of the senators for each state are open to election each three years, under the system of rotation. Electoral arrangements for territory senators are described below.
Election timing – periodical elections
Section 13 of the Constitution provides that a periodical election for the Senate must “be made” within one year before the relevant places in the Senate are to become vacant. The relevant places of senators become vacant on 30 June. This means that the election must occur on or after 1 July of the previous year.
The question which arises is whether the whole process of election, commencing with the issue of the writs, must occur within one year of the places becoming vacant, or whether only the polling day or subsequent stages must occur within that period, so that the writs for the election could be issued before 1 July.
This question has not been definitely decided. In Vardon v O'Loghlin (1907) 5 CLR 201, the question before the High Court was whether, the election of a senator having been found to be void, this created a vacancy which could be filled by the parliament of the relevant state under section 15 of the Constitution. The Court found that this situation did not create a vacancy which could be filled by that means, but that the senator originally returned as elected was never elected. A contrary argument was raised to the effect that, under section 13 of the Constitution, the term of service of a senator began on 1 January [now 1 July] following the day of his election, and it would lead to confusion if it were held that the subsequent voiding of the election, perhaps a year or more after the commencement of the term, could not be filled as a vacancy under section 15. In dismissing this argument, the Court, in the judgment delivered by Chief Justice Samuel Griffith, made the following observation:
It is plain, however, that sec. 13 was framed alio intuitu, i.e., for the purpose of fixing the term of service of senators elected in ordinary and regular rotation. The term “election” in that section does not mean the day of nomination or the polling day alone, but comprises the whole proceedings from the issue of the writ to the valid return. And the election spoken of is the periodical election prescribed to be held in the year at the expiration of which the places of elected senators become vacant. The words “the first day of January following the day of his election” in this view mean the day on which he was elected during that election. For the purpose of determining his term of service any accidental delay before that election is validly completed is quite immaterial.
This part of the judgment has been taken to indicate that, in interpreting the provision in section 13 whereby the periodical Senate election must be made within one year of the relevant places becoming vacant, the Court would hold that the whole process of election, not simply the polling day or subsequent stages, must occur within that period. This question, however, has not been distinctly decided. It would still be open to the Court to hold that only the polling day or subsequent stages must occur within the prescribed period, and there are various arguments which could be advanced to support this interpretation. The view that the requirement that the election “be made” within the relevant period means only that the election must be completed in that period is quite persuasive.
If it were decided, however, to hold a periodical Senate election with only the polling day or subsequent stages occurring within the prescribed period, there would be a risk of the validity of the election being successfully challenged and the election held to be void. This would lead to the major consequence that the whole election process would have to start again. It may be doubted whether the Court would favour an interpretation which would bring about this consequence.
Section 13 of the Constitution, as has been noted, also provides that the term of service of a senator is taken to begin on the first day of July following the day of the election. In this provision, the term “day of …. election” clearly means the polling day for the election. This is in accordance with the finding in Vardon v O'Loghlin. The day of election is polling day provided that the election is valid; if the election is found to be invalid then no election has occurred and the question of what is the day of election does not arise.
Election timing – simultaneous general elections
The provision for dating a senator's term from 1 July preceding simultaneous general elections for both Houses has been seen to be the source of a problem stemming from the preference of governments, for financial reasons as well as others of party advantage, to avoid separate dates for a general election of the House of Representatives (the term of which is governed by the date of the simultaneous dissolution) and an ensuing periodical election for half the Senate. The consequence in most cases has been to hold an “early” general election of the House to coincide with the next periodical Senate election. An instance where an “early” general election for the House was not subsequently held in order to synchronise with the next periodical election for the Senate was May 1953; the 1955 general election for the House is the only occasion when an “early” general election has been called to coincide with election of senators to fill the places of second class (long term) senators elected following simultaneous elections for both Houses.
Elections arising from simultaneous dissolutions, held in August 1914, July 1987 and July 2016 did not give rise in significant form to the issue of keeping elections for the two Houses synchronised because of the close proximity of the commencing dates for Senate and House terms in the relevant circumstances. However, the simultaneous dissolution of May 2016, only days before the last possible date to dissolve both Houses under section 57, led to a longer than usual campaign period to ensure a July election and minimal backdating of senators' terms.
The early dissolution of the House of Representatives in November 1929 had, in the event, no effect on synchronisation of Senate and House elections because another early dissolution, occasioned by defeat of the Scullin Government on the floor of the House, was needed in December 1931, a date when a periodical election for the Senate was convenient.
The House of Representatives was prematurely dissolved in 1963; as a consequence there was a periodical election for the Senate the following year. Subsequently there were general elections for the House in 1966, 1969 and 1972, and periodical elections for the Senate in 1967 and 1970. This sequence of unsynchronised elections ended with the simultaneous dissolutions of April 1974.
The case for synchronisation of elections for the two Houses is more a question of convenience and partisan advantage than one of institutional philosophy. Financial considerations simply buttress arguments of party advantage. In a truly bicameral system there is no requirement at all for synchronisation of elections. Proposals to make this a requirement of the Australian Constitution have four times failed at referendum, even though “expert” opinion continues to favour a constitutional amendment of this character.
If there is to be change, a more practical approach would be an alteration of the Constitution to provide that the terms of senators elected in a simultaneous dissolution election should be deemed to commence on 1 July following (rather than preceding) the date of election. Provided that the House of Representatives was not subsequently dissolved within two years of election, synchronisation of a general election for the House and a periodical election for the Senate could be restored with relative ease. Such a proposal, if adopted, would remove the current defect in simultaneous dissolution arrangements of circumscribing the standard six-year term for senators by anything up to one year. This approach would, on the other hand, avoid the two major deficiencies posed by simultaneous election proposals: the augmented power placed in the hands of a prime minister by extending executive government authority over the life of the House of Representatives to half the Senate; and diminishing bicameralism by irrevocably tying the electoral schedule for the Senate to that of the House of Representatives. Effective bicameralism requires that the second chamber should have a significant measure of autonomy in its electoral cycle, as well as distinctive electoral arrangements.
Issue of writs
Writs for the election of senators are issued by the state governor under the authority of the relevant state legislation. The practice is for the governors of the states (when the elections are concurrent) to fix times and polling places identical with those for the elections for the House of Representatives, the writs for which are issued by the Governor-General.
In practice, the Prime Minister informs the Governor-General of the requirements of section 12 of the Constitution, which provides that writs for the election of senators are issued by the state governors, observes that it would be desirable that the states should adopt the polling date proposed by the Commonwealth, and requests the Governor-General to invite the state governors to adopt a suggested date. Theoretically, a state could fix some date for the Senate poll other than that suggested by the Commonwealth, provided it is a Saturday. Different states, too, could fix different Saturdays for a Senate poll.
This power vested in the states to issue writs for Senate elections, fixing the date of polling, gives expression to the state basis of representation in the Senate.
The Constitution provides that, in the case of a dissolution of the Senate, writs are issued within ten days from the proclamation of the dissolution.
The Governor-General issues the writs for elections of territory senators.
Electoral rolls
Under changes introduced in the 2007 election, claims for enrolment or transfer of enrolment could not be considered if lodged after 8 pm on the date of issue of the writs, and the rolls closed on the third working day after the writs were issued. These provisions were ruled invalid by the High Court in Rowe v Electoral Commissioner (2010) 243 CLR 1 and replacement legislation providing for the rolls to close seven days after the date of the writs was enacted in 2011. A claim for enrolment or transfer of enrolment received between the close of rolls and polling day (“the suspension period”), and that was delayed in the post by an industrial dispute, is regarded as having been received before the rolls closed. Claims received during the suspension period are not considered until after polling day. Potential disenfranchisement of claimants for enrolment or transfer during the suspension period was the subject of a challenge before the 2016 election but the challenge was dismissed by the High Court in Murphy & Anor v Electoral Commissioner [2016] HCA Trans 111.
In Getup Ltd v Electoral Commissioner [2010] FCA 869, the Federal Court held that an online enrolment form signed with a digital pen was in order.
Nomination
Nominations close at least 10 days but not more than 27 days after the issue of the writ.
A candidate for election to either House of the Parliament must be at least 18 years old; an Australian citizen; and an elector entitled to vote, or a person qualified to become such an elector.
A person meeting the three qualifications may be disqualified for several reasons. Members of the House of Representatives, state parliaments or the legislative assemblies of the Australian Capital Territory or the Northern Territory cannot be chosen or sit as senators. Members of local government bodies, however, are offered some protection by s. 327(3) of the Commonwealth Electoral Act, but the High Court has not ruled conclusively on this matter. Others disqualified under the Constitution, section 44, are:
- anyone who is a citizen or subject of a foreign power;
- anyone convicted and under sentence, or subject to be sentenced, for an offence punishable by Commonwealth or state law by a sentence of 12 months or more;
- anyone who is an undischarged bankrupt;
- anyone who holds an office of profit under the Crown; and
- anyone with a pecuniary interest in any agreement with the Commonwealth Public Service (except as a member of an incorporated company of more than 25 people).
A person convicted of certain electoral-related offences is disqualified for 2 years.
For cases of the disqualification of senators and senators elect, see Chapter 6, Senators, Qualifications of senators.
No one may nominate as a candidate for more than one election held on the same day. Hence it is not possible for anyone to nominate for more than one division for the House of Representatives, or more than one state or territory for the Senate, or for both the House and the Senate.
Nominations must be made by 12 noon on the day nominations close and the onus is on candidates to ensure nominations reach the electoral officer in time. Candidates may withdraw their nominations at any time up to the close of nominations, but cannot do so after nominations have closed.
Nominations of candidates for the Senate, made on the appropriate nomination form (or a facsimile of the form), are made to the Australian Electoral Officer for the state or territory for which the election is to be held.
A candidate may be nominated by 100 electors or the registered officer of the registered political party which has endorsed the candidate. Nomination of a candidate of a registered political party not made by the registered officer must be verified. Sitting independent candidates require only one nominee.
Nomination forms are not valid unless the persons nominated:
- consent to act if elected;
- declare that they are qualified to be elected and that they are not candidates in any other election to be held on the same day;
- state whether they are Australian citizens by birth or became citizens by other means; and
- provide relevant particulars.
Candidates in a Senate election may make a request on the nomination form to have their names grouped on the ballot paper.
A party name or abbreviation (or for a group endorsed by more than one registered party, a composite name) may be printed on the ballot paper adjacent to the group voting square and any party logo.
A deposit must be lodged with each nomination. The deposit, payable in legal tender or banker's cheque only, is $2,000 for a Senate nomination.
The deposit is returned in a Senate election if, in the case of un-grouped candidates, the candidate's total number of first preference votes is at least four percent of the total number of formal first preference votes; or, where the candidate's name is included in a group, the sum of the first preference votes polled by all the candidates in the group is at least four percent of the total number of formal first preference votes.
Where the number of nominations does not exceed the number of vacancies, the Australian Electoral Officer, on nomination day, declares the candidates elected.
In a Senate election, if any candidate dies between the close of nominations and polling day, and the number of remaining candidates is not greater than the number of candidates to be elected, those candidates are declared elected. However, if the remaining candidates are greater in number than the number of candidates to be elected, the election proceeds. A vote recorded on a Senate ballot paper for a deceased candidate is counted to the candidate for whom the voter has recorded the next preference, and the numbers indicating subsequent preferences are regarded as altered accordingly.
In a House of Representatives election, if a candidate dies between the close of nominations and polling day, the election in that division is deemed to have wholly failed and does not proceed. A new writ is issued for another election in that division, but this supplementary election is held using the electoral roll prepared for the original election.
The statutory provisions regarding death after the close of nominations of a nominated candidate for the Senate could seriously prejudice the prospects of a political party unless a sufficient number of candidates is nominated to avoid disadvantage in the event of a death.
The constitutionality of the statutory requirements for the registration of a political party (500 members, no overlapping membership with other parties) was upheld in Mulholland v Australian Electoral Commission (2004) 220 CLR 181.
Polling
Polling takes place on a Saturday between the hours of 8 am and 6 pm.
The Divisional Returning Officer for each electoral Division arranges for appointment of all polling officials for the Division and makes all necessary arrangements for equipping polling places with voting screens, ballot boxes, ballot papers and certified lists of voters.
Candidates are prohibited from taking any part in the actual conduct of the polling. They may appoint a scrutineer to represent them at each polling place. The scrutineer has the right to observe the sealing of the empty ballot box before the poll commences at 8 am; observe the questioning of voters by the officer issuing ballot papers; object to the right of any person to vote; and observe all aspects of voting by voters in polling places, hospitals, prisons and remote mobile teams.
Voting
Voting is compulsory for all electors with the exception of those living or travelling abroad, itinerant electors and electors located in the Antarctic.
Contrary to the widely held belief that an elector only has to attend a polling place and have their name marked off the roll, the electoral Act specifically states that it shall be the duty of every elector to vote in each election and is quite specific about how ballot papers must be marked. The fact that voting is a private act performed in public means that the identity will never be discovered of electors who may deface their ballot paper or place it unmarked in the ballot box. Nonetheless, the law is still very clear on this point.
Some prisoners are excluded from voting although some of the relevant provisions of the Commonwealth Electoral Act were ruled invalid in the case of Roach v Electoral Commissioner (2007) 233 CLR 162. Replacement legislation was enacted in 2011. The penalty for failing to vote without a valid and sufficient reason is $20 or, if the matter is dealt with in court, a fine not exceeding $50.
Electors may vote at any polling place in the House of Representatives electorate for which they are enrolled, at any polling place in the same state or territory (absent voting) or at an interstate voting centre if they are travelling interstate on election day. Under prescribed circumstances electors may vote by post or cast a pre-poll vote.
Special arrangements are also made for ballots to be cast by eligible voters in hospitals, prisons and remote locations including Antarctica, and those travelling or residing abroad.
The ballot paper
A ballot paper for a Senate election has two parts, each reflecting particular methods of registering a vote. Electors may use only one method. The two parts are separated by a thick horizontal line known as the dividing line, and the two methods are referred to as voting “above the line” or “below the line”.
Introduced in 1983 to addresss an increasing proportion of informal votes for the Senate, the provisions for group voting tickets simplified voting for the Senate if electors chose not to indicate their order of preference for all candidates for that state or territory. By placing the number 1 in a box above the line for their chosen party, group or incumbent senator, voters could thereby adopt the registered preferences of the object of their choice. The constitutional validity of this method of voting was upheld in McKenzie v Commonwealth (1984) 57 ALR 747, Abbotto v Australian Electoral Commission (1997) 144 ALR 352 and Ditchburn v Australian Electoral Officer for Queensland (1991) 165 ALR 147. In due course, however, the potential for the system to be exploited by micro-parties with appealing names whose exchanges of preferences resulted in the election of candidates with miniscule primary votes became increasingly apparent. Recommendations by the Joint Standing Committee on Electoral Matters in an interim report on the conduct of the 2013 federal election for the abolition of group and individual voting tickets and the adoption of optional preferential voting both above and below the line were given effect in the Commonwealth Electoral Amendment Act 2016. The new provisions were the subject of an immediate challenge that was unanimously dismissed by the High Court which found that they did not impinge on the constitutional requirements for there to be one method of choosing senators which shall be uniform for all the States (s. 9) or for senators to be directly chosen by the people of the State (s. 7).
Where groups of candidates or individual incumbent senators have registered as such, a series of boxes is printed on the top part of the Senate ballot paper above the candidates' names. The voter may vote above the line by numbering at least 6 of the boxes in the order of his or her choice, starting with the number 1.
Alternatively, where the voter wishes to indicate preferences among individual Senate candidates on the bottom part of the ballot paper, the voter must place a number 1 in the square opposite the name of the candidate most preferred, and give preference votes for at least 11 other candidates by placing the numbers 2, 3, 4 (and so on, as the case requires) in the squares opposite their names so as to indicate an order of preference for them. The top part of the ballot paper is left blank.
Counting the vote
At the close of the poll each polling place becomes a counting centre under the control of an assistant returning officer who will have been the officer-in-charge of that polling place during the hours of polling.
Only ordinary votes (not postal, pre-poll or absentee votes) are counted at the counting centres on election night. Votes for the House of Representatives are counted before Senate ballot papers, as there is widespread community interest in the formation of government and usually considerable time before the Senate terms begin.
Furthermore, the nature of the Senate voting system means that a quota cannot be struck on polling night, so only provisional figures can be calculated from the ballot papers counted at polling places.
Ballot papers are sorted by the polling officials according to the formal first preference votes marked and the results are then tabulated and sent to the Divisional Returning Officer. Results are relayed through a computer network to the AEC's Virtual Tally Room where progressive figures are displayed. When scrutiny of ordinary votes at each counting centre ends, ballot papers are placed in sealed parcels and delivered to the Divisional Returning Officer.
Other votes are counted at the office of the Divisional Returning Officer after election night. In recent times, amendments to the electoral Act have permitted the computerised scrutiny of votes in Senate elections which has reduced the time taken to calculate results, particularly in the larger States.
After the 2013 election, during the course of a recount of the Western Australian Senate vote, it was discovered that 1370 ballot papers had been lost. An official inquiry failed to locate the papers or identify the circumstances of the loss. Given the closeness of the results and the different outcome from the recount, the AEC itself lodged a petition with the High Court sitting as the Court of Disputed Returns asking for the election result to be declared void. Two other parties lodged similar petitions. The Court declared the election void, holding that it was precluded by the Commonwealth Electoral Act 1918 from reconstructing the result from earlier records of the lost ballot papers, the loss of which, combined with the closeness of the count inevitably affected the result. The election was held again on 5 April 2014, with a date for the return of the writs that allowed all elected or re-elected senators to begin their terms on 1 July 2014.
Candidates may appoint scrutineers who are entitled to be present throughout the counting of votes. The number of scrutineers for a candidate at each counting centre is limited to the number of officers engaged in the counting.
Formal voting in a Senate election
Following a 2008 decision of the Federal Court sitting as the Court of Disputed Returns, a series of principles have been set out by the Court to be applied to the consideration of the admission or rejection of ballot papers. In summary, these principles are to (i) err in favour of the franchise; (ii) only have regard for what is on the ballot paper; and (iii) the ballot paper should be construed as a whole. Subsection 268(3) limits the reasons for informality to those specified and requires a ballot paper to be given effect to according to the voter's intention, so far as it is clear. However, the tests which apply to acceptance of a Senate ballot paper as formal are complicated because a Senate vote can be recorded either by numbering of preferences for individual candidates below the line or for parties or groups above the line. Additionally, a ballot paper may be accepted as formal even where the voter has erroneously attempted to record both types of votes. Thus three distinct cases may arise.
The first case is a vote above the line. A ballot paper is formal if:
- the numbers 1 to at least 6 are written in the squares printed above the line in order of preference for the parties or groups represented; or
- if there are 6 or fewer squares printed above the line, they are numbered consecutively from 1.
Specific allowances are made for voters who deviate from these requirements. A ballot paper is formal if the voter marks only the number 1 in a box above the line, or the number 1 and one or more higher numbers. In addition, a tick or a cross in a box above the line is accepted as the equivalent of the number 1. If a number is repeated, that number and any higher number are disregarded. If a number is missed, any numbers higher than the missing number are disregarded.
The second case is a vote below the line. A ballot paper is formal if:
- the numbers 1 to at least 12 are written in the squares printed below the line in order of preference for individual candidates; or
- if there are 12 or fewer squares printed below the line, they are numbered consecutively from 1.
Specific allowances are again made for voters who deviate from these requirements. If there are more than 6 squares printed below the line on a ballot paper, a vote is formal if the voter has numbered any of those squares consecutively from 1 to 6. In addition, a tick or a cross in a box below the line is accepted as the equivalent of the number 1. If a number is repeated, that number and any higher number are disregarded. If a number is missed, any numbers higher than the missing number are disregarded.
Finally, if a ballot has been marked both above and below the line and each vote would have been formal if recorded on its own, the vote below the line is included in the scrutiny rather than the party or group vote above the line.
As noted in Chapter 6, upon the finding that Senator Wood had not been eligible to contest an election for the Senate in July 1987, it was determined that the place should be filled by counting or recounting of ballot papers cast for candidates for election for the Senate at the election. It was held “that the ballot papers for an election to the Senate, conducted under the system of proportional preferential voting prescribed by Part XVIII of the Commonwealth Electoral Act, for which an unqualified person was a candidate, were not invalid but indications of voters' preference for the candidate were ineffective”.
Determining the successful candidates
The essential features of the Senate system of election are as follows:
Step 1.
To secure election, candidates must secure a quota of votes. The quota is determined by dividing the total number of formal first preference votes in the count by one more than the number of senators to be elected for the state or territory and increasing the result by one. A quota cannot be determined until the total number of formal ballot papers is calculated, which means waiting until the statutory period (13 days) for the receipt of postal votes has passed.
Step 2.
Should a candidate gain an exact quota, the candidate is declared elected and those ballot papers are set aside as finally dealt with, as there are no surplus votes.
Step 3.
For each candidate elected with a surplus, commencing with the candidate elected first, a transfer value is calculated for all the candidate's ballot papers. All those ballot papers are then re-examined and the number showing a next available preference for each of the continuing candidates is determined. Each of these numbers, ignoring any fractional remainders, is added to the continuing candidates' respective progressive totals of votes. Surplus votes are transferred at less than their full value. The transfer value is calculated by dividing the successful candidate's total surplus by the total number of the candidate's ballot papers.
Step 4.
Where a transfer of ballot papers raises the numbers of votes obtained by a candidate up to a quota, the candidate is declared elected. No more ballot papers are transferred to that elected candidate at any succeeding count.
Step 5.
When all surpluses have been distributed and vacancies remain to be filled, and the number of continuing candidates exceeds the number of unfilled vacancies, exclusion of candidates with the lowest numbers of votes commences. Bulk exclusions are proceeded with if possible; otherwise exclusions of single candidates take place. Excluded candidates' votes are transferred at full value in accordance with their next preferences to the remaining candidates. Under certain circumstances the transfer of a surplus may be deferred until after an exclusion or bulk exclusion.
Step 6.
Step 5 is continued, as necessary, until either all vacancies are filled or the number of candidates in the count is equal to the number of vacancies remaining to be filled. In the latter case, the remaining candidates are declared elected.
In counting votes in a Senate election, if only two candidates remain for the last vacancy to be filled and they have an equal number of votes, the Australian Electoral Officer for the state or territory has a casting vote, but does not otherwise vote in the election.
Recounts
Recounts normally occur only when the result of an election is very close. At any time before the declaration of the result of an election, the officer conducting the election may, at the written request of a candidate or on the officer's own decision, recount some or all of the ballot papers. The Electoral Commissioner or an Australian Electoral Officer may direct a recount.
A recount last occurred in 2013 after the result of the count in Western Australia was so close as to raise questions about the safety of the original result. The election was ultimately declared void.
Return of the writ
Writs must be returned within 100 days of issue.
Following the declaration of the result in a Senate election, the Australian Electoral Officer for a state or territory certifies the names of the candidates elected for the state or territory, and returns the writ and the certificate to the Governor of the state or, in the case of the ACT and the Northern Territory, to the Governor-General. The State Governors forward their respective writs to the Governor-General whose Official Secretary in turn passes them to the Clerk of the Senate for tabling at the swearing in of new Senators.
Meeting of new parliament
Under the Constitution, section 5, after any general election (for the House of Representatives and usually a periodical election for the Senate) the Parliament shall be summoned to meet not later than 30 days after the day appointed for the return of the writs.
Disputed returns and qualifications
Under the Commonwealth Electoral Act the validity of any election or return may be disputed only by petition addressed to the Court of Disputed Returns. The High Court of Australia is the Court of Disputed Returns and it has jurisdiction either to try the petition or to refer it for trial to the Federal Court.
A petition must:
- set out the facts relied on to invalidate the election;
- sufficiently identify the specific matters on which the petition relies;
- detail the relief to which the petitioner claims to be entitled;
- be signed;
- be attested by two witnesses whose occupations and addresses are stated;
- be filed in the Registry of the High Court within 40 days after the return of the writ or the notification of the appointment of a person to fill a vacancy;
- be accompanied by the sum of $500 as security for costs.
The Court has wide powers which include power to declare that any person who was returned was not duly elected; to declare any candidate duly elected who was not returned as elected; and to declare any election absolutely void. The requirement for a petition to be lodged within the 40 day limit cannot be set aside. The Court cannot void a whole general election.
The Court must sit as an open Court and be guided by the substantial merits and good conscience of each case without regard to legal forms or technicalities, or whether the evidence before it is in accordance with the law of evidence or not. Questions of fact may be remitted to the Federal Court. All decisions of the Court are final and conclusive and without appeal and cannot be questioned in any way.
If the Court of Disputed Returns finds that a candidate has committed or has attempted to commit bribery or undue influence, and that candidate has been elected, then the election will be declared void.
Any question arising in the Senate respecting the qualification of a senator or respecting a vacancy may be referred by resolution to the Court of Disputed Returns. For cases on the qualifications of senators, see Chapter 6, Senators, under that heading.
Division of the Senate following simultaneous general elections
After a general election for the Senate, following simultaneous dissolutions of both Houses, it is necessary for the Senate to divide senators into two classes for the purpose of restoring the rotation of members.
On all eight occasions that it has been necessary to divide the Senate for the purposes of rotation, the practice has been to allocate senators according to the order of their election. In 2016, the effective part of the resolution provided as follows:
- Senators listed at positions 7 to 12 on the certificate of election of senators for each state shall be allocated to the first class and receive 3 year terms.
- Senators listed at positions 1 to 6 on the certificate of election of senators for each state shall be allocated to the second class and receive 6 year terms.
[update: The division of the Senate is a matter for the Senate itself. However, there was speculation during the 45th Parliament, with the disqualification of numerous senators under section 44 of the Constitution, whether the High Court might have a role.
If a senator is found to have been disqualified at the time of election, their election is void and the vacancy is filled by a recount of the ballots under the supervision of the Court (a “special count”) to determine the person validly elected: see Chapter 6—Senators, under Qualifications of senators. The usual form of the court order following a special count was that a person is “duly elected for the place for which” the ineligible candidate was returned. One question agitated in hearings in December 2017 was whether such an order also had the effect of granting the incoming senator the term (that is, the 3- or 6-year term) initially allocated by the Senate to the ineligible candidate.
Nettle J described as “an attractive proposition” the view put by the Commonwealth Solicitor-General that there is “…a very real question as to whether anyone other than the Senate has a role in determining the three- or six-year issue. It may be that the Court has a role in declaring who the people are, and the Senate then chooses who gets three and who gets six years”: Re Parry; Re Lambie [2017] HCATrans 258 (13 December 2017).
Moreover, the High Court has held that a person invalidly returned in an election does not have a “term of service” at law for the purposes of section 13 of the Constitution: Vardon v O’Loghlin [1907] 5 CLR 201 at 211, 214. That being the case, it is hard to see how an order of the Senate under section 13 could have any effect in relation to that person, and similarly hard to argue that an incoming senator inherits that (non-existent) term.
In February 2018, the Senate moved to remedy any uncertainty by modifying the effect of the August 2016 resolution, so that it would operate by reference to the revised order of election produced in any relevant special count: 13/2/2018, J.2690-1. In doing so, the Senate preserved the principle adopted at the beginning of the Parliament, that the longer terms be allocated to the senators first elected in the count, and asserted the conventional view that the division of the Senate is a matter for the Senate itself.]
[update: Alternative method for dividing the Senate]
In its report of September 1983 the Joint Select Committee on Electoral Reform proposed that “following a double dissolution election, the Australian Electoral Commission conduct a second count of Senate votes, using the half Senate quota, in order to establish the order of election to the Senate, and therefore the terms of election”. The committee also recommended that there should be a constitutional referendum on “the practice of ranking senators in accordance with their relative success at the election” so that “the issue is placed beyond doubt and removed from the political arena”. The Commonwealth Electoral Act was subsequently amended to authorise a recount of the Senate vote in each state after a dissolution of the Senate to determine who would have been elected in the event of a periodical election for half the Senate.
Following the 1987 dissolution of the Senate, the then Leader of the Government in the Senate, Senator John Button, successfully proposed that the method used following previous elections for the full Senate should again be used in determining senators in the first and second classes respectively.
The Opposition on that occasion unsuccessfully moved an amendment to utilise section 282 of the Commonwealth Electoral Act for the purpose of determining the two classes of senators, in accordance with the September 1983 recommendation of the Joint Select Committee on Electoral Reform. According to the leading Opposition speaker, Senator Short, the effect of using the historical rather than the proposed new method was that two National Party senators would be senators in the first (three-year) class rather than the second (six-year) class, whilst two Australian Democrat senators would be senators in the second rather than the first class.
On 29 June 1998 the Senate agreed to a motion, moved by the Leader of the Opposition in the Senate, Senator Faulkner, indicating support for the use of section 282 of the Commonwealth Electoral Act in a future division of the Senate. The stated reason for the motion was that the new method should not be adopted without the Senate indicating its intention in advance of a simultaneous dissolution, but it was pointed out that the motion could not bind the Senate for the future. An identical motion was moved by Senator Ronaldson (Shadow Special Minister of State) on 22 June 2010 and agreed to without debate.
No such resolution preceded the 2016 dissolution and the order of election method was again followed. The recount method would have resulted in two minor party senators being allocated six-year terms at the expense of two major party senators.
Casual vacancies
Casual vacancies in the Senate are created by death, resignation or absence without permission.
In the case of resignation, a senator writes to the President, or the Governor-General if there is no President or the President is absent from the Commonwealth. A resignation may take the following form—
(Date)
Dear Mr/Madam President
I resign my place as a senator for the State of , pursuant to section 19 of the Constitution of the Commonwealth of Australia.
Signature
Where the letter of resignation is sent to the Governor-General, the form may be as follows:
(Date)
Dear Governor-General,
Section 19 of the Constitution provides —
“A senator may, by writing addressed to the President, or to the Governor-General if there is no President or if the President is absent from the Commonwealth, resign his place, which thereupon shall become vacant.”
As the President of the Senate is absent from the Commonwealth, I address my resignation to you.
I resign my place as a senator for the State of ..........., pursuant to section 19 of the Constitution of the Commonwealth of Australia.
Signature
If the President resigns as a senator, the resignation is addressed to the Governor-General.
The following principles have been observed in relation to the manner in which senators may resign their place:
- a resignation by telegram or other form of unsigned message is not effective;
- a resignation must be in writing signed by the senator who wishes to resign and must be received by the President; whether the writing is sent by post or other means is immaterial;
- it is only upon the receipt of the resignation by the President that the senator's place becomes vacant under section 19 of the Constitution;
- a resignation cannot take effect before its receipt by the President;
- a resignation from a current term may not take effect at a future time;
- the safest procedure is for the resignation, in writing, to be delivered to the President in person in order that the President can be satisfied that the writing is what it purports to be, namely, the resignation of the senator in question; resignations transmitted by facsimile or other electronic means and confirmed by telephone are accepted.
On 5 July 1993 Senator Tate, having just commenced a new term as a senator for Tasmania, resigned before taking his seat in the Senate. The resignation of Senator Tate before his swearing in did not affect the procedure for his replacement. The interesting questions that would have arisen had he resigned before the end of his term were deferred till 2013 when Senator Bob Carr resigned, having just been elected to a new term starting on 1 July 2014. He submitted what was in effect a “double resignation”, resigning both from his place in respect of his term ending on 30 June and also in respect of his new term commencing on 1 July. Notification of both vacancies was provided to the Governor of NSW by the President of the Senate pursuant to section 21 of the Constitution.
The resignation of a senator-elect in Senator Bob Carr's case was taken as giving rise to a double vacancy in respect of his current term and the term to which he had been elected. The death of a senator-elect has also been regarded as creating a casual vacancy to be filled in accordance with section 15 of the Constitution. Presumably a senator-elect could become disqualified and similarly create a casual vacancy. The disqualification of a senator at the time of election, however, does not create a vacancy but a failure of election which is remedied by a recount of ballot papers.
The Constitution, section 20, states that the “place of a senator becomes vacant if for two consecutive months of any session of the Parliament” a senator fails to attend the Senate without its permission. In 1903 the seat of Senator John Ferguson was declared vacant owing to absence without leave for two months. For the purposes of section 20, a record is kept in the Journals of the Senate of senators' attendance.
Method of filling casual vacancies
Casual vacancies are filled in accordance with section 15 of the Constitution.
The purpose of the current section 15, inserted by an amendment of the Constitution in 1977, is to preserve as much as possible the proportional representation determined by the electors in elections for the Senate.
The main features of the section are as follows:
- When a casual vacancy arises, the Houses of the Parliament, or the House where there is only one House, of the state represented by the vacating senator chooses a person to hold the place until the expiration of the term.
- If the Parliament is not in session, the Governor of the state, with the advice of the Executive Council thereof, may appoint a person to hold the place until the expiration of 14 days from the beginning of the next session of the parliament of the state or the expiration of the term, whichever first happens.
-
A person chosen is to be, where relevant and possible, a member of the party to which the senator whose death or resignation gave rise to the vacancy. The pertinent paragraph of section 15 states:
Where a vacancy has at any time occurred in the place of a senator chosen by the people of a State and, at the time when he was so chosen, he was publicly recognised by a particular political party as being an endorsed candidate of that party and publicly represented himself to be such a candidate, a person chosen or appointed under this section in consequence of that vacancy, or in consequence of that vacancy and a subsequent vacancy or vacancies, shall, unless there is no member of that party available to be chosen or appointed, be a member of that party.
-
Section 15 also provides:
Where —
- in accordance with the last preceding paragraph, a member of a particular political party is chosen or appointed to hold the place of a senator whose place had become vacant; and
- before taking his seat he ceases to be a member of that party (otherwise than by reason of the party having ceased to exist),
he shall be deemed not to have been so chosen or appointed and the vacancy shall be again notified in accordance with section twenty-one of this Constitution.
[update: This last provision gives the recognised party of a departing senator effective control over the choice of a replacement, including by deeming the choice of the state parliament void if “before taking his seat he ceases to be a member of that party”. Following the resignation of Senator Xenophon in 2017, reports that a party member other than the chosen nominee might press a claim to the position came to naught, so the operation of that part of section 15 remains untested: see also Delay in filling casual vacancies, below. In 2020 the vacancy caused by the resignation of then independent Senator Bernardi was filled by a nominee of the Liberal Party; the party he had represented at the 2016 election: 4/2/2020, J.1158; 10/2/2020, J.1271.]
Casual vacancies arising in the Senate representation of the Australian Capital Territory or the Northern Territory are filled by the respective territory legislative assemblies. If the legislature is out of session, a temporary appointment can be made in the case of the Australian Capital Territory by the Chief Minister, and in the case of the Northern Territory by the Administrator. Provisions relating to political parties, similar to those of section 15 of the Constitution, also apply.
The term of a senator filling a casual vacancy commences on the date of his or her choice by the appointing body. When a senator is appointed to a vacant place by the governor of a state and the appointment is “confirmed” by the state parliament within the 14 days allowed by section 15, the senator is not regarded as commencing a new term on the appointment by the parliament and is not sworn again. The 14 day period is regarded as commencing on the day after the first day of the session, in accordance with the normal rule of statutory interpretation. If there is a “gap” between the expiration of the 14 day period and the appointment of the senator by the parliament, the senator is sworn again.
The “double resignation” of Senator Bob Carr in 2013 created interesting questions for the Parliament of New South Wales in choosing a replacement. Senator Carr's party nominated one person to fill both the remainder of his current term and the new term to which he had been elected, but the Parliament, after considering advice from the Crown Solicitor, determined that it could fill the current vacancy only and could not act prospectively to fill a future vacancy. The advice was tabled in the New South Wales Legislative Council on 12 November 2013.
With the NSW Houses not scheduled to sit between 17 June and 12 August 2014, further advice was sought from the NSW Crown Solicitor about whether an appointment could be made by the Governor and whether a resolution of the Senate encouraging the NSW Parliament to fill the vacancy could somehow act as a “trigger” for the Houses to meet and fill the vacancy. Not surprisingly (NSW having always taken a strict view of when a governor's appointment could be made) the advice on both questions was negative. In any case, the Senate did not contemplate such a resolution. However, the NSW Houses resolved to meet on 2 July 2014 and again chose Senator O'Neill to fill the second vacancy created by the resignation of Senator Bob Carr. For the avoidance of doubt, the President, on 1 July 2014, reminded the NSW Governor of his earlier notification of the vacancy existing from that date.
Delay in filling casual vacancies
The 1977 alteration of the Constitution has not entirely solved all problems in the filling of casual vacancies. There is nothing to compel a state parliament to fill a vacancy. This was illustrated in 1987 following the resignation of Tasmanian Senator Grimes, who had been elected to the Senate as an endorsed candidate of the Australian Labor Party. In accordance with the Constitution, section 15, the Parliament of Tasmania met in joint sitting on 8 May 1987. The Leader of the Australian Labor Party in the House of Assembly and Leader of the Opposition, Mr Batt, nominated John Robert Devereux to fill the vacancy. In the ensuing debate it became apparent that government members as well as a number of independent members of the Legislative Council intended to vote against the nomination. The basis for doing so, in terms of the Constitution, was expressed as follows by Mr Groom, Minister for Forests:
It has been suggested by some people that there is a convention which requires us to accept Mr Devereux's nomination without question, but section 15 of the Constitution clearly states that it is for the Parliament to choose the person to fill the vacancy and not the party. We can choose only a person who is a member of the same party as the retired senator — that is well recognised — but we are not bound to accept the nomination of the party concerned.
The matter shortly came to a vote. Votes were tied at 26 each. The question was thus resolved in the negative in accordance with the rules adopted for the joint sitting.
Subsequently a member of the Legislative Council who had voted “No” in the division nominated William G McKinnon, a financial member of the Australian Labor Party and former member of the Tasmanian Parliament, to fill the vacancy and produced a letter from the nominee agreeing to the nomination. After a brief suspension the chair of the Joint Sitting declared that the “letter is not in order”. He continued:
It does not comply with rule 16(6) in that the letter does not declare that the person is eligible to be chosen for the Senate and that the nomination is in accordance with section 15 of the Constitution of the Commonwealth of Australia. Therefore I am in the position of being unable to accept the nomination.
The joint sitting adjourned soon afterwards without any further voting.
The filling of the casual vacancy was, in the event, overtaken by simultaneous dissolutions of the Senate and the House. In the subsequent election John Devereux was among the endorsed ALP candidates in Tasmania who were elected.
In the Senate itself, the Opposition granted a pair to the government following Senator Grimes' resignation so that in party terms relative strengths were maintained. The Opposition's position on the matter was stated in the following terms: “the person appointed to fill casual vacancies of this kind ought to be the person nominated by the retiring senator's political party”.
There was no certainty as to the outcome of the dispute. According to Senator Gareth Evans, representing the Attorney-General in the Senate, “we have all the makings, however, of a deadlock, and that is what will prevail in the absence of legal challenge and in the absence of a change of heart in Tasmania at the moment”.
Failure to fill a casual vacancy promptly means that a state's representation in the Senate is deficient and the principle of equality of representation infringed. The Senate itself takes a keen interest in prompt filling of casual vacancies and has on several occasions expressed by resolution concern about delay. On 19 March 1987, in the case of the Tasmanian vacancy, the Senate expressed the view that the nominee of the relevant party should be appointed. Because of the delay in filling a casual vacancy created by the resignation of Senator Vallentine on 31 January 1992, the Senate passed a resolution on 5 March 1992 expressing its disapproval “of the action of the Western Australian Government for failing to appoint Christabel Chamarette [the candidate endorsed by the relevant political group] as a Senator for Western Australia, condemns the Western Australian Government for denying electors of that state their rightful representation in the Senate, and condemns the Western Australian Government for the disrespect it has shown to the Senate”.
On 3 June 1992 the Senate passed the following resolution:
That the Senate —
- believes that casual vacancies in the Senate should be filled as expeditiously as possible, so that no State is without its full representation in the Senate for any time longer than is necessary;
- recognises that under section 15 of the Constitution an appointment to a vacancy in the Senate may be delayed because the Houses of the Parliament of the relevant State are adjourned but have not been prorogued, which, on a strict construction of the section, prevents the Governor of the State making the appointment; and
- recommends that all State Parliaments adopt procedures whereby their Houses, if they are adjourned when a casual vacancy in the Senate is notified, are recalled to fill the vacancy, and whereby the vacancy is filled:
- within 14 days after the notification of the vacancy, or
- where under section 15 of the Constitution the vacancy must be filled by a member of a political party, within 14 days after the nomination by that party is received,
whichever is the later.
This resolution was passed because the government of Western Australia had adopted the “strict construction” referred to in the resolution, that the state governor could not fill the vacancy because the state Parliament was not prorogued but the Houses had adjourned. Other states from time to time have adopted the view that their governors fill vacancies when their Houses are adjourned. This resolution was reaffirmed in 1997.
The Senate passed a resolution on 4 March 1997 calling on two states to fill casual vacancies expeditiously. The resolution was prompted largely by statements by the Premier of Queensland that a casual vacancy in that state caused by a mooted resignation of a senator might not be filled in accordance with section 15 of the Constitution. A resolution of 15 May 1997 referred to the tardiness of the Victorian government in filling vacancies. In 2015, a resolution agreed to on 26 March reaffirmed earlier resolutions and called on NSW to take all necessary steps to fill the vacancy caused by the resignation of Senator Faulkner. Despite the 1991 precedent, a governor's appointment was not made after the state Parliament was prorogued, and the vacancy remained unfilled until after the NSW Houses met following the state election.
The obligation on states to fill casual vacancies as expeditiously as possible is matched by an obligation on the Senate to swear in and seat the appointees at the earliest possible time. The Senate has always adhered to this principle.
[update: On the Senate's final sitting day in 2021 the Victorian Parliament chose a senator to fill a casual vacancy, but too late to enable the senator to be sworn in. He took his seat on the first sitting day in 2022. Two casual vacancies arising in 2022 were filled following the prorogation of the parliament for a general election, and the Senate did not meet again before the terms of those senators expired on 30 June.]
A list of casual vacancies filled under section 15 of the Constitution is contained in Appendix 7.
Territory senators
Until 1975 all members of the Senate were elected to represent the people of the states. In the elections in December 1975 following simultaneous dissolution of the two Houses on 11 November 1975 the Australian Capital Territory and the Northern Territory each elected two senators for the first time.
Legislation for election of territory senators was enacted in the Senate (Representation of Territories) Act 1973. This legislation was based on the Constitution, section 122, which provides that, in relation to territories, the Parliament “may allow the representation of such territory in either House of the Parliament to the extent and on the terms which it thinks fit”. The provisions for the representation of the territories in the Senate are now contained in the Commonwealth Electoral Act, ss 40-44.
The legislation was not enacted without controversy. Indeed, it was one of the bills cited as a ground for the simultaneous dissolutions of 1974 and was eventually passed into law at the joint sitting of that year. It was subsequently twice challenged in the High Court, surviving the first challenge by a majority 4 to 3 decision, and the second by a majority of 5 to 2.
The principal issue in dispute was the contention that territory senators would undermine the constitutional basis of the Senate as a house representing the people by states and that territory representation would disrupt the numerical balance between large and small states. Other questions related to the voting rights of territory senators; the effect of territory senators on the nexus between the sizes of the two Houses and on quorums in the Senate; and applicable criteria in determining whether a territory should be represented in the Senate. A full account of the matter is contained in ASP, 6th ed. That edition concluded that “the broadest possible representation of all the people of Australia best serves that [the Senate's] checks and balances role”.
Given that each territory's representation is currently limited to two senators, the practice of electing both at the one election by proportional representation preserves the Senate's role as a House which enhances the representative capacity of the Parliament and provides a remedy for the defects in the electoral method used for the House of Representatives. As indicated in Chapter 1, since the 1980 general election all members of the House of Representatives for ACT electorates have usually been members of the Australian Labor Party. Throughout much of this period, one senator has been a member of the ALP, the other senator from the Liberal Party. One-party representation in the House has also been common for the Northern Territory, so that its two senators are also essential to providing that territory with balanced representation.
The writ for election of senators for a territory is issued by the Governor-General and is addressed to the Australian Electoral Officer for that Territory; following declaration of the result of a Senate election in a territory, the writ is returned to the Governor-General.