Bicameral representation

Dr Scott Brenton

  < - Contents - Chapter: Conclusion - >

Bicameral representation

‘If a second chamber dissents from the first, it is mischievous; if it agrees it is superfluous’ – eighteenth-century political philosopher Abbé de Sièyes[1]

In most countries, the upper house is subordinate to the lower house, exemplified by the terms ascribed to the Senate: an ‘upper house’ is symbolically further from the people than the lower house; ‘second chamber’ is secondary; and even ‘house of review’ is reactive to the ‘house of government’.[2] Even the name House of Representatives implies that members are the real representatives, not the senators. House of Representatives Practice contends that the Senate ‘is not an equitably representative body in the sense that the House is’.[3] Uhr notes that the Australian Senate’s powers are remarkable and unlike that of any other upper house in a Westminster-derived parliament.[4] In Westminster-derived parliamentary systems, Budget bills are generally regarded, at least by the government, as sacrosanct and equivalent to a vote of confidence. Yet as Bach, among others, note, the Senate can compel the government to resign by denying supply, and therefore the executive is not only responsible to the lower house.[5] In Australia, the Senate can undermine the government’s budgetary authority, and has in 1975 precipitating a Constitutional ‘crisis’, and in 1993 when minor parties in the Senate demanded input into the budget and the budget process. [6] The dismissal of the Whitlam Labor Government in 1975 after the Senate blocked supply challenged the convention that government can only be brought down in the House of Representatives.

Evolution of the Australian Senate

Australia’s system of government is imaginatively described as a ‘Washminster’ mutation, transplanting the federal system from the United States onto the Westminster system of responsible parliamentary government. The constitutional conventional debates of the 1890s focused on the United States (and even Canada) and adapting their system to the existing British-derived systems of government as existed in the colonies at the time.[7] Yet they did prefer the British cabinet system of ministerial membership of parliament rather than the American presidential system, although America’s powerful Senate, or more precisely its federal system and state-based representative chamber, proved an irresistible model. However, the stature of the United States’ Senate was not successfully transplanted, with the Australian Senate retaining a sense of illegitimacy like the British House of Lords. Hewitt argues that the Australian Senate has not been able to attain the status and prestige of its American counterpart as it does not exercise the same level of oversight in relation to the administration and foreign affairs.[8] However, one notable feature of the Senate that Australia did not adapt from the United States, or from the British House of Lords, was the direct election of senators.[9] Thus, unlike in many other bicameral systems around the world, Australia has always had two houses of directly-elected representatives. In the Senate, where representatives are elected by the people of a state, are they representatives for the people or for the state?

Electoral legislation was reformed to introduce proportional representation for the Senate in 1948, with Labor Attorney-General H. V. Evatt describing the reform as ‘one most likely to enhance the status of the Senate’.[10] Yet there were no minor parties like the Democratic Labor Party, the Democrats or Greens to worry about at the time of the reforms. Odgers’ Australian Senate Practice is unsurprisingly complimentary in its description of the reforms:

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.[11]

As Bach argues, the 1949 electoral law reforms increased the likelihood of different balances of party power in each house.[12] Thus the Senate has the capacity to frustrate a government’s agenda. In recent decades, the Opposition has not controlled the Senate; rather minor parties and independents have held the balance of power between the government and opposition (or the government has controlled both houses). At every Senate election since 1955, candidates from outside the Labor, Liberal and National parties have been elected.[13] Minor parties and independents have held the balance of power from 1981–2005 and since 2008. While minor parties have been more electorally successful in the Senate, there have been more Independents elected to the House of Representatives than the Senate.[14]

This wider range of parties has given rise to the description of the Senate as the ‘Parties’ House’, even though the strength of party discipline in the major parties is practically the same in both houses. Indeed, Bogdanor argues that there is a fundamental weakness in the representative principle itself, as lower house members do not necessarily represent the popular will of the people, particularly under conditions of strong political party discipline.[15] However, party voting is more pronounced in the Senate, since the 1984 electoral reforms introduced ‘above the line’ or ‘Group Ticket Voting’. Under this system, voters can simply vote ‘1’ for a party grouping, rather than having to allocate preferences themselves, which can often be a cumbersome process. The parties not only control the order of their own candidates, but also the order of preferences for other candidates. Blom argues that:

Except for this accountability as an institution, individual senators may experience few pressures to account for their actions, as the elective element is weak or non-existent. At the most they have to respond to an exclusive audience to promote their re-election and even then, their activities as senators might be relatively unimportant. Whatever they do, so it seems, there is almost no survival-pressure to account for it. They might as well snore![16]

While Blom was referring to upper houses in general, including many non-directly elected upper houses, the nature of the Australia’s electoral system and dominance of the major parties means that Senate candidates in the top two positions of the major parties state tickets are virtually guaranteed election.

Among current and former senators there was considerable disagreement about whether  state-based representation is relevant. One senator believed that states are on a path to oblivion, but most were more circumspect. Another senator argued there will be always be distinct states, and a former party leader was sceptical about whether Australia could ever move away from state-based representation in the Senate. One long-serving senator argued that the role of the Senate as a ‘states’ house’ has diminished, particularly as free voting is less common and party discipline is much more regimented, as senators are elected by the party so their first allegiance is to the party.

As one senator put it, some issues are just state issues, some are national, and some start as state and become national. One senator stills sees it as a ‘states’ house’, in the sense of looking at state issues through a national framework and placing them in the bigger picture. For example, South Australian senators cited the Murray-Darling basin and the issue of water security as a clearly identifiable state issue. The budget was another area where Senators saw opportunities to ensure that their states received their fair share.

One interviewee observed that most ‘states’ issues’ are actually state government issues. One senator argued that in terms of state representation it is ‘horses for courses’—when it suits senators they claim that the Senate is a states’ house but it just as much a parties house and ‘to deny that is just a folly’. One interviewee explained that while state-based representation has been notional since Federation, equal representation means that the small states get more voice in the party room and that there is substantial debate in the party room. However, the ACT is disadvantaged as it is underrepresented in terms of population, and the ACT’s voice can be drowned out in the party room (there is only one ACT representative in the Liberal Party room compared with five Tasmanians, and three ACT representatives in Labor’s party room compared with ten Tasmanians).

Uhr argues that: ‘The common rhetoric of the Senate as a (failed) states’ house has got it no more than half right, and has done a disservice in suppressing the wider justification of the Senate as a brake on the misuse of majority power’.[17] Bach adds that: ‘In any event, whether it is accurate to say that the Senate was intended to be the House of the States, observers of the Parliament in practice are virtually unanimous in stressing that whatever the Senate may be, a House of the States it is not’. He goes onto reject the names House of the States or House of Review, preferring the distinctive names House of Responsibility (lower house) and House of Accountability (upper house).[18] If senators are not always directly representing their states, then who or what do they represent? Parties? While members of the House of Representatives have identifiable constituents, given the strict party discipline that exists in Australian politics and the diversity of large electorates, do they represent a constituency as much as a party? These questions are not new but if senators and members are both representing the same thing and if they sit in houses with almost the same powers, then do they actually differ in terms of their representative roles and responsibilities? Figures 27a and 27b show parliamentarians attitudes towards the institution of parliament in terms of the different functions of both houses.

Figure 27a: Current parliamentarians’ views of the roles of the houses

Members: Current parliamentarians’ views of the roles of the housesSenators: Current parliamentarians’ views of the roles of the houses

Figure 27b: Former parliamentarians’ views of the roles of the houses

Members: Former parliamentarians’ views of the roles of the housesSeantors: Former parliamentarians’ views of the roles of the houses

Many senators discussed the importance of the review role of the Senate, and strongly identified with that role. Many mentioned holding the government to account and checks and balances. A Senate interviewee was interested in reviewing legislation and holding the executive accountable to the legislature. One interviewee who has served in both houses observed that the Senate’s transformation into a house of accountability is only relatively recent, and that it was once quite a staid place and that estimates was boring. Indeed a long-serving parliamentarian observed that the relationship between the houses has changed and the status of the Senate has increased as the government has often not controlled the numbers in recent decades. Who holds the balance of power in the Senate appears to be a crucial factor. One senator claimed that when the government had control of the Senate there was not much difference between the houses but when the government does not have control, the workload and presence of senators increases. However, one current parliamentarian observed that during the last term of the Howard government, when the Coalition controlled both houses, that government backbenchers would question the executive more. One former government senator conceded that democracy is enhanced when the government does not have a majority in the Senate.

Yet advocates of the Senate were matched by critics. One former member described the Senate as the ‘B team’ as the House of Representatives is the House of Government. One senator conceded that the Senate is still seen as a second house and a bit of a sleepy place and it is difficult to get attention, particularly from the Opposition, as the media do not follow the debates closely and are more interested in the Prime Minister and Treasurer, who sit in the lower house. A current senator considered changing to the lower house to become a party leader and believes that one has greater influence in the lower house. However, another senator argued that most members do not understand the Senate or the roles of the senators. Unsurprisingly, minor parties were the biggest advocates of a strong and powerful Senate, but many did recognise the limitations of being able to frame an agenda, particularly in relation to budget bills. However, crossbench senators still actively initiate bills and propose amendments, and occasionally are successful in their endeavours.

The Senate’s review role can sometimes overshadow the House. While the government’s legislation always passes the House it is reviewed by an unpredictable Senate, which one House interviewee described as an over-simplification that does not acknowledge the contribution of House debate in forming the basis for Senate amendments. Ministers often listen intently and frame amendments themselves to be presented in the Senate. One member argued that accountability, however imperfect in the House, is still there with Questions on Notice and Without Notice, which is a very important part of the process. The House committee system is underrated but significant, as it checks the power of the executive. Flaws are exposed during the course of debates in and the government listens and put amendments in the Senate. This member claimed that the Australian parliament is one of the few in the world where the Prime Minister and ministers front parliament every day.

Upper House membership

Advocates of bicameralism stress the review role of upper house, variously described as: the chambre de réflexion, the ‘embodiment of considered opinion’, that ‘provides for second opinion’ and ‘sober second thought’. In performing the role, the membership often differs from that of the lower house. Shell argues that:

In principle a second chamber ought to provide for the mobilisation into a legislature of people whose experience is different, decidedly different, from that which is normal for the first chamber. The argument here would be that if a second chamber is going to exist at all it ought to be so composed that its membership is dissimilar in important respects from that of the first chamber.[19]

Couwenberg proposes a division of labour to improve the functioning of bicameralism, based on the Montesquieu model where there is a division of tasks between the chambers and a different composition of the second chamber.[20] Tsebelis and Money state that:

Montesquieu also noted the distinctive qualities of a senate, whose members shared the characteristics of age, virtue, wisdom, and service to the community. Such a council would reinforce the stability of the polity through its sound advice. In other words, a senate was a wise body that could remind the society of its first principles and ensure that new legislation improved rather than corrupted the old.[21]

Couwenberg argues that there is ‘an undeniable difference in abilities and interests among politicians: some are more interested in legislation and others more in policy making and policy checking’.[22] Indeed, many senators cited this as the primary reason for contesting the Senate in the first place. John Sturt Mill conceived of one House representing popular feeling and the other personal merit, in the sense that upper House members should have special training or knowledge to comprise a chamber of ‘statesmen’ counterbalancing the chamber of the people.[23] Indeed, one interviewee argued that there are different skill sets required for each house. Eligibility for upper house membership can be restricted, including higher minimum age requirements, social and economic restrictions, or the need for professional qualifications.[24] Yet on the basic measures of age, length of service (even though the Senate has longer terms), qualifications and previous occupations, the distributions of current senators and members are quite similar. This is shown in Figures 28a to 28f.

Figure 28a: Age distribution of current parliamentarians*

Age distribution of current parliamentarians

Data Source: Parliamentary Library, Parliamentary handbook of the Commonwealth of Australia, Thirty-first edition 2008, Department of Parliamentary Services, Canberra, 2008, p. 256.

* as at beginning of the 42nd parliament


Figure 28b: Years served by current parliamentarians*

Years served by current parliamentarians

Data Source: Parliamentary handbook of the Commonwealth of Australia, Thirty-first edition 2008, p. 256. (The graph includes one member who served four years in the Senate; two members who served six and eight years respectively in the Senate, and one senator who served eleven years in the House of Representatives.)

* as at beginning of the 42nd parliament

Figure 28c: Levels of qualifications of current parliamentarians

Levels of qualifications of current parliamentarians

Data Source: Parliamentary handbook of the Commonwealth of Australia, Thirty-first edition 2008, p. 264.

Figure 28d: Proportions of current parliamentarians with prior elected service in local and state government

Proportions of current parliamentarians with prior elected service in local and state government

Data Source: Parliamentary handbook of the Commonwealth of Australia, Thirty-first edition 2008, p. 263.

Figure 28e: Previous managerial & administrative occupations of current parliamentarians

Previous managerial & administrative occupations of current parliamentarians

Data Source: Parliamentary handbook of the Commonwealth of Australia, Thirty-first edition 2008, p. 263.

Figure 28f: Previous professional occupations of current parliamentarians

Previous professional occupations of current parliamentarians

Data Source: Parliamentary Handbook of the Commonwealth of Australia, Thirty-first edition 2008, pp. 268–270.

The most notable differences are that more than twice the proportion of members have previous experience in elected government and more senators have party/union management/administration experience, which is further evidence of Senate preselection process favouring organisational stalwarts. Where there are differences between senators and members is in terms of more personal characteristics.

Mirror representation

The 1996 Australian Parliamentary Fellow, Gianni Zappalà, examined ethnic politics and political representation in his monograph entitled Four Weddings, a Funeral and a Family Reunion: Ethnicity and Representation in Australian Federal Politics.[25] Zappalà discusses both system-wide approaches to representation and normative theories of representation. Of particular relevance to this monograph is Zappalà’s critique of mirror representation, that is, the idea that Parliament should proportionally reflect the different demographic groups in Australian society. While such critiques are beyond the scope of this monograph, of interest in this study is whether one House mirrors Australia’s diversity more than the other House. Mirror representation is often practised electorally through proportional representation, which is the method of election for the Australian Senate.

The idea that representatives should share the characteristics of the people they represent dates back to the American Revolution, with John Adams arguing that the legislature ‘should be an exact portrait, in miniature, of the people at large, as it should think, feel, reason and act like them’.[26] He believed the best way of avoiding corruption and tyrannical monarchies, with diversity increasing popular involvement, democratic responsiveness and political accountability.[27] Mirror representation is also commonly known as descriptive representation, or what Pitkin describes as ‘standing for’.[28] ‘True’ representation is where the composition of the representative body corresponds to, mirrors or reflects the whole nation. Pitkin writes that:

In politics, too, representation as “standing for” by resemblance, as being a copy of an original, is always a question of which characteristics are politically relevant for reproduction … the history of representative government and the expansion of the suffrage is one long record of changing demands for representation based on changing concepts of what are politically relevant features to be represented. The nation is not like a geographic area to be mapped—solidly there, more or less unchanging, certainly not changed by the map-making process.[29]

Pitkin notes that this conception of representation is significantly different from the formalistic authorisation and accountability views as it has nothing to do with how the representative acts but depends on the personal characteristics of the representative.[30]

In many respects the Senate is more representative of the wider populace than the House of Representatives. Not only are more parties represented but also more minority groups. The only two Indigenous Australian parliamentarians were senators. Women occupy just 27 per cent of the seats in the House of Representatives but 36 per cent of the Senate seats (see Figure 29). The youngest women to be elected and sit in parliament were senators. The first Australians of Asian ethnicity to sit in parliament were senators and the only openly gay and lesbian parliamentarians are senators. The list goes on but the theme is the same—that a wider range of groups have been better represented in the Senate, or as some more radical proponents have suggested, the Senate is the real house of representatives.

Proportions of current female parliamentarians

Data source: J Wilson and D Black, Women parliamentarians 1921–2009, 4 May 2009, viewed at http://www.aph.gov.au/Library/pubs/BN/2008-09/WomenParliamentarians.pdf

Women are arguably the most underrepresented group, and differences between houses are replicated at a state level. In the five states with upper houses, four (NSW, Vic., WA and Tas.) currently have higher proportions of female members in the upper house. Only South Australia has a higher proportion of women in the lower house (and Queensland does not have an upper house).

One female parliamentarian claimed that conservative party leaders believed that women are not as exposed to a campaign in the Senate. A long-serving female parliamentarian argued that it would take a long time to achieve equal representation, as sitting members vacate seats and more women contest and win preselection. In this respect, Senate preselections for election ensure a higher turnover and more opportunities for women, due to the multiple positions available. Another major party senator contended that winning a lower house seat and as member of a minority group was more challenging, though not clearly impossible. However, it was not only (male) party powerbrokers who guided women towards the Senate, with some female politicians also preferring the chamber. One female parliamentarian saw the Senate as more attractive in terms of achieving a better work, life and family balance as electorate demands are less.

Geography provides an easy base upon which to construct representative institutions, but why is geography more important in Australia’s electoral design than gender, for example? A certain number of seats are reserved for certain states, sometimes over and above their population, but the notion of reserving half the seats in parliament for women, is contentious. As Dovi acknowledges:

Most theorists do recognise that members of certain groups must be present in democratic institutions in order for good democratic representation to take place. In other words, descriptive representation is considered necessary, albeit not sufficient, for good democratic representation.[31]

Yet it is dependent on the political actors—essentially the political parties—to recognise and act upon this, rather than through institutional changes. Furthermore, while Saward observes that in Australia and the United States some parliamentarians claim they represent constituencies such as women, lesbians and gay men and indigenous people, which transcend geographically defined electorates,[32] Dovi concedes that not every member of a historically disadvantaged group is committed to advancing the concerns of that group.[33] They are not necessarily the best representatives of the group and can indeed be hostile to such group identities. Furthermore, it assumes a high degree of homogeneity within the group.



[1].          D Shell, ‘The history of bicameralism’, Journal of Legislative Studies, vol. 7, no. 1, 2001, p. 13.

[2].          SC Patterson and A Mughan, ‘Senates and the theory of bicameralism’, in SC Patterson and A Mughan (eds), Senates: bicameralism in the contemporary world, op. cit., p. 12; J Uhr, ‘Generating divided government: the Australian Senate’, Ibid., p. 102.

[3].          Quoted in IC Harris (ed.), House of Representatives Practice (Fifth Edition), Department of the House of Representatives, Canberra, 2005, p. 469.

[4].          J Uhr, ‘Generating divided government: the Australian Senate’, op. cit., p. 94.

[5].          S Bach, Platypus and parliament: the Australian Senate in theory and practice, op. cit., pp. 102–3.

[6].          L Young, Minor parties … major players?: the Senate, the minor parties and the 1993 Budget, 1996 Parliamentary Fellow monograph, Parliamentary Library, Canberra, 1997.

[7].          See EE Hewitt, Cameral government: unicameral and bicameral legislatures, Viridia Books, Melbourne, 1992, pp. 129–35.

[8].          Ibid., p. 191.

[9].          At the time of Australia’s Federation, United States’ Senators were chosen by state legislatures.

[10].        J Uhr, ‘Generating divided government: the Australian Senate’, in SC Patterson and A Mughan (eds), Senates: bicameralism in the contemporary world, op. cit., pp. 107–8.

[11].        H Evans (ed.), Odgers’ Australian Senate Practice (Twelfth Edition), Department of the Senate, Canberra, 2008, pp. 90–1.

[12].        S Bach, Platypus and parliament: the Australian Senate in theory and practice, op. cit., p. 5.

[13].        S Bennett, The Australian Senate, Research Paper, no. 6, 2003–04, Parliamentary Library, Canberra, 2004, p. 2.

[14].        M Rodrigues and S Brenton, The age of independence? Independents in Australian parliaments, Forthcoming Research Paper, Parliamentary Library, Canberra.

[15].        V Bogdanor, ‘The problem of the upper house’, in HW Blom, WP Blockmans and H de Schepper (eds), Bicameralisme: Tweekamerstelsel vroeger en nu, op. cit., p. 412.

[16].        HW Blom, ‘Ethos and interests. Arguments for a representational differentiation in a changing society’, in HW Blom, WP Blockmans and H de Schepper (eds), op. cit., p. 150.

[17].        J Uhr, ‘Generating divided government: the Australian Senate’, op. cit., p. 116.

[18].        S Bach, Platypus and parliament: the Australian Senate in theory and practice, op. cit., p. 351.

[19].        D Shell, ‘The history of bicameralism’, op. cit., p. 17.

[20].        SW Couwenberg, ‘Bicameralism, historical background, alternative conceptions and current relevance’, in HW Blom, WP Blockmans and H de Schepper (eds), Bicameralisme: Tweekamerstelsel vroeger en nu, op. cit., p. 141.

[21].        G Tsebelis and J Money, Bicameralism, Cambridge University Press, Cambridge, 1997, pp. 11–2.

[22].        SW Couwenberg, ‘Bicameralism, historical background, alternative conceptions and current relevance’, op. cit., p. 141.

[23].        D Shell, ‘The history of bicameralism’, op. cit., p. 11.

[24].        HW Blom, ‘Bicameralism – history – theory – problems’, in HW Blom, WP Blockmans and H de Schepper (eds), Bicameralisme: Tweekamerstelsel vroeger en nu, op. cit., p. 22.

[25].        G Zappalà, Four weddings, a funeral and a family reunion: ethnicity and representation in Australian federal politics, op. cit.

[26].        Adams cited in S Dovi, The good representative, op. cit., p. 29.

[27].        Ibid., p. 29.

[28].        HF Pitkin, The concept of representation, op. cit., pp. 60–1.

[29].        Ibid., p. 87.

[30].        Ibid., p. 61.

[31].        S Dovi, The good representative, op. cit., p. 23.

[32].        Sawer 2001 and Mansbridge 2003 cited in M Saward, ‘The subject of representation’, Representation, vol. 44, no. 2, 2008, p. 94.

[33].        S. Dovi, op. cit., p. 34.

Back to top