Document Details
Conference Paper
Publication Year:
The Australian Senate

The Australian Senate
Professor John Uhr
Australian National University
Paper presented for roundtable
“Senate Reform: Options and Comparative Perspectives”
Sponsored by the Institute for Research on Public Policy
and the Forum of Federations
November 18, 2008
Sheraton Ottawa Hotel
Ottawa, Ontario
This document is a preliminary draft.
1. Introduction
1.1 The Australian Senate with 76 elected members is one of two houses of the Parliament of the
Commonwealth of Australia. The other elected house is the 150 member House of
Representatives. The Australian Constitution establishes the Commonwealth of Australia as a
federation of six States, with the Senate composed of an equal number of senators from each
State. Each State forms one multi-member constituency, with senators serving for fixed six year
terms. House members represent single-member seats distributed nationally according to
population, and serve three year terms, subject to early dissolution by the prime minister. Members
of both houses are elected on the same franchise and are subject to the same qualifications of
office. Australian electoral practice is historically distinctive: voters have long operated under a
system of compulsory voting, with a range of voter-friendly measures (in reality: ‘party-friendly’
measures) implementing ‘preferential’ voting (often called ‘the alternative vote’ outside Australia)
for both houses, and a ‘proportional representation’ (effectively, a list system using the ‘single
transferable vote’ method) for Senate elections.
1.2 The Australian Constitution sets out the powers of the two houses. The two houses share
‘legislative power’, with the Senate having virtually equal legislative powers with the House of
Representatives. Although there are restrictions on what types of laws the Senate may introduce
(eg, appropriation or taxation bills) or amend (eg, appropriation for the ‘ordinary annual services’ of
government), there are no restrictions on the power of the Senate to reject bills. The Constitution
contains deadlock-resolving provisions involving a ‘double dissolution’ (ie, a dissolution of all
members of both houses) with the prospect of a subsequent ‘joint sitting’ of all members to
determine the fate of disputed measures. By convention, the House of Representatives is often
referred to as ‘the house of government’ and the Senate as the ‘the house of review’. Another set
of terms has the House called ‘the house of the people’ and the Senate ‘the house of the States’.
The prime minister and most of the ministry have historically been drawn from the House of
1.3 The Parliament itself has legislated to ‘fill out’ many of the constitutional provisions, notably in
relation to the system of electoral representation used by each house. For example, the number of
senators for each state has been expanded from the original six to the current twelve, and two
senators have been allowed to be elected for each of two mainland Territories: the Australian
Capital Territory and the Northern Territory. The Senate system of proportional representation
introduced by ordinary statute for the 1949 elections has broadened the range of political parties
and independents represented in the Senate. The constitutional provisions relating to Parliament
may only be altered through referendum: from time to time, governments attempt to use referenda
to curb the powers of the Senate but historically only eight of 44 referenda proposals have been
successful, with some impact on Senate operations but no real diminution of the Senate’s powers.
2. Aim and Scope
2.1 This paper is a report on ‘the Australian Model’ of a Senate. The aim of the report is to clarify
the nature of the Australian Senate for a Canadian audience curious about lessons that ‘the
Australian Model’ might have for Canadian discussion of reform of the Canadian second chamber.
2.2 Given that the report is designed for oral delivery to a non-specialist audience, I have tried to
keep the text relatively uncluttered with historical or statistical data, preferring instead to provide a
snapshot of current operations of the Australian Senate. Wherever possible, I use examples to
illustrate general trends, suggesting where readers find more comprehensive data which helps
explain how these trends have arisen. I use selected averages rather than provide historical runs of
data, hoping to give a flavour of ‘the Australian Model’ without trying to explain institutional history
or the Senate’s many interesting variations year by year.
2.3 My basic aim has been to provide a Canadian audience with a picture of how the Australian
Senate impacts on Australian national politics. My focus is on the institutional qualities that the
Australian Senate brings to Australian national politics, so that readers can get a reliable first sense
of what sort of impact ‘the Australian Model’ of a Senate makes to parliamentary politics as
practiced in Australia. As I note in the next section, we can not assume that ‘the Australian Model’
would have similar impacts if adopted elsewhere.
3. Three Qualifications
3.1 A number of qualifications should be mentioned at the outset. First, this is a personal report by
an academic analyst and not an authorized statement by an official spokesperson reflecting the
institutional interests of the Australian Senate. The Australian Senate does have something of an
authorized version defending its contribution to Australian parliamentary democracy: Odgers’
Australian Senate Practice, which is a large manual of parliamentary practice now in its 12th
edition, providing readers with unrivalled knowledge about Senate history, procedure and practice.1
Although I write as an admirer of the Australian Senate, this report is an exercise in advice rather
than advocacy: advice about how the Australian Senate impacts on Australian politics, rather than
advocacy about the export potential of ‘the Australian Model’.
3.2 Second, and building on this first qualification, I note that ‘the Australian Model’ is an Australian
response to Australian problems, with possible lessons for other countries but probably very few
easy or non-controversial applications to non-Australian circumstances. Put simply: ‘the Australian
Model’ is not designed along the lines of any other model, and it is unlikely to perform well as a
model for other countries, even so-called Westminster countries, to try to replicate. Australian
parliamentary commentators have increasingly rejected the terms and categories of ‘the
Westminster system’ because Australian political practices do not really resemble those of classic
Westminster. The presence of an elected Senate in a constitutionally-entrenched federal
parliament is far from classic ‘Westminster’. True enough, many governments of the day appeal to
Westminster norms when trying to justify the prevailing power of the political executive in what is
loosely called a regime of ‘responsible government’. Also true is that opposition parties often
appeal to ‘Westminster’ norms to justify an increased share of parliamentary power by nongovernment
parties. The fact that Australian governments so rarely share significant parliamentary
power with opposition parties suggests the limits of the ‘Westminster’ analogy for Australian
politics. Bicameralism has many virtues; so too, bicameralism has many varieties, allowing
Canadian and Australian students of parliamentary democracy to learn from the experiences of
each other. Both are capable of adapting and innovating in ways that overcome many of the
deficiencies of grafting alien institutions on native stock.
3.3 Third, the current characteristics ‘the Australian Model’ have developed or grown up in the 107
years since Australian Federation in 1901, reflecting the work of many generations of parliamentary
actors. These actors built on the constitutional foundations spelt out in the 1901 Constitution for the
Commonwealth of Australia but often in ways not necessarily anticipated by the constitutional
framers. Although the black-letter provisions of the Australian Constitution might not have changed
all that much since 1901, the practical operations of the Australian Senate most certainly have
changed. These institutional changes have been driven partly by changes in parliamentary law on
such core operational issues as electoral mechanisms and driven partly by changes in the
parliamentary ambition of the political parties competing for place and power in Australian politics.
3.4 History certainly matters; but accidents of history probably also matter. It is possible that
elements of ‘the Australian Model’ rest on accidental developments or, more likely, unintended
consequences of almost forgotten developments. Even the Australian embrace of proportional
representation is something of a happy accident, with its Labor initiators in the late 1940s unaware
of many of its potential effects.2 The practical implication of this historical blend of intention and
accident is that the current version of ‘the Australian Model’ is such an amalgam of law and politics
that observers are uncertain how particular elements of the model (eg, Senate Estimates hearings)
might operate as stand-alone features taken out of their historical context.
3.5 Why labour these qualifications? The answer is that ‘the Australian Model’ is a work in
progress. This report tries to convey an accurate picture of current operations. Canadian audiences
requesting similar reports in earlier decades would have received very different reports, depending
on which wave of institutional change was occurring at the time of the report’s preparation. Thus
readers of this report should bear in mind that ‘the Australian Model’ being discussed is in reality
the current version of a model that has been developing for over a century. Further, many of the
current features reflect often-forgotten decisions by earlier generations of political innovators or
reformers. In other cases, current features derive from attempts to respond to problems deriving
from earlier generations of institutional innovators and reformers. Thus ‘the Australian Model’ is far
from being a kit of deliberately crafted components that one can, after importing and assembling,
expect to operate in Canadian circumstances exactly as it does in Australian circumstances.
4. Canada and Australia
4.1 Canada and Australia deserve the close comparison they receive. Canada and Australia share
many aspects of the larger heritage of parliamentary government. Both were British colonies
attracted to the promise of responsible parliamentary government around the mid-19th century.
Both are federations. Both are members of the Commonwealth of Nations. Both are constitutional
monarchies. And both have had to struggle for many of the rights of self-government. Canada as
the older British colony was something of an inspiration to 19th century Australian colonists:
‘Canada Bay’ in Sydney is named in honour of the Canadian colonists who took temporary refuge
in Sydney after the initial failure of the upper Canada struggles for self-government. Both countries
have a long history of stable parliamentary government at both national and provincial/state levels,
including early reliance of second chambers at provincial/state level.
4.2 But the historical developments diverged at some point, with the Australian colonies/states
showing greater interest in modernizing and democratizing their second chambers. By contrast,
Canadian provincial second chambers were discarded: a process that only one Australian state
(Queensland) has followed.
4.3 The relevance is that the Australian Senate, unlike the Canadian Senate, is nested in a
framework of bicameral Australian parliaments. Over recent decades, many of the Australian state
second chambers have been further reformed to resemble ‘the Australian Model’ pioneered by the
Australian Senate. Thus the Australian Senate should be understood as part of a larger package of
bicameral arrangements in the Australian federation. Australian political parties have learnt to use
bicameralism for their own purposes: the existence of second chambers is accepted a part of the
institutional environment of parliamentary politics and is presumably welcomed by parties,
particularly as it increases opportunities for paid public office open to political activists. I know of no
evidence of popular interest in reform of the Australian Senate, certainly not in repeal of the Senate
or State second chambers, although there is evidence of popular suspicion of sustained
government interest in reform of the Australian Senate. I will say more about this in the Conclusion.
5. A States House?
5.1 One of the major misconceptions relating to the Australian Senate is the contention that the
Senate has somehow failed to live up to supposedly-original intention of acting as a ‘States House’.
The claim is that the primary purpose of the Senate was to inject State-wide blocs of State
representatives into the national Parliament and that these State-wide blocs would be expected to
protect their respective States’ interests by voting en bloc as State delegates. While it is true that
the Senate has never (or very very rarely) voted along State lines, and while it is true that party
divisions quickly arose as the predictable sources of division within the Senate, it does not
necessarily follow that the Senate has ‘failed’ as a States House.
5.2 First, the Senate does provide for equal representation of each State and this constitutional
equality strengthens the political representation of the smaller and hence more vulnerable States.
These smaller States receive a greater number of parliamentary representatives than they would
deserve solely on the basis of representation by population. Second, each of the major parties of
government draws into its federal party caucus a greater number of representatives from the
smaller States than they otherwise would without a Senate. Thus the Senate broadens the State
representation of the major political parties. Third, the standard misconception gets the original
intention wrong. The original intention was to have the Senate promote States interests not through
uniformity of voting but through diversity of views represented within each State body of senators.
5.3 The Constitution was written by serving politicians who fully appreciated the rising power of
party and of the normality of party competition in a emerging system of party government. But they
also appreciated the facts of political geography and knew that the national Parliament needed to
know the diversity of views within each State if the Parliament was to contribute to the new federal
Commonwealth. Fourth, the very idea of a Senate was favoured by many early federalists on the
assumption that proportional representation would make the second chamber a distinctive house of
minorities. Just as the equal representation of each State in the Senate would protect the minor
States, so too proportional representation would protect minorities within each State body of
senators. This frequently-forgotten version of the Senate as a States house is in many ways the
basis of its greatest enduring public legitimacy.3
5.4 Arguments over the Senate as a States House eventually come face to face with the fact that
the Senate has developed very much as a party house, and more particularly as a State party
house. That is, State party officials tend to dominate who gets elected to the Senate. They exercise
this power through their selection of who gets nominated on the State party list. Current electoral
arrangements allow, indeed encourage, voters to elect senators by endorsing the party-ticket of
their preferred party, right down to that party’s often-undisclosed order of ‘preferences’ as required
under the Australian system of preferential voting. Voters have the option of ranking their
candidates according to whatever merit ranking the voter favours. But the political parties do all
that they can to encourage voters to limit their involvement to authorizing their favoured party’s
internal rank order of candidates.
5.5 One important consequence of this party-list development is that the Senate can be seen from
the perspective of political parties as something of a nominee house: voters get to authorise those
on their favoured party list, even though they might never hear of many beyond those near the very
top if the ticket. Of course, similar observations could be made about House of Representatives
elections, to the extent that voters tend to vote for party labels rather than known candidates and
so simply authorize choices made by party officials. But in lower house elections, voters tend to
see more of the small number of candidates competing in their riding (or ‘division’ in Australian
language) and are better placed to form their own view of the who is the one candidate best
qualified to be their representative. Such calculations are considerably more difficult to do on the
basis of reliable knowledge in Senate elections when voters are typically electing not a single
representative but six representatives. Thus it is easier for many voters simply to take on trust the
rank ordering determined by their preferred party, which illustrates the considerable power of the
State party officials who determine who (or what factional representative) gets to stand for Senate
6. A Current Snapshot
6.1 I want to turn now from what the Senate is to what the Senate does. I think it might be useful to
use current examples to convey the ‘feel’ and ‘presence’ of the Australian Senate and the ‘spirit’ of
its approach to parliamentary politics.
6.2 The current Rudd Labor government was elected in November 2007 when it defeated the
conservative Howard government, which had won a rare double majority in both parliamentary
houses at the previous 2004 election. Apart from the Howard government in its fourth and last term
in office (2004-2007), no Australian government in the last 30 years has enjoyed a Senate majority.
The distinctive Senate electoral system of proportional representation has the effect of denying
either of the two major party blocs (Australian Labor Party; the Liberal-National coalition parties) a
Senate majority. Typically, governments do not have a majority of Senate seats to guarantee
passage of their own initiatives; equally, the official opposition does not have a majority of votes to
get its own way. With neither of the two major party blocs enjoying majority power, the ‘balance of
power’ typically falls to the third parties: the so-called ‘cross benches’ comprising the minor parties
and independents who manage to win Senate seats through the remarkable fairness of
proportional representation which allocates seats proportional to the share of votes.
6.3 The 2007 election of the Rudd government restored the Senate to its usual non-government
majority. Remember that this situation rarely if ever means a Senate majority for the official
opposition. The 76-member Senate currently comprises: 32 government senators; 37 opposition
senators; and 7 cross-bench senators (5 Greens; two independents). Unlike the lower house
Speaker, the Senate President (by convention, a government senator) has no casting vote,
consistent with the strict reading of federalism as meaning equal voting power of each state. Half of
the total Senate 76 votes (ie, 38 votes) is sufficient to block a measure. Any party wanting to
secure passage of its initiatives requires one more than half of the Senate votes: 39 votes.
6.4 To wield a bare winning majority of 39 votes, the current Rudd government needs all seven
cross-bench votes. The current opposition can block government initiatives by gaining one
additional vote (38 votes) but it needs yet another vote to get a majority of 39 votes for passage of
its own initiatives. The situation of dispersed power is equally demanding for the minor parties. The
Greens are the largest of the cross-bench forces and they need two more votes than the ‘greenfriendly’
government can provide in order to get a Senate majority in favour of Greens’ initiatives.
The Greens could secure a majority with the support of the official opposition, but this would
involve an unusual blending of right and left political orientations: not impossible but not what either
orientation would initially favour. And then there are the two independents: either of the two
independents can join forces with the opposition to secure 38 votes to block government initiatives;
both can join forces and provide the official opposition with the required 39 votes for Senate
passage of opposition initiatives.
7. Budget Blues
7.1 Time now for some recent examples of the Senate’s impact on Australian national politics. I will
highlight examples of the Senate’s response to the Rudd government’s first budget. This story tells
a larger tale about the Senate’s impact on government law and policy.
7.2 The parliamentary side of the Australian budget process begins with the Treasurer’s budget
speech in early May, in anticipation that Parliament will pass the budget as soon as possible in the
new financial year which begins on 1st of July. Passage through the House of Representatives is
generally smooth because the government of the day holds office by virtue of its House majority,
which Australian governments are not shy to use. The trick is getting the budget smoothly through
the Senate, which under the Constitution has no time limit within which to pass legislation,
including government budgets. Governments have learnt to tolerate a fair degree of delay in the
Senate, because they know that both major party blocs use their time in opposition to use the
budget process (particularly the 40 year-old practice of Senate estimates committee hearings) as
their primary opportunity to hold the party in government to account.
7.3 What goes around, comes around; and both major party blocs have learnt to use this power of
delay in ways that generally fall short of what public opinion might see as willful and irresponsible
obstruction. The line between open accountability and willful obstruction is, of course, a matter of
political convention rather than black-letter constitutionality. In the history of the Australian Senate,
the year 1975 stands out as the year constitutionality trumped convention when the Senate
delayed considering the government’s budget to the point of deadlock, triggering the Governor-
General’s intervention to dismiss the Whitlam Labor government on the basis that it lacked
parliamentary confidence.4 The opposition took office as caretaker government and resoundingly
won the subsequent election demanded by the Governor General. ‘The Dismissal’ of 1975 is an
atypical example of the Senate’s impact on Australian politics. Let me now proceed to give a few
examples of the typical forms that Senate impact takes, drawing on examples from the last few
7.4 The simple version of the current Senate story is that the Senate continues to have a significant
impact on government legislation. By ‘significant’, I mean that the Senate has recently made
repeated amendments to the Rudd government’s package of budget bills: not ‘money bills’ or
supply as such, but budget measures introduced as part of the government’s overall budget
package.5 This tendency towards challenging or even amending budget measures was initially
pursued by opposition senators whose unusual period of Senate mastery did not come to an end
until the newly-elected senators took up office from July 2008. But the tendency was reinforced by
‘the new Senate’ where the balance of power was held by the Greens and two independents. With
the budget still under legislative consideration by the Senate, a number of prominent budget
measures suffered at the hands of the non-government forces. Bills were often defeated at second
reading: for instance, a national health taxation measure was defeated on 28 August; a package of
bills to increase taxation on luxury cars was also defeated on 4 September (later passed with
amendments on 17 September); a medicare levy surcharge bill was also defeated on 24
September, although later passed on 16 October following cross-party agreement on a
compromise package of amendments.6
7.5 Other budget measures were passed but only after amendment, including amendments that
take the form of ‘requests’ to the House of Representatives in those cases where the Constitution
places limitations on the Senate’s capacity to amend directly: eg, the government’s budget
measure to remove excise exemptions for a range of fuel condensates. This constitutional
limitation on the Senate’s power to amend taxation bills is contained in s53 of the Constitution. The
cryptic words of the Constitution have provided hours of enjoyment (and years of employment) for
constitutional lawyers. Government lawyers usually take the strict interpretation that any bill relating
to taxation may not be amended by the Senate, although the Senate is in its rights to ‘request’ that
the House of Representatives amend such bills. And so the Senate does.
7.6 But a more radical challenge to conventional interpretations of s53, and to the Rudd
government’s budget, came when the non-government parties in the Senate engaged in their own
budget-making exercise by passing legislation to increase the age pension. Section 53 states in
part that appropriation or taxation legislation ‘shall not originate in the Senate’. The Senate passed
this non-government pension bill on 22 September 2008, with its supporters claiming that the bill
itself did not appropriate money but simply increased the rate of age pensions which were formally
appropriated under standing provisions (or ‘special appropriations’) in existing social security
legislation. The Rudd government argued that the House of Representatives was under no
obligation to consider the Senate bill because it was ‘unconstitutional’. The Speaker of the House
of Representatives tabled advice from the Clerk of the House of Representatives supporting the
government’s contention that the House was under no obligation to consider the Senate bill
because it was ‘not in accordance with the constitutional provisions’ of s53. The opposition in the
House of Representatives has little if any opportunity to debate the Speaker’s ruling as the
government used its numbers to close debate, which had the effect of dividing the House on the
Speaker’s ruling, to the convenience of the government.7
7.7 Beneath the surface of partisan dispute over the government’s parliamentary tactics lurked a
deeper dispute over the clashing opinions of the two parliamentary Clerks, with each bringing
forward evidence to support the constitutional claims of their own house. A careful reading of the
Clerks’s opinions would show just how difficult it is to eliminate administrative no less than policy
disagreement from bicameral parliamentary arrangements.8
8. Impact on Legislation
8.1 I turn now to a more general review of the impact of the Senate on Australian parliamentary
politics. The first and in many ways most characteristic impact is on the legislative process,
particularly on the fate of government legislation. After briefly dealing with the Senate’s legislative
impact, I shall look even more briefly at two related areas of Senate impact: on government
accountability; and on parliamentary representation. These three areas of legislation, accountability
and representation are three central testing-grounds for evaluating the effectiveness of any
parliamentary system. The comments here are only the modest first steps towards an evaluation of
the institutional effectiveness of the Australian Senate.
8.2 When considering the Senate’s impact on the legislative process, the starting point is to
recognize that in most years the Senate passes around two-thirds of government bills without
amendment. The Senate’s impact on the content of these non-controversial bills might well be
considerable, causing governments to anticipate non-government interests and to modify their own
initial drafting. That is, the very fact that Senate consent is required for legislation is itself sufficient
for governments not to introduce bills or provisions in bills that have no prospect of ‘getting through
the Senate’. Critics like to point out that approximately one-third of the government bills that are
amended are altered through changes moved by the government itself. But who moved the hand
that moved the amendment? As is well known, many government amendments take up issues
originally raised by non-government interests and are to that extent involuntary or enforced actions.
The conclusion is not black or white but grey. On the one hand, most of what governments want,
governments get. On the other hand, much of importance to non-government parties is also
secured through that very process of government adoption of non-government interests.
8.3 Remember that governments need Senate assent to all their legislative initiatives. Not
surprisingly then, most of the formal time available to the Senate is spent in what is classified as
‘government business’: primarily the passage of government legislation. In the years since 2001,
around 52% of the Senate’s timetable has been devoted to ‘government business’. This figure
nicely illustrates one of the fundamental functions of the Senate, which is to process whatever the
government wants processed, although not necessarily in ways or with results favoured by
governments. The Senate has passed on average 165 bills each year, almost all being government
bills. On average, 67 bills each year are referred to one of the Senate’s set of eight investigative
committees for inquiry and report. These are inevitably the bills that go on to attract amendments,
often although not always as consensus recommendations from the relevant committee.
8.4 Instead of providing comprehensive data on the Senate’s record of impact on proposed
legislation,9 I simply want to contrast two recent years in order to highlight the general story of
Senate legislative activism. This year’s legislative record is still incomplete. Last year was an
election year when the record is typically truncated because of a smaller number of sitting days. So
2006 self-selects as the most recent full year, except that it happens to be one of those rare years
when the Howard government had a majority in the Senate, so we might expect evidence of a
tame Senate. We can compare 2006 with 2003, the last non-election year before the arrival of the
rare Howard double majority. The two-year contrast is instructive.
8.5 Sure enough, the 2006 record shows no success in relation to any of the 39 second reading or
‘policy’ amendments moved, mainly by the then-Labor opposition. But when we look at the
subsequent ‘committee of the whole’ stage of the legislative process dealing with the details of
proposed legislation, we find a different story with evidence of Senate capacity and will to amend
many government bills. In 2006, the Senate dealt with 218 bills, 75% of which were government
bills. Around 45% of government bills were referred by the Senate to one of the eight ‘standing’ (or
subject-matter) committees for inquiry. Around 80% of bills passed both houses, some in amended
form. The Senate agreed to amendments in the case of around 15% of the bills under
consideration: a relatively low figure but remember that this reflects the brief period of Howard
government control of the Senate. Many legislative amendments originated as government
proposals: in fact, around 92% of successful amendments were government amendments: nongovernment
senators moved no more than 8% of committee stage amendments.
8.6 Contrast this story with the year 2003, a more typical year well before the Howard double
majority. The overall pace of legislative work is about the same, with 215 bills passing both houses,
compared to the 218 figure in 2006. But there are some interesting contrasts with 2006. For
example, whereas 2006 had only one bill caught in fundamental disagreement between the two
houses, in 2003 there were 25 such disagreed bills: that is, bills with no agreed resolution that
year. That is a significant proportion of a government’s legislative program. And instead of there
being 15% successfully amended bills, in 2003 there were nearly 30% amended bills. And most
interesting of all, the committee stage evidence shows a higher proportion of successful to
unsuccessful amendments, with 55% of proposed amendments being successful. The contrast is
highlighted when we notice that 2003 included 17 successful Senate ‘requests’ to those bills which
the Constitution says the Senate may not amend. There were no requests, successful or
unsuccessful, in 2006.
8.7 Of course, most Senate amendments are moved by the government, even though the
government rarely enjoys a majority in the Senate. This tells us that governments can read the
writing on the wall and do what they can to direct and steer the legislative momentum. What
happens to Senate amendments when they return to the House of Representatives? The fact that
most amendments are moved by the government would suggest that the House would accept
these amendments. True enough: in nearly 80% of the cases over the last decade, the House has
accepted the Senate amendments.
8.8 But what happens in the other 20% of cases when governments refuse to accept Senate
amendments? Stanley Bach is the latest authority on this topic whose recent research puts the
Senate’s power into fresh perspective.10 Reviewing the last decade or so of Senate amendments to
government legislation, Bach contrasts the high rate of Senate amendments with the interesting
pattern that emerges from the way the Senate reacts when the House (ie, the government of the
day) refuses to accept Senate amendments. In many such cases, governments simply stick to their
guns and do not counter-propose alternative amendments; and in most such cases, the Senate
yields. In other cases, where the government counter-proposes with alternative amendments, the
Senate also typically yields. Generally, the Senate either does not contest when a government
overrides Senate amendments.
8.9 This situation is a classic case of the glass being half full or half empty. Looked at from a
parliamentary perspective, one has to admire a second chamber that can secure the first
chamber’s support for 80% of its amendments. But looked at from a Washington perspective,
which Bach brings to the Australian scene, one wonders why the Senate does not hold its nerve for
the other 20% of the time. From Bach’s perspective, one finds ‘evidence of regrettable institutional
reticence’. As Bach writes: ‘…the pattern of Senate acquiescence…also may reflect persistent
doubts among many Australians, and perhaps some Senators, about the constitutional propriety of
the Senate’s determined exercise of its constitutional powers to legislate, especially in ways that
challenge the government’s legislative agenda…Some Senators may have believed, and may
continue to believe, that mounting regular and determined challenges to government legislation
might produce public anger and could even violate their own sense of the what the Senate’s
appropriate role in the legislative process should be’.11
9. Impact on Accountability
9.1 We turn now to the second arena of Senate impact, which is impact on government
accountability. Again, I begin with a current example. One of the most heated topics in recent
Australian debates over government accountability has been the unaccountable power of the
private office employed by government ministers. Australia has seen extraordinary growth in the
numbers of staff working directly for government ministers. The unaccountable power of ministerial
staff became an election issue in the last few national elections. Labor in opposition promised to
rein in the power of ministerial staff, based largely on the evidence of unaccountable power
unearthed by Senate committees of inquiry. Labor in the Senate made the running on this issue, so
it comes as now surprise that Labor senators in government have been at the forefront of
regulating the conduct of ministerial offices. The Rudd government introduced a new code of
conduct for ministerial staff on 26 June 2008.12 The fact that this initiative was promoted by a
senior government Senate minister is telling. Senator Faulkner is a former Leader of the Opposition
in the Senate and enjoyed high public prominence as a critic of the former Howard government’s
misuse of ministerial staff. When in opposition, Labor promised to bring greater public
accountability to ministerial staff, and the new code of conduct delivered on that promise.
9.2 How can we gauge the impact of the Senate on government accountability? One simple but
useful way is to track the record of Estimates hearings by the Senate’s eight ‘standing’ committees
responsible for holding hearings on the ‘estimates’ included in the government’s annual budget.13
These estimates hearings occur twice-yearly and allow non-government senators a welcome
opportunity to ask for whatever evidence from the heads of the public service they or their
committees consider necessary for them to make an informed decision about whether to support
the government’s proposed budget. The interesting dimension is that Senate estimates have
extracted commitments from ministers that senior public servants (but not ministerial staff
employed in the private office of government ministers) are expected to appear as witnesses.
Given the Senate situation, many of these witnesses will appear in the absence of their
departmental minister, who might well be a member of the other parliamentary house. After all,
two-thirds of the ministry are located in the House of Representatives. Hence, many public
servants appear alongside a duty minister from the ranks of the Senate ministry: such ministers
might or might not be across the policy issues facing the departmental witnesses.
9.3 Perhaps none of this really matters in strict legislative terms: the rude fact is that, to my
knowledge, no estimates hearing has ever generated a determination by an Estimates committee
that a government’s budget legislation should be amended. The political purpose of the estimates
exercise is not so much to examine the proposed budget as to use the formal legislative authority
of the Senate to hold governments accountable for alleged policy and administrative deficiencies,
independently of the merits (or otherwise) of the proposed budget. So what do Senate estimates
committees focus on when exercising their accountability functions? Perhaps surprisingly, they
focus on many matters that go very close to the heart of the norms of responsible parliamentary
government. A good example is the provision of policy advice to the political executive, which might
be thought to be an area out of bounds for parliamentary committees when examining the public
9.4 Observers of Estimates hearings note that across the eight committees which share
responsibility for this core business, one common theme is the reluctance of government witnesses
(ministers as well as officials) to report openly on the nature of public service to government.14 One
particular committee, the Senate Legal and Constitutional Committee, tracks this issue with regular
reports on the stand-off between the government and the Senate over policy advice given by the
public service to government ministers. Earlier this year the public service secretary to the
Attorney-General’s Department tabled advice from the head of the Department of Prime Minister
and Cabinet to the effect that public service advice was confidential except when ministers might
decide to reveal it.15 The Senate regularly acknowledges that in many if not most cases, public
service advice will indeed be protected by public interest immunity; but the Senate’s point has been
that the mere claim to a general public interest immunity is hardly persuasive evidence of the
merits of a specific claim.
9.5 But there is a deeper underlying trend towards direct parliamentary accountability of senior
public servants. Over more than 40 years, the Estimates hearings have reframed accountability
relationships between administrative officials and Parliament. In tension with the reluctance of
ministers to table public service policy advice is the increasing openness of public servants to
speak their own minds on matters of public service professionalism that are not confidential to the
inner workings of government. Observers note the rise of public service forthrightness in Estimates
hearings, where chief executive officers are displaying increasing boldness in revealing and indeed
explaining their own professional opinions. As a Senate procedural digest reports after a recent
round of Estimates hearings: ‘A notable feature of the hearings was the outspokenness of several
senior officers and their defence of their roles: the Secretary of the Treasury vigorously defended
his right to speak on economic issues; the Auditor-general spoke very frankly about the effect of
budget constraints on the ability of the Audit Office to perform its essential tasks…the Chief of the
Defence Force was as frank as ever, stating that foreign forces would be required in Afghanistan
for ten years’.16
9.6 A prominent recent example of how the Senate Estimates hearings can impact on the
accountability of government officials occurred last month, when the public service head of the
Treasury made one of his regular twice-yearly appearances before the relevant Senate committee
investigating his department’s budget estimates. The hearings are free to canvass any matters that
the Senate considers relevant to making decisions about the government’s expenditure plans on
the basis of informed consent rather than blind judgment. Over time, the Senate has made clear its
own expectations that public service heads should appear, to provide what has come to be called
‘the factual and technical background’ to government policy and practice. Provision of this type of
policy information can and should be distinguished from defence and justification of government
policy and practice, which is a task more properly expected of government ministers. Needless to
say, questions and answers about government practice can have very large policy ramifications.
The degree of difficulty is substantially increased when, as in this case, administrative officials
appear alongside a duty minister drawn from the ranks of the government’s crew of Senate
ministers, as distinct from the officials’ own departmental minister. The Treasury minister is by long
convention a very senior member of the House of Representatives, so that in this case, the public
service head of the Treasury was appearing alongside a non-Treasury minister who was at the
table providing a ministerial presence for a range of public agencies that were not part of his official
9.7 To cut to the chase, the head of Treasury spent 8 hours on 22 October in one long day
appearing before the Senate estimates committee.17 The big issue had little or nothing to do with
specifics provisions of the appropriation bills formally under examination. The issue of concern to
non-government senators who dominated the hearings was the Rudd government’s justannounced
relief package in response to the international credit crisis, and in particular the roles of
the Treasury bureaucrats and appointed heads of the central bank in advising the Rudd
government on policy options. Fuelling the hearing was the media, which strengthened opposition
doubts about ‘ministerial misspeak’: leaks published in the daily press suggested discrepancies
between what that the Rudd government was saying about acting on public service advice and
what unauthorized disclosures revealed about the rather different state of public service advice.
The Treasury head was caught between explicit claims by the Treasury minister and implicit claims
attributed in particular to the central bank, whose head was quoted in ways conveying a picture of
scepticism of the Rudd government strategy.
9.8 One final area of Senate impact on government accountability relates to the Constitution itself.
One could argue that the Constitution is the anchor of accountability for all government operations.
But what role does or should the Senate have in holding governments firm in their compliance with
the Constitution? Section 53 includes the provision that the Senate may not amend bills
appropriating money for ‘the ordinary annual services of the government’ (that is, core ‘supply’). It
is not hard to imagine governments being tempted to tack other budget provisions onto an ordinary
annual services bill in order to avoid Senate amendment. Certainly, there is evidence that Senate
Estimates committees often suspect that matters classified by government as part of the ordinary
annual services are misclassified in ways that are designed to deflect Senate scrutiny, especially
scrutiny of new policy proposals that are, strictly speaking, ‘extra-ordinary’ and ought therefore be
open to Senate scrutiny.
9.9 For example, early in the life of the Rudd government, the government introduced bills
appropriating monies to deal with an outbreak of equine influenza, claiming that these bills were for
‘ordinary annual services’ and therefore not amendable by the Senate. This claim attracted Senate
attention and was openly rejected by the relevant Senate Estimates committee as an unwarranted
use of the constitutional language of ‘ordinary annual services’.18 The Senate Finance and Public
Administration Committee tracks this issue and reports regularly on differences in interpretation
between the government and the Senate over this core constitutional concept. In a report tabled on
20 March 2008, this committee repeated the traditional Senate claim that ‘ordinary annual services’
does not include new policies and new projects. The government seemed sufficiently worried by
the entrenched nature of the Senate’s claims that it has taken on a retiring senator (former senator
Andrew Murray) as an adviser on this contentious constitutional issue.19
10. Impact on Representation
10.1 Third in this brief trilogy is a comment on the impact of the Senate on parliamentary
representation. Linking with the comments above on government accountability, I should begin by
noting the substantial representation of the government frontbench in the Senate. Labels like
‘house of review’ can mislead with pictures of non-government reviewers independently going
about their work of policy scrutiny and administrative oversight. Over the years, there have been
reform calls to free the Senate of government ministers but the governing parties have shown no
interest in denying their senators the chance of ministerial office. The current Rudd government is
typical in drawing a third of its cabinet ministers from the Senate. These are not junior ministries
but include (using their short titles) the cabinet secretary, the minister for climate change, the
minister for immigration, the minister for communications, the minister for industry, and the minister
for human services. Of course, the House can claim that it has twice as many cabinet ministers as
the Senate. Interestingly, this relationship of two to one reflects the constitutional ‘nexus’ provision
which holds that the House of Representatives should have twice the membership of the Senate.
But senators can put this in more positive terms by stating that the Senate contains half the
number of ministers as the House.
10.2 The figures for the current opposition are almost identical. In addition, one-sixth of the
government’s parliamentary sectaries (junior ministers) come from the Senate. Figures for the
opposition are even starker, with two-thirds of their parliamentary secretaries coming from the
Senate. Put differently, slightly less than one third of the government’s 32 senators hold executive
office (9 of 32). Once again, figures for the opposition are even more revealing, with slightly less
than half of the their Senate membership holding a position in the shadow executive (16 of 36). To
drive home my point, I note that almost exactly one-third of senators serve in executive offices,
defined as membership of either the political executive of the governing party or the shadow
executive of ‘the alternative government’. This itself has a very real impact on Senate
effectiveness, as two-thirds of the relatively small parliamentary house of 76 members are then
expected to do the lion’s share of work on Senate committees.
10.3 Obviously, the Senate impacts very directly by increasing the representation of the smaller
States in the Commonwealth Parliament. So too, the Senate’s inclusion since 1975 of two senators
from the two mainland Territories increases the representation of those regions. Traditionally, the
Senate has better represented women than has the House of Representatives, as one would
expect from a system of proportional representation which allows the party hierarchy to promote
female candidates with less risk of negative voter reaction than with local contests. Currently,
around 35% of senators are female and around 27% of House members are female: the patterns
are slowly converging but the historic leader has been the Senate in the years since the adoption
of proportional representation. Neither house has special provisions for indigenous representation.
The Senate has however had a number of indigenous representatives as members of political
parties with Senate seats.
11. How so much comes from so little
11.1 The real puzzle about the Australian Senate is how such a small house can try to do so much
and so often achieve so much. The answer is that the basis of the Senate’s functionality is its
public legitimacy. The Canadian and Australian Senates do not differ all that much in terms of their
formal constitutional powers. What is clearly distinctive about the Australian Senate is not so much
its busy involvement in the three domains of law-making, accountability and representation as the
high public expectations that the Senate provide the checks and balances otherwise missing in
Australian parliamentary government. To exaggerate for effect: senators know that they occupy
valued and highly-honoured public offices, that they are elected to perform as agents of open
government and public accountability, and that they are important bridges facilitating open traffic
between executive government and wider community interests. This fascinating situation came
about because of a century-long convergence of expectations inside and outside of the Senate:
institutional leadership from within by activist politicians got the Senate involved in legislative and
policy scrutiny; and demands by outside interests for rights of parliamentary access widen the
public legitimacy of the Senate’s step-by-step pursuit of open government and public
11.2 One can read this story in many places, but the one that I will mention here is the Labor
narrative of its love/hate relationship with the Senate. Labor at Federation was suspicious of, if not
hostile to, the prospect of a Senate which seemed designed as a brake on democracy. Labor early
adopted a Senate-abolition plank in its party platform. And for good reason: Labor was the first
party in government to be frustrated to the point of electoral defeat after suffering at the hands of
an obstructive Senate, during the brief but inglorious years of the Scullin government (1929-1931).
Not surprisingly, Labor took the initiative to deliver on the Federation promise of proportional
representation for the Senate as a safe and sensible way of giving the second chamber a
distinctive character in the Australian parliamentary system. Yet Labor was also the party,
interestingly when in opposition in the late 1960s, that moved to establish a Senate committee
system and to accept Estimates committees as the conservative parties adjunct to that reform.
Even more pointedly, Labor was the party that rescinded its Senate-abolition plank in 1979: four
years after the 1975 Dismissal, which one might have thought would have confirmed all its worst
fears about the Senate. Why the historic about-turn on the Senate? The short answer is that public
legitimacy had attached to the Senate, not so much because of its rare capacity for deadlock and
obstruction but because of its increasingly normal capacity for slow drilling of the hard boards of
open government and public accountability.
11.3 A proper explanation of this institutional history of deepening public legitimacy is beyond us
here and now. Not everything done by the Senate or by senators has been of high public value.
But if a social-democratic political party like the Australian Labor Party has made its peace with the
Senate, then we are safe to assume that the Senate has done enough to earn its place in terms of
public legitimacy. Between most elections, there is usually a crisis facing the party in government
that attracts the attention of enough non-government senators that one of the many committee
structures emerges as the primary accountability arena for public investigation of that current crisis.
The value of the Senate is unrelated to any capacity it might have to resolve this or any other crisis:
the Senate is not a court of law. Instead, the Senate’s real value is that it can find a forum for the
public investigation of misgovernment, using its power of publicity to bridge the gap between
government secrecy and open government, and to use its power to frustrate government to bridge
a related gap between government decision-making and accountable government. When push
comes to shove, the Senate’s vast constitutional powers certainly get the attention of government
focused on what senators want government to focus on; but the Senate’s sustained public
legitimacy derives more from the prudent political management of its various arenas of public
accountability. The Senate rarely ‘solves’ government problems: exercises by Senate committees
in these accountability arenas inevitably give rise to more questions than answers, but that itself is
an important contribution to the cause of open government.
11.4 There are only 76 senators, after all. They might have the backing of the Constitution. They
might also have the backing of a well-developed committee system, currently including eight
‘standing’ or portfolio committees which also take responsibility for Estimates hearings, four ‘select’
or temporary committees, two legislative scrutiny committees, participation on an expanding
number of ‘joint’ committees with membership from both houses, and a number of ‘house’
committees of which the most significant is the Senate Privileges Committee which manages the
increasing case load of friction between the Senate and government officials. Perhaps the very
smallness of the number of available senators to serve on these committees means that the
committees tend only to investigate matters of genuine public importance.
12. Conclusion
12.1 When prime minister Harper spoke to a joint meeting of Australian parliamentarians in
September 2007, he confessed that he was one of those Canadians who suffer from ‘Senate envy’
when looking at the Australian Senate.20 Prime minister Harper told his Australian audience that:
‘Australia’s Senate shows how a reformed upper house can function in our parliamentary system’. I
have reason to think that prime minister’s Harper’s praise of the Australian Senate was not shared
by the Australian prime minister John Howard, who was then enjoying his remarkable double
victory with a rare majority in both parliamentary houses. But the Howard government went on to
lose their power at the November 2007 election: they lost their Senate majority; more importantly,
they lost government; more personally, John Howard lost his own parliamentary seat.
Commentators believe that one reason for the Howard government’s defeat was, paradoxically,
their remarkable double victory at the previous election in 2004.
12.2 First elected to government in 1996, the Howard government suffered its own form of ‘Senate
envy’ associated with their inability to steamroll legislation through the Senate. Many of the Howard
government’s most prized policy initiatives, particularly workplace relations, were frustrated in the
Senate. When the double victory did arrive, the Howard government knew that such commanding
parliamentary power would not come again and so they pushed through with more daring
legislative proposals than would have been possible earlier. Opposition critics claimed that the
government had no mandate for some of the more far-reaching proposals which has never been
openly declared at election time. The community watched as the government seemed to be
overplaying its hand with a legislative program reshaped along more fundamentalist lines than
earlier programs. It is plausible that voters punished the Howard government for, among other
perceived mistakes, misusing its Senate power to ignore or at least marginalize the rights of nongovernment
parties to be treated with due parliamentary process.
12.3 But, as said earlier what goes around, come around. The new Rudd government has just
announced its own Senate reform proposals. Senator John Faulkner, cabinet secretary and special
minister of state, revealed the new vision at a conference on bicameralism just last month.21
Faulkner admitted that the current legitimacy of the Senate reflects not so much the merits of the
original Constitution as the period of Senate reform in the late 1940s that introduced proportional
representation. But the praise was grudging at best: in senator Faulkner’s view, ‘The Senate does
not reflect that fundamental, democratic, Chartist principle of one vote, one value’. Senator
Faulkner conceded that the same complaint could be made of the US Senate, and he also
conceded that the US Senate had no outstanding problem with public legitimacy.
12.4 Reporting the views of former Labor prime minister Paul Keating about senators as
‘unrepresentative swill’, senator Faulkner contrasted his open admiration for ‘democratic principles’
with his unhidden dislike of the Senate’s power to frustrate democratically-elected governments,
with the 1975 events front and centre in his picture of the Senate’s ‘constitutional restrictions’. What
are the reform options? First, curbing the Senate’s power to block supply, in order to eliminate the
1975 option. Second, substituting fixed four terms for both houses to ‘make the Senate more
reflective of the will of the electorate at the most recent election’, in place of the current mixture of
three years for the House and six years for the Senate. There is nothing more concrete on offer as
yet but this recent ministerial speech indicates where government thinking is heading. Senate
reform has been a staple theme of Australian politics and it is only fair to end by reminding readers
that the current Senate is the product of substantial reforms over many generations. All the reforms
are consistent with the original if quite permissive words of the Constitution which remain the final
authority on the power of the Australian Senate.22
1 Full text available at:
2 J Uhr. ‘Why we chose proportional representation’, chapter 2 in Marian Sawer and Sarah Eds, Representation and
Institutional Change: 50 Years of Proportional Representation in the Senate. Canberra 1999. available at:
3 J Uhr, Proportional Representation in the Senate: Recovering the Rationale’, Australian Journal of Political Science,
v30, 1995, 127-141.
4 Odgers’ Australian Senate Practice, pp564-573
5 Senate Procedural Informational Bulletin, no. 222, 27 June 2008, p2, available at:
6 Senate Procedural Information Bulletin, no. 225, 27 October, p2, available at:
7 House of Representatives Hansard, 23 September 2008, pp8274-8281, available at:
8 Senate Procedural Information Bulletin, no. 224, 26 September 2008, pp5-7, available at:
9 Full statistics are available at:
10 Stanley Bach, ‘Senate Amendments and Legislative Outcomes in Australia, 1996-2007’, Australian Journal of
Political Science, 43/3, September 2008, 395-423.
11 Bach, ‘Senate Amendments’, at 418.
12 Senator John Faulkner, Special Minister of State, Code of Conduct for Ministerial Staff, available at:
13 Odgers’ Australian Senate Practice, pp366-371
14 Senate Procedural Information Bulletins, no. 218, 25 February 2005, p4; no. 219, 25 March 2008, p2; no. 222, 27
June 2008, p3; and no. 225, 27 October 2008, p3 available at:
15 Senate Procedural Information Bulletin, no. 221, 6 June 2008, p1, available at:
16 Senate Procedural Information Bulletin, no. 221, 6 June 2008, p3, available at:
17 Full coverage in The Australian, 23 October 2008, available at:
18 Senate Procedural Information Bulletins, no 218, 25 February 2008, p3, no. 219, 25 March 2008, p2, available at:
19 Senate Procedural Information Bulletins, no 219, 25 March 2008, p3, no. 222, 27 June 2008, p3, available at:
20 House of Representatives Hansard, 11 September 2007, p4, available at:
21 Senator John Faulkner, Special Minister of State, The Senate: Blessing or Bane?, 9 October 2008, available at:
22 The full text is available at:
Additional References
Bach, Stanley 2003, Platypus and Parliament: the Australian Senate in Theory and Practice.
Canberra: Department of the Senate, extracts available at:
Castles F G and Uhr, J 2007, ‘The Australian Welfare State: Has Federalism Made a Difference?’,
Australian Journal of Politics and History, 53/1, 96-117
Uhr, John 1989, ‘The Canadian and Australian Senates: Comparing Federal Political Institutions’,
chapter 5 in , B W Hodgins et al, Eds, Federalism in Canada and Australia: Historical Perspectives
1920-1988. Peterborough: Frost Centre for Canadian Heritage and Development Studies, 130-146
Uhr, J 1995 ‘Proportional Representation in the Australian Senate: Recovering the Rationale’,
Australian Journal of Political Science, 1995 special issue on party systems, vol 30, pp127-141.
Uhr, John 1998 Deliberative Democracy in Australia: the changing place of parliament. Cambridge
University Press.
Uhr, J 1998 ‘Generating Divided Government: The Australian Senate’, ch 4 in Senates:
Bicameralism in the Contemporary World eds S C Patterson and A Mugan. Ohio University Press.
Uhr, J 2001 ‘Rules for Representation: Parliament and the Design of the Australian Electoral
System’, chapter in Geoffrey Lindell and Robert Bennett eds Parliament: The Vision in Hindsight.
The Federation Press, 249-290.
Uhr, J 2002 ‘What’s so responsible about responsible government?’, in D Burchell and A Leigh
eds, The Prince’s New Clothes: why do Australians dislike their politicians?. University of New
South Wales Press, 155-166.
Uhr, J 2002 ‘Explicating the Australian Senate’, Journal of Legislative Studies, 8/3, Autumn, 3-26
Uhr, J 2004 ‘Parliament and the Executive’, Adelaide Law Review, University of Adelaide, 25/1, 51-
Uhr, J 2005 ‘Rethinking legislative powers’ chapter in H Charlesworth, M Chiam, D Hovell and G
Williams eds, The Fluid State: international law and national legal systems. The Federation Press,
Uhr, John 2006 ‘Bicameralism’, ch 24 in R A Rhodes, S A binder and B A Rockman, Eds, The
Oxford Handbook of Political Institutions. Oxford University Press 2006, 474-494
Uhr, J 2006 ‘Appropriations and the Legislative Process’, 17 Public Law Review 2006, 173-177.
Uhr, J 2006 ‘Constitutions and Rights’, chapter 9 in B G Peters and J Pierre eds, Handbook on
Public Policy. Sage.
Uhr, J 2006 ‘The Performance of Australian Legislatures in Protecting Rights’, in T Campbell, J
Goldsworthy, and A Stone eds, Protecting Rights without a Bill of Rights.. Ashgate, 41-59.