Chapter 8: Upper houses

The concept of a review structure to watch over political decisions is certainly not new. The Romans had a Senate from the sixth century BC. Less sophisticated societies had cruder structures with a similar purpose. ‘The ancient Goths of Germany had all of them a wise custom of debating everything of importance to their state, twice; this is, once drunk-that their councils might not want vigour; and once sober-that they might not want discretion.’[73]

The executive government is not responsible to the upper house. The upper house has no role in the choosing of a government, and should have none in a dismissal. But the executive government should be answerable to the upper house for its proposed legislation and its administration. If the upper house is to be effective in these roles, it must have some control over the executive government, otherwise it will simply be ignored.

Of the twenty parliaments we are considering, only eight still have upper houses. All but two of these (the House of Lords and the Canadian Senate) are elected. Of the seven upper houses which have been abolished since 1867, only one was elective, that of tiny Prince Edward Island. The Canadian provinces of British Columbia, Alberta and Saskatchewan have never had upper houses. Ontario chose not to have one on partition from Quebec in 1867, and the Newfoundland upper house was not re-created when responsible government was restored in 1949. There is no doubt that the non-elective nature of the upper houses diminished their political prestige, and made possible their abolition by governments impatient with any check to their power.

For the perceived role of upper houses has changed since Bagehot’s day. In the eighteenth and nineteenth centuries an upper house was seen as a check on change and the onward march of democracy. ‘All second chambers have been instituted ... not from any disinterested love of mature deliberation, but because there is something their makers wished to defend against the rest of the community, especially inherited possessions and status.’[74] The Canadian Senate and most of the upper houses in the Canadian Provinces and the Australian colonies were quite openly structured so as to protect the interests of property owners. Madison campaigned for a strong American Senate as ‘check on the democracy ... it cannot therefore be made too strong.’ It was this anti-democratic role of the upper houses which led Abb Sieys to claim that ‘if the Upper house agrees with the Lower it is superfluous; if it disagrees it ought to be abolished.’

The march of democracy could not be halted, despite rearguard actions, and other justifications of upper houses came to the fore. The Bryce Report of 1918, for instance, held that a main function of second chambers was ‘the interposition of so much delay (and no more) in the passing of a bill into law as may be needed to enable the opinion of the nation to be adequately expressed upon it.’ But who is to decide when the will of the people has been adequately expressed, and what it is? And how is the second chamber to be limited to imposing no more delay than is necessary? These were-and are-the unsolved questions.

This role, too, has been overtaken by events. In all Westminster-style parliaments, the past century has seen, through the strength of party discipline and the development of party machinery, the increasing dominance of the lower house by the Cabinet. This has been reflected in a dramatic growth in departmental bureaucracy and an extraordinary expansion of administrative law and delegated legislation. The power of the courts to constrain the executive has been reduced. The lower house has almost none. The restraining role, if it is to be performed at all, must fall on the upper house. ‘The House of Lords, which once stood guard over the actions of a too powerful House of Commons, now stands guard over a too powerful Cabinet.’[75] How effective a guard is another question.

Some upper houses are moving into the political space abandoned by the lower houses; others seem moribund. Many possible roles for upper houses have been suggested. They could provide an additional pool of ministerial talent. They could perform the legislative role with regard to bills and delegated legislation, to the extent that the lower house is unable or unwilling to perform it. They could protect the special interests of states or provinces, and minority groups. They could inquire publicly into government activities, particularly those the government does not want to be scrutinised. They could monitor the activities of government business enterprises and other statutory organisations. And they could force a government which is performing badly or irresponsibly to face an immediate election.

Before the desirability and practicality of these various activities can be discussed, it is necessary to describe the different methods by which individuals become members of upper houses, for they affect both the quality of membership and the effective political power of those bodies.

Composition of upper house memberships

United Kingdom

The House of Lords is of course the prototype for all the other upper houses, but its composition has never been copied, although there were a few stray attempts in Canada and Australia. As a result of the deal struck with Lord Cranborne in 1998, the number of hereditary peers in the House of Lords was reduced from 758 to 92, two of them ceremonial posts. There were also 26 bishops and archbishops (all Church of England, of course), 27 judges (created life peers under the Appellate Jurisdiction Act of 1976 and making up the country’s highest court of appeal) and the 515 other life peers. Before the changes were made, the House of Lords had 1326 members, of whom 750 held their seats because they were born into the aristocracy. Eight more had been created. These figures do not include 63 Lords without Writs of Summons, and 52 (including four life peers) who were on leave of absence. Contrary to popular belief, most of the hereditary peerages are not ancient. Before the Stuarts, the number of English peers ranged between 23 and 55. These days only 30 per cent date from the eighteenth century or earlier. It should be noted that in the crucial vote, less than a quarter of the eligible peers actually voted.

The 90 hereditary peers who were to sit in the reformed House of Lords were elected by colleagues who were members of their party. Any of the existing hereditary peers could nominate, and each was given 75 words to state an election manifesto. These ranged from the eccentric to the pompous. One promised ‘action against cruelty to animals, particularly fishing with rods. All cats to be muzzled outside to prevent the torture of mice and small birds.’ At the other extreme, one merely said that he hoped his ‘fellow peers will honour me with their vote, on account of their knowledge of my endeavours.’ The elections were for 42 Conservative peers, 28 crossbench, three Liberal Democrats and two Labour. Fifteen Deputy Speakers and other office holders were elected separately, and there were two ceremonial posts.

The royal commission, set up by Tony Blair as the second stage in the reform of the House of Lords, reported in December 1999. It recommended that there should be some elected members of the new second chamber but rejected the idea of a wholly democratic house. The Commission thought that the new House of Lords should have about 550 members with, according to the chairman of the Commission, John Wakeham, ‘a significant minority chosen on a basis which reflects the views of the regional electorates’, the number of regional members being somewhere between 65 and 195. A third of the regional members should be chosen at the time of each general election, and their term as regional members should be three terms of Parliament-that is, not more than fifteen years.

The Commission recommended that the remainder of the 550 members should be appointed by a genuinely independent Appointments Committee, aiming to achieve a political balance matching the political opinion in the country as a whole, with about 20 per cent of the total number of members being Independent ‘cross benchers’. The appointments should include members with such qualities as ‘a breadth of experience outside the world of politics’, ‘personal distinction’ and ‘a non-polemical and courteous style.’[76] The Commission also thought that one third of the members should be women and a broad range of religious and ethnic backgrounds should be represented.

Despite the promise by Tony Blair that a reformed upper house would be in place by the next general election there was in fact no further change before the 2001 election. Whether there will then be public support for further reform remains to be seen.

The House of Lords is the final court of appeal for most legal cases, and up to eleven ‘Lords of Appeal in Ordinary’ may be created. They retire as Lords of Appeal at the age of 75, but remain members of the House of Lords for life. The appellate function of the House of Lords is now quite separate, and only ‘Law Lords’ may hear appeals. Bagehot favoured removing judicial functions from the Lords, and in fact this was done in 1873. Unfortunately the Disraeli Government restored them in the following year, so there is the confusion of having two courts of appeal in England. The royal commission did not see any reason to tackle this problem, thinking rather that there was some advantage in having senior judges in the Parliament where they could be educated in social developments. The Commission did not discuss the desirability of the separation of the legal and legislative powers.

The bishops and archbishops are at present members of the Lords until retirement, which is now the age of 70 if not earlier. The Archbishops of Canterbury and York and the Bishops of London, Durham and Winchester are there ex officio, and the remaining bishops are selected on the basis of seniority of appointment. The royal commission recommended a broadening and deepening of religious representation in the second chamber, to include Christian denominations other than the Church of England as well as representatives of other faiths.

Life peers have been regularly created since 1958, at a rate of about twenty a year. These peerages have been effectively in the prime minister’s gift, but although most prime ministers have tended to favour their own party, all have appointed life peers from the other parliamentary parties and from Independents. Indeed, opposition leaders are regularly consulted before new life peers are created. The power of the prime minister to offer someone a life peerage has been a useful way of bringing a suitably qualified individual into the ministry.

The presence of life peers has done much to raise the intellectual level and prestige of the Lords. The hereditary peers did occasionally produce individuals of ability, and, because an hereditary peer was entitled to his seat at the age of 21 also occasionally represented youth or early middle age, but such peers are unlikely to be elected under the transitional scheme. The Law Lords added a certain judicial dignity and legal knowledge, while the bishops exuded rather unfocussed social compassion. There is no doubt, though, that the Lords have been hampered by their non-representative character, and have been reluctant to take proper action on some occasions for fear of being pilloried in the media as ‘the Lords against the people’.

The House of Lords is unique in that its members are unpaid (except for daily attendance expenses),[77] and are under no constraint to attend the House. ‘Whipping’ is therefore a very imprecise science in the Lords, and party strengths can be deceptive. The royal commission recommended the introduction of a modest attendance payment.


Canadian senators, although they must reside in the provinces they represent and must have property worth four thousand dollars, do not really represent anyone except the prime ministers who appointed them. Since 1965 senators have been obliged to retire at 75. As prime ministers nearly always appoint senators from their own party, a prolonged period of rule by one party causes serious party distortions in the Senate. A prime minister with a secure majority in the Senate sometimes appoints a distinguished ‘Independent’, and party members sometimes desert the party after appointment. A prime minister also sometimes appoints an opposition MP to the Senate, in the expectation (not invariably achieved) that the government would win the ensuing by-election for the House of Commons.

Before 1984 senators did not take rigid party positions, but between 1984 and 1990 the Liberals in the Senate used their majority to frustrate the Conservatives on several occasions. In 1990 this culminated in the Senate attempting to block the Goods and Services Tax Bill. Prime Minister Mulroney used the deadlock procedure to appoint eight additional senators (all Conservatives of course), and by filling some vacancies which had been left unfilled, he managed to gain control of the Senate.

Another significant (but short-lived) development occurred in 1990. The Meech Lake Accord had provided that the provinces should propose lists of suitable persons to fill Senate vacancies as they occurred. Alberta held a province-wide election to choose a candidate and forwarded the winner’s name to the prime minister. Brian Mulroney was in a dilemma. Was a single name a ‘list’? Besides, the nominee was a member of the Reform Party, which was becoming a dangerous rival to his Conservatives in the prairies. Very reluctantly, and after a long delay, Mulroney appointed the new senator. The Meech Lake Accord has since died, so there will be no further elected senators, for the present at least.

According to a Canadian authority, the result of the system of appointing senators is that Canadian senators are old, biased in favour of a long-lived government, almost totally unrepresentative of minor parties, male-dominated (though less so than the Commons), and composed of many wealthy people who on average have more political experience than the MPs in the House of Commons. The average age of Australian senators in December 1999 was 49 years, and MPs 47. The average age of British MPs was 52 and peers 65. In Canada the figures were 51 for the House of Commons and 64 for the Senate. The Canadian Senate is becoming less male-dominated; in 2000 there were 34 women senators.


The Australian Senate was patterned on the model of the US Senate, though during the federation debates some doubted whether the concept of a powerful Senate representing the states was compatible with responsible government. Yet there was really no option: without such an upper house there was no way the four less populous colonies would have agreed to federation in the 1890s, and probably thereafter. The Constitution therefore provides for equal representation in the Senate for each of the six original states. Because the Constitution also provides for a two-to-one ratio between the two houses, as the population grew and the number of MPs needed to be increased, the number of senators had to be increased accordingly.

Initially there were six senators from each state, elected on a state-wide basis, rising later to ten senators and in 1985 to twelve. The term of a senator is six years, with half the senators retiring every three years. The Northern Territory and the Australian Capital Territory, not being states, have two senators each. Election is by proportional representation, with preferential voting. The quota to elect a senator is 14.3 per cent of the votes in a normal half-Senate election, and 7.7 per cent when the whole Senate is dissolved after a legislative deadlock between the two houses. Robson rotation[78] is not used, so the party organisations have control over the order in which their candidates are elected. Indeed, if a voter wishes to follow a party ticket, all that is needed is to tick a box, and as the alternative may be to fill in 60 or more sequential numbers on the voting form, it is not surprising that over 90 per cent of voters do tick a party box.

The result of all this is that the senators are very similar in age and background to the members in the Representatives, the choice of house being largely dictated by political opportunity, with the proviso that those who see themselves as potential leaders of their parties will try to head for the House of Representatives. Because of proportional representation and the even number of senators to be elected in each state and territory, it is most unlikely that any government will have a majority in the Senate. The balance of power will be held by minor parties and Independents. The Senate is thus in some ways much more representative of the political views of the community than is the House of Representatives, though it is somewhat distorted by the requirement (copied from the US Constitution) of equal representation from each state, so that Tasmania (population 473 000) has the same number of senators as New South Wales (population 5.8 million). Fortunately the party voting patterns across Australia are remarkably consistent, very different from those in Canada and the UK, for instance.

The Australian states

Of the five legislative councils in the Australian states, three are elected by proportional representation. In New South Wales and South Australia the electorate is state-wide; the term of a councillor is two terms of the lower house, with half the councillors facing the electorate at each general election. Western Australia continues its tradition of rural bias by having six regions, four returning five councillors each and two returning seven. The number of voters in the various regions is manipulated so as to give a two-to-one bias in favour of the rural and remote areas, with the result that the Council is significantly more conservative than the lower house, the Legislative Assembly, though after the 1996 election the balance of power was held by two minor parties. The life of the Council is a fixed four years, although the elections are held at the same time as the lower house.

Tasmania and Victoria do not use proportional representation for their upper houses. In Victoria there are two councillors for each electorate, standing at alternate elections. The Council electorates are created by lumping together four Assembly electorates.

The Tasmanian system is distinctly curious. For many years there were nineteen electorates for the upper house, with wild discrepancies developing in the number of voters-by the 1990s the electorates varied in size from 6000 to 21 000. In 1998 there was a reduction in the Council’s size from nineteen to fifteen, to be achieved over three years. The number of voters in each electorate has been equalised, with a 10 per cent tolerance, which triggers a redistribution if it is exceeded. More unusual is the electoral system. Councillors are elected for fixed six year terms, with elections being held on the fourth Saturday in May each year. When the Council size has come down to fifteen, two councillors will be elected one year and three in the next year and so on, the electorates which are voting each year being selected so that they are as far as possible spread around the state.

The extraordinary result of the by-election atmosphere created by this system of voting has been that the Council is almost entirely filled with Independents. In 1990, for instance, the Council consisted of seventeen Independents (about a third of them leaning to the Liberals and another third to Labor), one Labor councillor and one Liberal. There were of course no party caucuses, though the two councillors who were party members attended the meetings of their colleagues in the Assembly.

The Tasmanian Legislative Council can be rather reactionary. When a bill to decriminalise homosexual acts was being debated, there were claims that homosexuals should not sully Tasmania with their behaviour, and that they should go to other states where such disgusting behaviour was legal. When a councillor suggested that the opponents of decriminalisation were using Old Testament arguments and ‘in those days a person was put to death for it’, another councillor interjected ‘not a bad idea either.’[79] The decriminalisation clauses were defeated.

Representation of community opinion

Although all the upper houses were set up to protect special interests, this role has diminished. Nevertheless the composition of upper houses gives various interests formal representation they would not have in the lower house. The House of Lords still gives power to some representatives of the hereditary peers, and to some judges and senior clergy of the Church of England. Appointments by the prime minister can place distinguished individuals in the House of Lords and the Canadian Senate, although the Canadian appointments seem more designed to reward party service than to enhance the quality of the Senate. Life peers have done much to enhance the quality of the House of Lords. They certainly come from diverse backgrounds.

The largest category is of politicians, whether ex-ministers, ex-MPs or others active elsewhere in party politics, especially in local government. But the many other occupations honoured include businessmen, trade unionists, civil servants and other public servants, diplomats, senior servicemen, industrialists, scientists, economists, journalists, newspaper proprietors, doctors, lawyers, farmers, technologists, actors and artists, clerics, accountants, nurses, social workers, and a composer.[80]

If the recommendations of the royal commission are accepted, the representation would become even more diverse, with regional members as well as increased representation of women and of religions other than the Church of England.

The Australian Senate was originally designed as a states’ house. ‘Not only by its express powers,’ said Barton, Australia’s first prime minister, ‘but by the equality of its representation of the states, the Senate was intended to be able to protect the states from aggression.’ In fact, as some prescient members of the constitutional conventions had predicted, the Senate immediately split on party lines. The Senate has never been effective as a states’ house. Of course senators raise state issues, but not more frequently-in fact rather less frequently-than their colleagues in the House of Representatives. There is however one state advantage: the existence of the Senate, and its election by proportional representation, ensure that there will be someone from each state in each major parliamentary party. Tasmania is the problem. From 1972 to 1975 all five members of the House of Representatives from Tasmania were Labor, from 1975 to 1987 all were Liberal, and since 1998 all have been Labor.

The effect of proportional representation is to allow representation of minor parties and Independents, and to make it unlikely that a major party will have an absolute majority. The extent to which minor parties or Independents can gain seats depends on the size of the quota. In Australia federally, when there is a double dissolution and all twelve senators in each state are up for election, the quota is 7.7 per cent, but the drift of preferences from other parties will permit the election of someone with an even lower proportion of the primary vote. After the 1987 double dissolution, for instance, the state of the parties in the Senate was: Labor 32, Liberal-National Coalition 34, Australian Democrats seven, Nuclear Disarmament Party two, and one Independent. One of the Nuclear Disarmament Party (from New South Wales) received only 1.5 per cent of the primary vote. The person elected was later discovered not to be a citizen, and his election was declared void. His place was taken by the person who was second on the party ticket. The strength of the Nuclear Disarmament Party in New South Wales was estimated to be under 100 at the time, and it was divided into several factions.

Although the electoral systems in most of the Australian upper houses do result in the representation of minority groups and special interests, most suffer from the defect that they lag in following changes in public opinion. This applies whether their members are appointed for life or until aged 75, or whether a portion of the house (usually half, but in Tasmania one-sixth) retires at each general election. The continuing nature of elected upper houses was originally designed to maintain stability. As one delegate to the 1891 Constitutional Convention in Australia put it, it would be undesirable for ‘the people’s will in its burst and flush of impetuosity to reach the statute book. That is not what the sober-thinking, solid and-I will use the word-conservative Senate is for.’[81] A modern upper house must be aware that its membership may not reflect changing public opinion, and must be particularly careful not to obstruct important measures which the government can reasonably claim were critical to its election program.

The possibility of abolition

The possibility of abolition concentrates the minds of upper houses, though the consequence of concentration is sometimes paralysis. Yet there is no serious move against any of the remaining upper houses. The House of Lords could be abolished by an act of Parliament, which the Lords could do no more than delay for a year. It is possible, but these days very unlikely, that the Queen could insist on the issue being put to the voters before she would approve such a bill. In 1977 the Labour Party conference voted overwhelmingly for the abolition of the House of Lords ‘as quickly as possible’, but this pledge was dropped from the 1987 manifesto. The Labour Party adopted a policy in favour of a ‘second chamber’, though it was not entirely clear about how it was to be constituted, except that it should have no hereditary component. The Labour manifesto for the 1992 election suggested the replacement of the House of Lords with an elected second chamber which would have a delaying power over bills which reduced individual or constitutional rights. The Conservative manifesto did not mention the House of Lords. With the reforms being implemented by the Blair Government, abolition is not now an issue.

Everyone seems to agree that the Canadian Senate should be reformed, but there agreement stops. ‘It would be idle to deny that the Senate has not fulfilled the hopes of its founders; and it is well also to remember that the hopes of its founders were not excessively high.’[82] There is no real move to abolish the Senate, though its behaviour over the Goods and Services Tax bill made it many enemies. Abolition would require an amendment to the Constitution which would have to be passed by the House of Commons and two-thirds of the parliaments of the provinces that have between them at least 50 per cent of the total population. It would be very difficult to obtain such a vote. The Senate would not be able itself to block such a bill, for it has only a 180-day suspensive veto over amendments to the Constitution.

In Australia the Senate is powerful, and is entrenched in the Constitution. There is no way the voters in a majority of the states[83] would vote for its abolition, for the four less populous states firmly believe that the Senate protects their interests. The Labor Party used to have the abolition of the Senate as part of its platform, but in 1979 recognised reality and deleted it.

Three of the five legislative councils in the Australian states are firmly entrenched in their state constitutions. The New South Wales, South Australian and Western Australian legislative councils could be abolished only by referendum, and past experience suggests that such a referendum would surely fail. In Victoria the Council is only slightly entrenched, its abolition requiring an absolute majority in each house. However, no party is currently committed to its abolition. Tasmania, as usual, is the odd one out, for its Council could be abolished by an ordinary act of Parliament, but the prospect of the Independents who dominate the Council voting for their own destruction seems remote.

It can be seen that none of the surviving upper houses should fear abolition, anyway in the short term, or permit that fear to prevent them carrying out the jobs that should be done.

Ministers in upper houses

Although the House of Lords does provide a pool of ministerial talent, and a convenient method, through a life peerage, of bringing an eminent non-politician into the government, the expertise in the Lords is not often used in this way. It may occasionally be used as a quiet refuge for a busy minister. Prime Minister Macmillan sometimes moved a minister to the Lords to shield him from ‘the time-wasting turmoil’ of the Commons. Typically there are two or three Cabinet ministers in the Lords (out of 22) and five or six ministers of state (out of 30). As all ministers in the Commons are required to have someone in the Lords to answer for them, the balance of the Lords front bench team is made up of parliamentary under secretaries and whips. A typical government front bench team in the Lords would have 22 members, including seven whips.

Cabinet ministers in the House of Lords are more often than not life peers. In Conservative governments the other ministers and parliamentary under secretaries tend to be hereditary peers, because life peers have usually had successful careers outside politics, are not as young as they were, and are not very interested in full-time political drudgery without compensating power. In the 1990 Conservative government under John Major, for instance, both of the Cabinet ministers in the Lords were life peers, but only one of the six ministers of state and none of the remainder of the front bench. Suitable hereditary peers are much less available to a Labor government, and in the 1997 Blair Government there were none.

The elimination of hereditary peers will of course change this balance, but during the transitional period, where there are 90 hereditary peers elected by their colleagues, there may not be much change, for nearly all those hereditary peers who would have been considered for the front bench would be among the 90 chosen. The royal commission recommended that some ministers should continue to come from, and be answerable to, the new second chamber, and that senior ministers from the House of Commons should make statements to an appropriate committee of the second chamber, and answer questions from it.

In Canada there is normally only one minister in the Senate, the leader of the government party, although in the past both Liberal and Conservative governments have used Senate ministers to cover gaps in their ranks in the Commons when there is no suitable MP from a particular province.

All the Australian upper houses, except Tasmania’s, are substantially represented in the various ministries. In the 1990 Hawke Labor Government, for instance, the federal ministry contained 21 MPs (out of 76 members of the government party) and nine senators (out of 32). The figures for the second Howard Liberal-National Party Government in 1998 were 20 MPs out of 80, and nine senators out of 34. In the smaller parliaments (the Australian House of Representatives has 148 members compared to 659 in the British House of Commons and 301 in the Canadian) there is no doubt that the upper houses usefully increase the pool of potential ministers. Some of the most effective ministers in the various Australian parliaments have come from their upper houses, but there is no scope for the British technique of bringing a talented outsider into the ministry by appointing him to the upper house.

There are nevertheless substantial problems in having ministers in the upper house. A basic concept of responsible government is that the Cabinet is a committee chosen by the lower house, and is answerable to it. This has effectively prevented a non-parliamentarian remaining a minister, though this would be technically legal in the UK and Canada and some of the Australian states. The possibility of providing an MP in the lower house to represent such a minister is not thought to be a politically acceptable solution. Logically the same objection should surely apply to having ministers in the upper house.

At a practical level, if a minister is in the upper house he cannot be directly questioned by members of the lower house. Certainly he will be represented there, but the minister answering the questions may not be directly involved in the affairs of the department, or involved in only part of it. Though he will undoubtedly have a thick book of briefing notes, it is not at all the same as being able to confront the responsible minister directly. The same problem arises when the house debates the performance of departments or policies for which an upper house minister is responsible. In Australia, for instance, during the Gulf War the ministers for foreign affairs and defence were both in the Senate.

Senate ministers in Australia, like the ministers in the House of Representatives, wish to keep the sittings of the Parliament as brief as possible, so that they can get on with their administrative work in their departments, and not give the non-government parties opportunities for easy publicity. The consequence has been that the Senate sits for an absurdly short time-only half the sitting days of the House of Lords, for instance-and much important Senate business is either rushed or neglected. It is surely wrong for the sitting days of the Senate, which is the only effective scrutineer of government legislation and administration, to be determined by the body being scrutinised.

Another problem is the effect that the presence of ministers has on the numbers available in the smaller upper houses for committee work. In the Australian Senate, for instance, where there are only 76 members, the unavailability of eight or nine ministers for committee work is a significant loss.

Most serious of all is the effect that the possibility of becoming a minister might have on the zeal with which upper house members approach their other activities, all of which have the potential to question, constrain or expose the government. The task of a minister in an upper house is to try to ensure that government legislation is passed promptly, without amendments, and with as little embarrassing debate as possible. Similarly, government activities must be presented in the best light, and any inquiries potentially harmful to the government firmly headed off.

All of this is of course totally incompatible with an effective role for the upper house, but the opposition will usually support this attitude, for they hope to be in government themselves one day, and they have no desire to see really effective controls established. Worse still, upper house members who are or aspire to be ministers-and that means virtually all upper house members-will try to use the upper house to support their party’s continuing electioneering in the lower house, and such campaigns do great damage to an upper house as a serious legislature. In the Australian Senate, for instance, the policy of senators from the major parties on serious issues is determined at a joint meeting of their party, at which the lower house members outnumber the senators two to one. And if they have any ministerial ambitions, senators have to vote in accordance with these decisions.

It is difficult to see these attitudes changing while the aspiration of most upper house members is to be promoted to the ministry.

Upper houses as legislatures

United Kingdom

The royal commission did not recommend any significant changes to the legislative arrangements in the second chamber, so it may be assumed that the present performance of the House of Lords in this area will continue. These days, of an average of about 85 bills passed by the UK Parliament each year, about twenty are introduced in the Lords. This figure includes a handful of ‘consolidation’ bills, which make no substantive changes to the law, but reorganise various acts to make them more accessible. These bills are always introduced in the Lords. A few important bills are usually introduced in the Lords so as to avoid a traffic jam in the Lords at the end of the session, but these cannot include supply and money bills, which must be introduced in the Commons. The Lords sit for some seven hours a day, four (or, towards the end of the session, five) days a week, and into the evenings each sitting day except Fridays. The total sitting days average about 150 a year, which puts other upper houses to shame.

About half of the available time is now spent on legislation, a remarkable increase occurring in the last two decades of the twentieth century. Only about a fifth of the time available for legislation is spent on second reading debates, a very low figure compared with other upper houses. The Lords do not normally divide on the second reading of a government bill. They follow a doctrine, first formulated by Lord Salisbury during the term of the Attlee Labour Government, that the Lords would not oppose bills which implement undertakings given in the government’s election manifesto. Other bills-the deplorable retrospective War Crimes Bill, for instance-may be opposed, though the Lords have only a delaying role, not a veto. The impressive legal talent in the Lords effectively demolished the War Crimes Bill during the second reading debate, but the Thatcher Government was driven by political rather than legal standards, and after the necessary interval the Commons passed the bill again. When the Lords rejected it for a second time, the full provisions of the Parliament Act of 1949 were invoked for the first time, and the bill became law.

The committee stages of bills are much more thorough, and taken on the floor of the House. It is very rare for a bill to be referred to a committee; only ten have been referred in the past three decades. There is no opportunity to hear public evidence or receive submissions. In July 1992 a Select Committee on Committees recommended a special standing committee to take evidence on technical bills and then consider them clause by clause. These committees were later renamed special public bill committees, but only three bills have been considered by them, none since the 1994-95 session. None were referred to any other committee in the 1990s.

There are numerous amendments made to the bills considered on the floor of the House. Nearly all of them are initiated by the government, dealing with problems with poorly drafted bills which have nevertheless passed the House of Commons, second thoughts on technical detail or even policy by the sponsoring department, or mistakes noticed or promises given to make certain amendments. The House of Lords is a convenient place to make such amendments, though if it did not exist other ways to make them could be found, as has to be done in unicameral parliaments.

There are, though, some amendments which would probably not be made but for the House of Lords. The first type occurs when problems which have been raised in the Commons but dismissed there are raised again in the Lords, and the government introduces amendments to resolve them, either wholly or in part. The second type occurs when a peer points out a defect which no one had noticed until then. The range of specialist expertise in the House of Lords make this not uncommon. The third type is different, for it is an amendment resisted by the government but passed anyway by the Lords. Before the 1998 transitional reform of the Lords about ten bills a year were amended in this way. For the House of Lords to press the issue, it had to be concerned with one of the Lords’ special interests, which were ‘in summary: the treatment of pensioners and the disabled, rural/countryside issues, matters of constitutional etiquette, respect for existing minority rights and, to a lesser extent-because it rarely leads to defeats-consumer rights.’[84] It remains to be seen how the new House of Lords will use its power.

Some, but by no means all, of the third type of amendments were accepted wholly or in part by the Thatcher and Major governments, which was a significant change from the attitude of the Labour governments of the 1970s, when most such amendments from the Lords were rejected. The Lords did not insist on them, though they did manage to negotiate some changes to bills such as those on the dock-work scheme, devolution proposals, and aerospace and shipbuilding nationalisation, from which the Lords were able to exclude ship-repairing.

The Blair Labour Government also accepted some amendments from the House of Lords, one of which significantly changed the Crime and Disorder Bill (1998), which contained wide ranging provisions on law and order. At the report stage in the House of Commons a motion was moved to reduce the homosexual age of consent from eighteen to sixteen, to bring it into line with that for heterosexuals. The amendment was passed on a free vote by 336 votes to 129, but was rejected in the Lords. The government accepted the Lords rejection so as to get the other 25 measures in the bill into operation.

Although the House of Lords rarely refers government bills to committees, it has set up a European Communities Committee to consider draft EU legislation. The committee has six sub-committees, and they consider selected draft legislation in detail, taking evidence from witnesses including UK ministers and European Commission officials and members of the European Parliament. Although the UK Parliament has no formal role in the EU legislative process, the House of Lords reports are regarded as authoritative and often superior to anything produced by the European Parliament.

Private bills are concerned with benefits to any person or body of persons. About half of the private bills to come before Parliament are introduced in the Lords, and if they are disputed they are referred to select committees which act in a quasi-judicial manner, hearing evidence and arguments from both sides. The Parliament Acts of 1911 and 1949 do not apply to these bills, and the Lords occasionally reject them, notably a 1976 bill to nationalise Felixstowe Dock, despite the bill being supported by the Labour government. There has been a considerable increase in private legislation in recent years, because railway, light rail and other works projects require such legislation, and although the Lords could cope with the workload the Commons was under pressure. In a typical reaction of a modern ‘responsible’ government, the system was changed so that Parliament no longer has to approve these works bills.

Several private members’ bills are passed each session but such bills frequently die in the Commons, for a private member’s bill from the Lords not only requires a sponsor in the Commons but must follow the same tortuous procedures as private members’ bills introduced in the Commons. Nevertheless private members’ bills from the Lords have raised important issues, such as abortion, homosexuality and sex discrimination, which have ultimately been dealt with one way or another.

As the Lords have no power over money bills, the estimates are not considered at all by them.


Until 1984 it was a much calmer picture in Canada. The Senate had not rejected a government bill since 1940, or insisted on an amendment since 1961. The Senate did do some useful work in tidying up defects in the bills which were overlooked in the Commons, or making last minute changes to bills-possibly to meet promises made in the Commons, though in fact only 69 government bills have been amended in the past 30 years. During the 1970s and early 1980s the Senate used a system of sending the subject matter of government bills (not the bills themselves) off to legislation committees for public input while the bills were still before the Commons, and tabling the reports in the Senate where they would be available to the responsible minister. This pre-study was helpful, as worthwhile amendments could be made to a bill before it arrived in the Senate, but its almost clandestine nature did nothing for the public image of the Senate. Many senators felt that it was more appropriate for the normal procedures to be followed, and pre-studies are now very rarely used.

Things changed dramatically in 1984, when the incoming Conservative government was faced with a Liberal majority in the Senate. On past experience this would not necessarily have been a problem, but the Liberal senators had been reinforced by new senators chosen from ex-members of the House of Commons, and they introduced partisan tactics to the Senate.

In the 1984-85 session the Senate delayed a 19.3 billion dollar borrowing bill until the government tabled its spending proposals. Although the Senate was technically in the right, its action provoked anger and the Senate responded with threats to delay legislation. Helpful procedures such as committees privately suggesting amendments to ministers were dropped. Bills were still referred to legislation committees, and there was substantial input to these committees, but the committees were controlled by the Liberal opposition, and they became very partisan, with almost no cross voting. Although the Senate never finally insisted on its amendments, they did cause considerable delays and frustration.

In 1988 there was a dramatic development. The Conservative government had negotiated a free trade agreement with the United States. The enabling bill was duly passed by the Commons, but the Liberal parliamentary leader announced that the Senate would not pass the bill until an election was held on the issue. Whether the Liberal majority in the Senate would have obeyed these orders (it almost certainly would have) was not put to the test, for Prime Minister Mulroney called an immediate election, which he won. The bill was then passed by the Senate.

This was nothing compared to what happened two years later. The Mulroney Government had passed a bill through the Commons to introduce a new indirect tax, a ‘goods and services’ tax. When the bill reached the Senate it was referred to a legislation committee. The committee, controlled of course by the Liberals, toured the country for two months stirring up opposition to the bill, which the committee then recommended should be rejected. Extraordinary scenes followed: the Senate sitting 24 hours a day for weeks on end while the Liberal opposition filibustered, with whistles being blown and old speeches and complete books being read; the use of the deadlock power to bring in eight new Conservative senators; sustained uproar for hours to prevent the Leader of the Government being heard so that he could move a motion to bring the matter to a vote; members of the Commons being invited, most improperly, on to the floor of the Senate to harangue senators; press photographers on the floor of the Senate; and fist fights almost breaking out between senators. The bill was eventually passed, but the whole process was scandalous and a dramatic change from the dignified and rather somnolent Senate of the past. What will happen in the future, if another government is faced by a Senate controlled by an aggressive opposition, is a critical question. The problem has been deferred, for the Liberal government elected in 1993 managed to regain control of the Senate two years later by filling four vacancies with party members.

Before we leave the question of the Senate’s handling of bills, two unusual procedures should be mentioned. Although the Senate can establish legislation committees, it has never yet done so. If a bill is to be considered by a committee-and this is rare-the bill is referred to one of the standing committees. Only one standing committee (the National Finance Committee) deals with estimates. Because of the absence of ministers in the Senate, the responsible minister from the Commons may give evidence to the committee handling the bill, and may even be invited to come into the Senate to handle a bill at the committee stage. Other upper houses could well consider such arrangements.

They might also consider the Canadian Senate’s arrangements for the pre-study of bills, which have already been described. The purpose of the procedure was to allow some Senate input into bills, which are frequently sent up to the Senate at the end of a session, and expected to be adopted quickly and without change. It falls far short of making the Senate an effective legislature, but it was better than nothing, until it fell into disuse.

The Canadian Senate does not really consider supply bills at all, thought it certainly has the power to do so if it chooses. Private bills are usually introduced in the Senate in order to balance the workload between the two houses. Private bills are considered carefully by a committee, which hears both petitioners and those opposed to the bill. There are many fewer such bills than there used to be, for most were divorce petitions from Quebec and Newfoundland, and these were stopped in 1963. There are few private members’ bills introduced in the Senate, because such bills have little chance of becoming law and there is no publicity value, for the Senate is little reported and there are no constituents.


The government has had a party majority in the Australian Senate for only five of the last 30 years (1976-81). With the House of Representatives having abandoned any pretence to being a legislature, it might have been thought that the Senate would move rapidly to fill the gap. Nevertheless the Senate was surprisingly slow to move into the field of careful committee examination of government bills.

In 1970 seven (later increased to eight) ‘legislative and general purpose’ standing committees were set up, and one might have thought that they would play a major role in reviewing government bills, whether originated in the Senate or received from the Representatives. In fact very few bills were referred to these committees-usually only one or two a year, none in some years-although it was generally agreed that the bills which had been considered by the committees had been considerably improved as a result.

Of course the government could be expected to resist such referrals, claiming delay, though the urgency was rarely apparent. The real reason seemed to be that the government felt that any detailed examination of its bills, with public evidence being taken, by an all party committee was an unreasonable intrusion by the legislature into the business of the government! Yet the government did not normally control the Senate, so how did it prevent a bill being sent to a committee? The secret was to do a deal with the Democrats or the Independent senators holding the balance of power, agreeing to allow them to make amendments which would be attractive to their supporters, on the implied condition that opposition amendments and attempts to refer the bill to a standing committee would be resisted. It was a pernicious system. A deal could not always be made, and bills were frequently amended by the combined votes of the opposition and the Democrats, or very occasionally referred to a standing committee for thorough examination.

Even attempts to refer bills to the standing committees for consideration in the lengthy winter recess never succeeded. The chairs of the legislative and general purpose standing committees were always government party senators, and they always maintained that their committees were already overloaded with inquiries and they could not possibly deal with any legislation.

After prolonged campaigning by those who wished to reform the Senate as a legislature, in 1989 a select committee was set to examine, among other things, the question of the referral of bills to standing committees. The committee recommended the setting up of a Selection of Bills Committee, to consider all bills except money and appropriation bills, which at this time were dealt with by the estimates committees. The Selection of Bills Committee was required to report on each of the other bills as to whether it should be referred to a standing committee, and if so which committee and by when it should report to the Senate. The select committee’s recommendations were unanimously adopted by the Senate, and the immediate result was a sharp increase in the number of bills considered by the standing committees. More bills were referred to committees in the first twelve months of this procedure that had been in the previous twenty years. Now about 30 per cent of bills presented to the Senate are examined by one or other of the committees.

Another problem that had to be tackled was how to provide adequate time for the Senate to consider a bill thoroughly. Governments of both persuasions tended to pass a mass of bills to the Senate at the end of a session (until 1994 there were two sessions a year) with the demand that they should be rushed through. The Senate, which was not controlled by the government, imposed increasingly stringent guidelines for the handling of government legislation, and the government eventually agreed that it would make ‘a concerted effort to ensure that in future most legislation is introduced in one sitting for debate in the next’, and that from 1994 the number of parliamentary sessions would be increased from two to three. The Senate strengthened this by passing a standing order which required that a bill passed in one period of sittings may be considered in that period of sittings only if it is introduced in the first two-thirds of that period, and any bills introduced after that are automatically adjourned until the next period of sittings. The government has to make a special plea for a bill to be exempted from this rule.

There was a further change in 1994. There is always a danger when a committee is examining legislation and also conducting inquiries into the activities of a government department that the committee will become overloaded and one or the other will have to give. The solution adopted was to set up two committees-one dealing with legislation, the other with references-in place of each of the eight legislative and general purpose committees. The chairs of all the legislation committees were government party senators, giving the government an effective majority on each of the committees.

The big increase in the number of bills referred to committees is all very well, but the quality of the examinations of controversial bills has fallen very sharply. While only a few bills were being considered by Senate committees, the party leaderships did not attempt to force committee members to toe the party lines. The bills that were referred were thoroughly examined, and the reports were usually unanimous, even on highly controversial bills such as the one on freedom of information.

With the greatly increased number of bills being considered since 1990, it is no longer acceptable to the party leaderships to have them examined in such an independent way. Effective pressure has been brought to bear on the senators concerned. Committee members now follow the party lines, and committee hearings on controversial bills tend to become electioneering stunts rather than serious examinations of the contents of bills.

There has also been a disturbing development in the behaviour of the chairs of legislation committees. All such chairs are held by the government party, and the senators concerned seem to be moving towards a belief that government bills should not be touched. Even the referral of the very important GST bills to legislation committees was resisted by the government, apparently on the grounds that the very complex bills had been carefully considered by the government, so why was the legislature trying to get involved in the detail of such legislation?

The attitude of the legislation committee chairs is that, if possible, bills should not be referred to a committee at all, the chairs using often dubious claims of work overload. If they are referred, the hearings should be brief and if possible no amendments should be made. The reason for this developing attitude is that a committee chair is seen as a stepping stone to the ministry, and the prestige of a senator is thought to be greatly increased by ruthless and successful protection of government bills. The reaction of the non-government parties was to use their majority to refer some controversial bills, not to the legislation committees but to the references committees, where the non-government parties had a majority.

These problems do not arise if a bill is not a matter of dispute between the parties. About 90 per cent of bills fall into this category. There is no doubt that the legislation committees do valuable work with these bills, and many of the non-controversial bills they consider are improved as a result of their investigations. In the committee hearings expertise in the community is brought in, and perhaps even more importantly the opportunity for members of the community to have some chance to make an input into laws which may have a significant effect on them is very important for the working of the democratic system.

The question, though, is whether the Senate committees can afford to spend time on these non-controversial bills. There are only 76 senators, and there are unlikely to be more than about 33 or 34 government party senators. By the time ministers and the president have been provided from the government party, there are usually not more than 23 or so government party senators available for committees. With the number of joint committees with the House of Representatives and other important Senate committees stretching the time of the available senators, it is important that the Senate committees should focus on investigations the House of Representatives cannot or will not undertake.

For this reason it was earlier recommended, when the performance of the House of Representatives as a legislature was considered, that consideration of non-controversial bills should be left to the committees of the House of Representatives, thus leaving Senate committees with time for more controversial investigations. At the moment the committees of the House of Representatives do nothing about such bills; some effective pressure will have to be applied to them.

Private members’ bills in the Senate used to be infrequent, but the development of the minor parties has made a sharp change, the number being introduced rising from about two a year in the 1970s and 1980s to twelve in the 1990s. This is because there is some prospect of the bills being passed, as a result of the lack of government control of the Senate, and senators have begun to make use of the publicity opportunities. Since 1970, 162 private members’ bills have been introduced in the Senate, and twenty have been passed by that chamber. In order to be taken up by the House of Representatives, a senator’s bill must be sponsored by an MP, and seven of the twenty bills passed by the Senate were ultimately accepted by the Representatives. It is worth noting that two of the seven successful Senate bills were moved by opposition senators, and two others by Australian Democrats.

The estimates are reasonably thoroughly dealt with. They are considered by the eight legislation committees, who are allocated departments to consider by a Senate resolution. They consider the main estimates (after the introduction of the budget in the House of Representatives in May) and the additional estimates (usually in November). The evidence is taken in public session, but ministers from the House of Representatives do not appear before these committees, with the result that most of the departments being examined are represented by a Senate minister who has little detailed knowledge of what is going on in the department. Public servants from the department being examined do attend, and may answer the questions of committee members if the minister approves, which he usually does, particularly if he is not the responsible minister.

These committees have been effective in forcing the government to produce estimates in a standard format with proper explanatory notes. They have also uncovered some waste and mismanagement, and certainly public servants view the hearings with trepidation. On the other hand, the committees have serious shortcomings. They are not permanent watchdogs, for they meet for only a few days a year. Worst of all, the non-government senators are anxious to embarrass the government by uncovering dramatic mistakes by public servants, rather than examining the policy and methodology to which the expenditure is related. Still, that is the nature of politicians.

The increased influence of the minor parties and Independents in the 1990s has resulted in a great deal of the negotiations on the budget being conducted directly between them and the government, and they have been able to negotiate specific changes to the budget in return for their support for the budget as a whole. This is a great change from the days when the budget was regarded as non-negotiable and its progress through both houses was regarded as automatic.

The consideration of annual reports by government agencies has also been greatly improved. Responsibility for considering these is assigned among the legislation committees, and the committee has to report to the Senate whether the report is apparently satisfactory. If it is not the committee has to conduct a detailed examination. There is no doubt that, although the system is far from perfect, the Senate has made a dramatic improvement in the supervision given to the very numerous government agencies, which a decade ago were scarcely scrutinised at all.

The Australian Senate has taken also three dramatic actions which have substantially improved its performance as a legislature. In 1981 it set up a Scrutiny of Bills Committee, with independent legal advice, with instructions to examine each bill when it is introduced and to report if any bills trespass unduly on personal rights and liberties; make rights, liberties or obligations unduly dependent upon insufficiently defined administrative powers; inappropriately delegate legislative powers; or insufficiently subject the exercise of legislative power to parliamentary scrutiny. The all party committee raises queries about an average of 85 bills a year, nearly 40 per cent of those introduced into the Parliament. In something like half of the cases the responsible minister provides a satisfactory explanation, but that leaves nearly 50 bills a year where there are concerns which the minister, in the opinion of the committee, cannot meet.

The committee does not yet have the political clout of the comparable Senate committee watching over delegated legislation, but its views are available to both MPs and senators, and about 30 per cent of its criticisms result in amendments to bills. Without this committee a good deal more defective legislation would be passed. It will come as no surprise that the establishment of the committee was strongly opposed by the government of the day, the Fraser Coalition Government. The government’s attitude was that the bills had been carefully drafted by the government and its advisers, so why should the Parliament want to get involved? A waste of time, the government said. The committee was established only after several Liberal senators cross voted.

The Senate has also taken action to limit two other abuses of power by governments. A custom had grown up by which the treasurer would announce a change of tax policy in a press release, and this change would come into immediate effect, even though the bill to enact the changes retrospectively might not be introduced for many months, sometimes more than a year. Worse still, the bill might differ significantly from the press release which until then taxpayers were being expected to obey. In 1988 the Senate passed a resolution which sharply restricted the amount of retrospectivity it would tolerate, and since then government behaviour has noticeably improved in this regard.

The other abuse of government power which has now been controlled concerns the failure to proclaim acts or parts of acts which have been duly passed by the parliament. Certainly delay is sometimes justified, to allow for such matters as the preparation of regulations, but the situation was often absurd, with acts which had been rushed through the Parliament as ‘urgent’ remaining unproclaimed for months or even years. Following Senate protests, all acts which are to commence on proclamation now have a provision that they come into effect after six months if the proclamation is not made. The Scrutiny of Bills Committee is very assiduous in drawing attention to any bills which have open-ended proclamations. The government provides the Senate with an annual list[85] which shows the details of legislation which is to come into effect on proclamation and has not yet been proclaimed. The reasons for the failure to proclaim the act are set out. As an example, in the 1999 report there were 24 such acts. The oldest was passed in 1974, and dealt with financial corporations. The government said that Part IV of the Act was inconsistent with new arrangements for the financial sector, and the need for it was being reviewed.

If the government, for some good reason, does not wish to proclaim an act, or part of it, it now has to come back to the Parliament, explain its reasons, and ask it to amend or repeal the act. That is as it should be.

State upper houses are generally less effective than the Senate as legislatures. None of the state upper houses regularly sends bills to committees for public examination. Western Australia has the only upper house with a standing legislation committee, intended to deal with contentious bills as well as revisions of acts. The committee is empowered to take evidence from the community, but only about 5 per cent of bills are referred to it. Tasmania, as usual, has an unusual arrangement. Although the Legislative Council does not formally take evidence from the public, there is an arrangement whereby the whole Council meets with pressure groups, hears their cases and asks questions. Although evidence from the public is desirable, this procedure is far too loose. There is no public call for submissions on a bill, the evidence is not given on oath, and the proceedings are not recorded.

When upper houses are not controlled by the government, bills are fairly frequently amended. The Tasmanian Legislative Council, which is never controlled by any government, amends 40 to 50 per cent of bills, many of them heavily, and meetings of managers from the two houses are often used to try to resolve differences.

Ministers from the lower houses are entitled to attend the upper houses in Victoria and New South Wales, to explain their bills and to answer questions, but not of course to vote or to move motions. The right has occasionally been exercised in New South Wales, but is not used in Victoria.

To show the contempt state governments sometimes show towards upper houses, it is worth looking at the recent behaviour of the New South Wales government towards its upper house. The Legislative Council is elected by state-wide proportional representation, and is rarely controlled by any government. The Legislative Council, against the wishes of the government, added clauses to a government bill. Rather than go through the complicated procedure for the resolution of a deadlock between the two houses, the government accepted the amendments in the lower house, and then proclaimed the Act excluding the amendments made by the Legislative Council. And there was nothing the Council could do about it.

Deadlocks between the two houses over legislation

If upper houses are to be effective legislatures, there will inevitably be deadlocks over legislation between the two houses, or, lower houses being what they are, more accurately between the government and the upper house. A method of resolving such deadlocks is desirable, but not easily devised. In the UK the problem has been solved by removing any power over money bills from the Lords and limiting them to a mere delaying power over most other bills. In Canada the government can nominate four or eight additional senators (one or two from each region) to resolve a deadlock. This has been used only once, in 1990.

In Australia federally there is the so-called double dissolution procedure, by which, if there is a deadlock over a bill, both the Representatives and the whole of the Senate may face an election if the government so desires. If the deadlock persists after the election, it can be resolved at a joint sitting of the two houses. A legislative deadlock is held to exist if the Senate rejects or unacceptably amends a bill passed by the House of Representatives, and after an interval of not less than three months the same thing happens again. The double dissolution method of resolving such deadlocks was adopted by the 1897 Constitutional Convention after other suggestions had been rejected. The idea of a referendum was decisively defeated, and almost in desperation the Convention accepted the concept of a double dissolution.

It was not very well thought out. Although the delegates to the 1897 Convention were quite clear that the Senate had to have the power to reject (but not amend) the budget, they did not consider what happened next. Perhaps they thought that a compromise budget would be produced by negotiations between the two houses. Whatever their thoughts, the deadlock procedure does not cope with the blocking of supply, because a government cannot afford the waiting period of three months, at any rate with the traditional budgetary timetable.

Nor has the procedure worked at all well for other deadlocks, for in an election campaign the dissolution issue is soon forgotten. It would be a bold person who held that, after such a campaign, the voters’ decision on who should form the government was also a decision on the particular piece of legislation which caused the election. Moreover, in some double dissolution elections, several deadlocked bills have been put to the community. What does a voter do if he or she agrees with some of the deadlocked bills and disagrees with others? The threat of a double dissolution is not even an effective way of bringing a recalcitrant Senate to heel, for governments hold elections when it is politically advantageous to do so, and would very rarely be prepared to risk loss of government over a particular piece of legislation. Besides, with proportional representation in the Senate, a double dissolution is an advantage to minor parties, and therefore unattractive to the major parties.

Victoria and South Australia are the only states with similar deadlock procedures. In each of these states the first three years of the four year term are ‘fixed’, and an election cannot be held unless a no-confidence motion has been passed in the lower house, or there is a legislative deadlock between the two houses. If the Victorian premier wants to resolve a deadlock issue (or use it as an excuse for an early election), an election can be held for the whole Assembly and half the Council. In South Australia an election must have been held between the first and second rejections by the Council, and the premier has the option of sending the whole Council to the polls, as well as the Assembly. But it is almost inconceivable that any re-elected government would want to have another election so soon, so it is not surprising that the procedure has never been tried. South Australia has an alternative method of resolving deadlocks, the election of two additional Council members, but this too is not very attractive and has never been tried.

The American model is not helpful either. In Washington the president can veto legislation passed by the Congress, though this veto can be overridden by a two-thirds majority of each house. This system is designed to allow the executive to prevent the legislature passing laws not desired by the executive. In modern responsible governments the problem is the opposite. There the problem is how to prevent an active Legislature not controlled by the executive from unreasonably frustrating its legislative program.

A system of resolving legislative deadlocks is clearly desirable, and general elections have not been a satisfactory solution. More use could undoubtedly be made of conferences between delegates from each to try to resolve differences over amendments. There have been only three such conferences in the Australian Parliament since 1901 and all produced solutions. It is certainly a better system than the government attempting to do secret deals with minor parties and Independents in the Senate. In Canada eight such conferences were held between 1925 and 1999. Five resulted in settlements, one in a stalemate, and two bills were abandoned when the government refused to accept the recommendations of the conferences.

But although conferences might be helpful, they certainly will not resolve all the legislative deadlocks, particularly with the tightening of party discipline which has occurred in most parliaments. There are two other possible ways of resolving deadlocks-by a joint sitting of the two houses, or by a referendum of voters. The problem with a joint sitting is that different methods of election are usually employed for the two houses. The most common is single member constituencies for the lower house and proportional representation for the upper. The narrow margins which normally result from proportional representation, and the decisive majorities which are the usual outcome of single member constituencies, would mean that the government would nearly always have a secure majority at a joint sitting, which would then become merely another party controlled rubber stamp for the government’s legislation.

A referendum of voters would be a better way to resolve a deadlock, for in a referendum on a particular piece of legislation the interest of voters would be focussed on the bill rather than electing a government. The only Parliament to have adopted the referendum as a means of resolving a legislative deadlock is New South Wales, but the procedure is so cumbersome that it is almost unusable. It takes at least nine months for the pre-conditions for holding such a referendum to be met. If a bill passed by the Legislative Assembly has not been dealt with by the Legislative Council after two months, or has been rejected or unacceptably amended, the Assembly may, after waiting three more months, pass the bill again, and the process is repeated. But that is not the end. There has to be a conference between managers (usually ten of them) from each house, but no votes are taken. If the managers fail to reach agreement on the bill there may be a joint sitting of the two houses, though again no votes are taken. If there is still no resolution, the Assembly may direct that a referendum be held, but such a referendum may not be held for at least two more months.

It is not surprising that such a complex procedure has been used only once since its adoption in 1933. That was in 1960, on a bill to abolish the Legislative Council. At the subsequent referendum the bill was soundly rejected. There are too many delays and safeguards in the New South Wales procedure. Governments will not hold referendums recklessly, for defeats are usually politically very embarrassing. A government will consider a referendum only if the bill in question is important, there appears to be strong community support for it, and there is no prospect of the upper house passing the bill in an acceptable form. Certainly the upper house should be given time to consider the bill properly, but surely a period of four months from the time a bill has first left the lower house should be ample. Governments deserve to have the power to implement key elements of their program in a timely fashion. The safeguard is that if what they want is out of tune with the community, the referendum will fail. New South Wales has the right solution, but the wrong method of applying it.

Monitoring of government administration

In monitoring the activities of the government, its departments of state, its business enterprises and other statutory and non-statutory government bodies, upper houses have the same weapons as lower houses: questioning ministers, either orally or in writing; setting up committees of inquiry; and forcing debates on particular topics by moving motions. One might have thought that, as lower houses have become more disciplined to the will of the government, the upper houses would have moved to fill the void, but this has by no means always happened.

The House of Lords uses select committees to meet particular needs, but with no coherent pattern. There are two committees, one (with six sub-committees) dealing with the scrutiny of European Union legislation and the other (with two sub-committees) with science and technology. They have a high reputation for objective, in-depth analysis among their somewhat specialist interest groups. There are occasional ad hoc select committees-about one or two every three years-dealing with general policy matters such as murder, life imprisonment, overseas trade, unemployment and also some domestic committees dealing with such matters as the broadcasting of the proceedings of the House of Lords. There is no systematic scrutiny of the activities of government departments and quasi-government organisations.

The Royal Commission on Reforming the House of Lords, which reported in January 2000, appeared to accept the present arrangements, but recommended the establishment of three new committees. Two of these were a Constitutional Committee to scrutinise the constitutional implications of all legislation, and a Human Rights Committee to examine all bills and delegated legislation for human rights flaws. There is no doubt that these committees would be valuable, after the experience of the Scrutiny of Bills Committee in the Australian Senate. The third proposed committee was potentially the most important, for it was to scrutinise treaties laid before Parliament and to draw attention to matters which should be considered by the Parliament before the treaties were ratified by the government. The royal commission also recommended the strengthening of the power and the support provided for two existing committees, one scrutinising delegated legislation and the other scrutinising ministers’ handling of European Union business. Finally, the Commission thought that there should be more consideration of the drafts of bills before they were introduced into the chamber.

The Canadian Senate uses standing committees fairly freely to investigate particular problems, and some useful reports are made. The committee reports are usually fully debated in the Senate, but government responses to the recommendations are infrequent. Critical examination by Senate committees of the activities of government departments and business enterprises (Crown corporations) is negligible. Perhaps the most useful recent Senate committee inquiry concerned a proposal to establish a security intelligence service. The bill was very controversial and heavily criticised. It was not debated by the House of Commons, but instead was referred to a special Senate committee, which proposed substantial amendments. These were accepted by the government, and a satisfactory new bill was drafted.

The Australian Senate has eight ‘references’ committees, covering the full range of government departments.[86] These committees have conducted some valuable inquiries, by no means always on subjects welcomed by the government-which are of course often the ones that should be investigated. The committees can inquire only into subjects assigned to them by the Senate, but in practice the Senate rarely gives a committee a task it does not want or refuses a committee a reference it does want. The committee reports are debated in the Senate and generally receive good media coverage. The government is expected to reply to a committee report within three months, and it usually does, giving its observations and intentions with regard to the committee’s recommendations. As part of the deal when the functions of the old legislative and general purpose committees were divided, all the references committees have a non-government chair. This gives the non-government parties control of the committees, and there has been some blatant electioneering by some of the committees, which has reduced their prestige and the value of their reports. The Senate also occasionally sets up a select committee to investigate a particular problem.

Senate supervision of government business enterprises and other statutory commercial and marketing bodies is negligible. Some questions are asked by estimates committees, though these are often inappropriate because there is no proposed appropriation for the enterprise concerned. In any case, the party political nature of estimates committee hearings makes any serious examination of their operations very difficult. The old Senate Finance and Government Operations Committee did some useful work in establishing how many such bodies there actually are. It uncovered a surprising number no one seemed aware of. The committee also did some excellent work in ensuring that annual reports were made and were presented to Parliament. There is an opportunity for senators to speak briefly when an annual report is tabled,[87] but the real responsibility rests with the Senate references committees, for each annual report is referred to the appropriate committee for it to make any investigations it thinks appropriate, and since 1989 the relevant committee has been required to examine each annual report referred to it. A committee could, if it wished, summon members of the management before it, and cross-examine them on their operations and the quality and timeliness of their reporting to Parliament. In fact, no committee has yet systematically exercised its power, partly because of the pressure of other work and particularly because of the tradition of very short parliamentary sittings. Meanwhile the organisations concerned go on, virtually unsupervised by the Parliament which represents their owners.

The Australian state upper houses generally do not have a very developed system of standing committees, though they do participate in joint committees with the lower houses on matters such as public accounts and delegated legislation. Western Australia’s system is the most extensive with five standing committees. The Western Australian committees cover legislation; estimates and financial operations; ecologically sustainable development; constitutional affairs; and public administration. Bearing in mind that there would typically be more than 600 state government agencies, it can be seen that the supervision is very limited, even in the one state which has a standing committee on the subject.

New South Wales has also made some useful advances in recent years. In 1988 two standing committees were created, on Social Issues and State Development, with a third standing committee, on Law and Justice, established in 1995. There are also three estimates committees to scrutinise the annual budget.

Question times

Seven of the eight upper houses have daily question times while the house is sitting, though in all of them its effectiveness is limited because most of the ministers are not available for questioning. Certainly a frontbencher represents each lower house minister, but this is not at all the same thing as directly questioning the responsible minister. In the House of Lords questions are not directed at a particular minister but at the government, and a frontbencher replies on behalf of the government. None of the upper houses has attempted to arrange for lower house ministers to visit the upper house, on a roster basis, to answer questions. This would no doubt be difficult to arrange, but in its absence question time in the upper houses remains of limited effectiveness.

In the House of Lords, as in the Commons, all the questions are printed on the order paper,[88] and the Lords deal with four questions each sitting day. Supplementary questions are freely allowed, and the period is fairly lively. The average time taken by the question period has risen from ten minutes in 1970 to 30 minutes in 1999.

The Canadian Senate has traditionally had a club-like atmosphere, with senators deciding among themselves who is to ask a question or a supplementary question. The Speaker has had little control unless appealed to, though he has been given a little more authority after the traumatic events of 1990. Question time is certainly not very satisfactory. Usually the only minister in the Senate is the Leader of the Government, and he has no departmental responsibilities. Committee chairs are sometimes asked questions. It is all a sort of ritual game, with opposition senators trying to induce the minister to depart from his brief and to make an admission which will embarrass the government, while the minister tries to stick to his brief while at the same time not looking ignorant or foolish. Question time used to be open ended, but it has been reduced to 30 minutes.

In the Australian Senate, as in the Representatives, all questions are in theory without notice but, unlike the Representatives, one supplementary question (never more) is allowed to the senator who asked the question (never to anyone else). An hour is allowed for question time each sitting day, and in that time an average of sixteen questions and four supplementaries is asked. To avoid lengthy policy statements by ministers in response to a ‘Dorothy Dix’ question, (named after an American ‘agony aunt’ who ran a column in an American newspaper answering questions most of which she had written herself) and lengthy speeches masquerading as questions from other senators, the Senate in 1992 imposed a one minute limit on questions and a four minute limit on ministerial answers, and on supplementary questions a one minute limit on both question and answer.

The questions alternate between the two sides of the chamber, and a minister in trouble with an opposition question will hope for a friendly one from his own side. Question time is heavily biased in favour of the ministers, but the opposition has an opportunity for 30 minutes after question time to debate the answers given by ministers. A senator may move to take note of an answer given that day by a minister, and may speak for up to five minutes on it, the call alternating between the two sides of the chamber.

In the state legislative councils question time is substantially the same as in the Senate, though except in New South Wales the period is shorter. Supplementary questions are not permitted in the Victorian Legislative Council, whereas as many as three have been permitted in Western Australia. The Tasmanian Legislative Council has no question time because it has no ministers.

Questions asking for written answers

In all the upper houses, there are arrangements for ministers to provide written answers to questions from upper house members. In the House of Lords, the number of such questions has risen from about 300 a year in 1971 to 4322 in 1998-99. Some of these questions are ‘planted’, so that a minister may make a written statement without going through the formality of making it orally in Parliament. All the questions are normally answered within a fortnight.

In Canada written questions are little used by senators, only about 40 a year being asked. The government certainly does not hurry in answering them, but there is usually a reply within three months. In the Australian Senate an average of about 1000 questions a year are placed ‘on notice’ to be answered in writing, and most are answered within 30 days, although some remain unanswered for many months. If a question is not answered within 30 days, and the minister has not given an acceptable reason for the delay, the matter may be raised at the end of question time, usually by moving a motion for the answer to be tabled by a specific date. Ministers usually comply with such orders, though nothing effective can be done to them if they do not.

All the legislative councils in the Australian states use such questions lavishly, with New South Wales councillors asking more than 900 a year. In Tasmania 150 are asked each year, for this is the only way a councillor can question a minister. In all the states, the delays in responding to upper house questions are the same as for lower house members.

Orders for ministers to produce government documents

If the executive government is to be accountable to the parliament, it must be prepared to produce relevant documents on matters of public concern. Some government documents of course should be exempt, such as Cabinet minutes or ones affecting national security, but most should be made available to the parliament. Governments of all persuasions resist such disclosure when the documents would be politically embarrassing, and mere requests for the documents are not likely to be effective.

An extreme example occurred in the New South Wales upper house in 1996. When an opposition councillor asked the treasurer, Michael Egan, for some documents about allegedly improper government handling of the Fox Film Studios agreement, Egan refused to produce them. When the Council passed a motion requiring Egan to produce the documents, Egan refused again, and the Council suspended him from the remainder of the day’s sitting. Egan disputed the validity of the order for his expulsion and refused to leave. He had to be escorted out by the Usher of the Black Rod, acting on the orders of the President. Egan took the Council to court (at considerable expense to the taxpayers) seeking a declaration that an unlawful trespass on his person had taken place when he was expelled, but the real issue was an attempt to prove that the Council did not have the legal power to require the production of documents, and that the government, having being elected to govern, had the right to determine which documents should be made public. Egan failed, both the New South Wales Court of Appeal and the High Court finding that the Legislative Council had the power to act as it did, and the Court of Appeal finding that claims of public interest immunity and legal professional privilege do not protect the government from the use of the Legislative Council’s power. The documents were eventually tabled on 26 November 1998, more than two years after they had been asked for.

Over the years the Australian Senate’s power to order the production-and even the creation-of documents has been used increasingly, and in recent years has averaged about fourteen a year. The reaction of the government has been predictable; in 1999 the government started freely to use ‘public interest’ as a reason for not producing the documents, although the grounds were nearly always very dubious. There were five refusals of requests for the production of documents during the year. Two were of particular interest. The first of these was when the Leader of the Government in the Senate, Senator Hill, was formally censured by the Senate for not producing documents which had been requested. The documents concerned the Jabiluka uranium mine. He did table some of the documents, but withheld others, and later said that only ‘key documents’ had been produced.

The Minister for Family and Community Services, Senator Newman, was involved in the second of these dramatic refusals to produce documents. She refused to release a draft document on changes to the welfare system, despite having earlier said she would release the draft at a Press Club speech. Among the many grounds she gave for refusing to release the draft to the Senate were that its disclosure would ‘confuse the public debate’ and ‘prejudice policy consideration’, whatever that may mean. The minister was censured by the Senate, and the government was put under pressure by the Senate majority who increased the length of question time and ordered a committee hearing on the minister’s behaviour. The draft document was eventually produced, but it certainly would not have been if the government had had a majority in the Senate.

The power to order the production of government documents is much less used in the Canadian Senate. One of the few significant examples occurred in 1995, when a special select committee recommended ‘that an humble address should be presented to His Excellency the Governor General praying that he will cause to be laid before the Senate a copy of the submissions to the Treasury Board in August 1993 relating to the Pearson Airport Agreement.’ The recommendation died on the order paper without being brought to a vote.

Select committees in the House of Lords can request the publication of government documents. The government usually produces the documents requested, but in fact requests are very rare.


In all the upper houses there are procedures by which debates can be initiated, to give information, to gain publicity, to draw attention to problems, or to attempt to force government responses. In the House of Lords ministerial statements are infrequent, because so few ministers in the Lords are responsible for policy. However, important statements made in the Commons are repeated in the Lords by junior ministers, and questions on them are allowed for up to 30 minutes. Wednesdays are set aside for debates rather than legislation. These debates may be initiated by the opposition parties, or by the cross-benchers, or by government backbenchers, but once a month there are two two hour debates on backbenchers’ motions chosen by ballot. There are other debates on government motions, such as to ‘take note’ of a green or white paper, and there are debates on select committee reports, usually moved by the chairman. These various debates occupy about a quarter of the sitting time. Peers can also initiate debates as the last business of the day (similar to adjournment debates in other parliaments) by giving notice of a so-called ‘unstarred’ question, and the debate on the question has no time limit, which makes it unpopular with the management. About 50 such debates occur each year, typically taking a total of 60 hours.

Important ministerial statements made in the Commons are sometimes debated in the Canadian Senate, but no ministerial statements are made there. Very little use is made, either by the opposition or by individual senators, of opportunities to initiate debates on other subjects, because the purpose of such debates is publicity, and such debates in the Senate would not attract it. The fact that senators have no constituents to appeal to is another reason for the lack of interest in such debates.

In the Australian Senate, as one might expect in a house not normally controlled by the government, there is generous time allowed for the opposition to debate matters of its choice, at least while the Senate is sitting, which is an average of only 73 days a year, as compared with the House of Lords average (1994-95 to 1998-99) of 148 days a year and the Canadian Senate average of 82 days. The Australian state upper houses meet even less frequently than the Senate, and for shorter hours when they do meet. Each sitting day in the Senate there is a period, usually two hours, when ‘matters of public importance’ or ‘urgency motions’ may be moved, and there is a further period on Thursday when motions or private members’ bills may be moved. Both these periods are used almost exclusively by the non-government parties, in proportion to their party strengths, but the subjects are chosen by the party leaderships. There are two other periods when backbenchers can air their personal interests and campaigns-the lunch hour on Thursday and the adjournment debate each day. The latter lasts as long as senators desire to speak, which is sometimes quite a long time. In all, the party controlled debates last roughly 200 hours a year, and the backbenchers’ debates 40 hours. Considering the relatively small number of sitting days a year, the non-government parties have very fair opportunities to put their point of view.

New South Wales, Victoria and Tasmania set aside one day each sitting week as a private members’ day though, except in the Tasmanian Legislative Council where there are no parties, it is usually pre-empted by the party leadership. New South Wales also permits ‘matters of public importance’ to be raised, but this is not much used, despite unlimited time being available for the debates. In the Western Australian Legislative Council there is no government business as such, and ministerial motions have to take their turn with the motions of other councillors. Backbenchers may be able to raise their own issues on private members’ day (if they can persuade their party leadership), and there is another opportunity on the adjournment debate each day.

In some upper houses the members use the device of giving notices of motions, knowing that there is no chance of them being debated, but hoping to gain favourable publicity for themselves, or their party, or their cause. The number of such motions ranges from 475 a year in the Australian Senate to an average of eight in the two legislative councils which use them. Such motions are little used in the House of Lords or the Canadian Senate, because the Lords and the Canadian senators have no constituents. The subjects of the motions may vary from serious matters such as international relations or abortion, to trivial publicity such as the victory of a football team, though some upper houses do not permit such motions. The aim is always publicity.

An upper house forcing an election

The last, and most controversial, of the possible activities of an upper house is to force a government which retains the confidence of the lower house to a premature and unwanted election. Justification for such action might be found if the government were acting illegally, or governing so incompetently that it had lost the confidence of the voters, or was proposing a major change of policy which had not been foreshadowed to the voters.

In the nineteenth century the head of state might have taken action on such matters, and as late as 1914 it was thought that King George V might order an election over the government’s proposals for Irish Home Rule, but it seems highly improbable that any head of state would now intervene in such matters, though questions might be asked and warnings given. The only exception might be if the government were acting illegally and refused to refrain. In all other cases, if any action is to be taken it will have to be taken by an upper house, if it has the power.

An upper house, disenchanted with the government and seeking to destroy it by forcing it to a premature election, has two weapons at its disposal: refusal of supply, or the rejection of so many government bills as to frustrate the government totally. The House of Lords has lost both these powers. The Canadian Senate has the power to reject appropriation bills, but it would be pointless for it to use this power, for the government would survive without difficulty by using special appropriations, which do not need parliamentary approval.

The only upper houses with effective control over supply are the Australian Senate and the legislative councils of Victoria, Tasmania, South Australia and Western Australia. The New South Wales Legislative Council has not had power over a bill for ‘the ordinary annual services of the government’ since 1933. Since federation in 1901 the supply-blocking power has been used six times: twice federally (1974 and 1975), twice in Victoria (1947 and 1952), once in South Australia (1912) and once in Tasmania (1948). On each occasion the outcome was an election, and in all but two of the elections the government which was refused supply (or threatened with refusal) lost the election. The exceptions were the 1974 federal election in Australia, in which the incumbent Whitlam Labor Government scrambled home in the House of Representatives but did not gain control of the Senate, a failure which would be fatal for it eighteen months later; and the 1948 election in Tasmania, where the Labor government was returned. There were dark mutterings in the Labor ranks, but no action was taken against the Legislative Council.

What is to be said for the upper house taking over the electoral college role of the lower house? Certainly governments sometimes govern badly, and lose the support of the voters, though oppositions are not always the best judge of this, as evidenced by the 1974 fiasco in Australia. Certainly governments sometimes make abrupt changes of policy, towards objectives not considered in the preceding election campaign. The Canadian Senate forced such an election in 1988, by threatening not to pass the bill on free trade with the United States unless an election on the issue were first held. But if an election is forced on such a change of policy, it cannot be guaranteed that the new policy will be a major issue in the campaign. Experience suggests that it will often be submerged in other issues. Besides, it is important for a government to be prepared to change its policy, radically if necessary, to meet changed circumstances. There was a story of Australian Prime Minister Gough Whitlam keeping a chart in his office of Labor’s election promises, and gleefully crossing them off as they were met, despite the fact that radically changed economic circumstances (the first oil crisis) had made many of them very inappropriate. If the threat of a snap election dissuaded the government from taking sensible administrative action, the nation would be the loser.

This is the nub of the problem. If a government, despite possessing the confidence of the lower house, is under constant threat of being forced to an election, it will behave very much like a public company faced with a hostile takeover. That is, it is likely to drop any long term investment plans and concentrate on immediate benefits to the shareholders. In a government, such behaviour would be very damaging.

The likely behaviour of a government deprived of supply must also be considered. After the events in Australia in 1975, it seems certain that no government would meekly accept being forced to the polls by the upper house, unless the government was satisfied that the opposition had made a misjudgement and that the government would win the election, as happened in 1974. A government is more likely to attempt to outface the opposition, refusing to recommend a dissolution to the Governor-General or Governor. Despite the fact that historically the failure of a government to secure supply from the parliament justifies the dismissal of the prime minister or premier, it is most unlikely that any Australian head of state would act as Sir John Kerr did in 1975. The memories are too bitter. Besides, a government might not accept dismissal in such circumstances, and what would the head of state do then? Call in the army, or the police? What would almost certainly happen if both sides remained intransigent is that supply would run out and essential government services would fall into chaos. Of course this could not continue indefinitely; there would have to be an election eventually, but the government would try to make sure that the opposition was blamed for the chaos-and would probably succeed, which is why any future blocking of supply would be an act of political insanity.

If its use would be so damaging, what should be done about the power of upper houses to block supply? Neither the British nor the Canadian model is desirable. Both the British and Canadian upper houses have lost power as legislatures because of their lack of control over supply, though their non-elective nature restricts their legislative role in any case. Victoria and South Australia have made attempts to deal with the problem. The legislative assemblies in those states have four year terms, with the first three years being ‘fixed’ and the final year having the usual arrangement whereby an election can be held at the whim of the premier. If the full four year term were made fixed, as it is in New South Wales, it would eliminate the possibility of the Legislative Council blocking supply in order to force an early election, for there could not be one.

Before it adopted the fixed four year term, New South Wales had had in place for many years an alternative arrangement for preventing its upper house from forcing a premature election by blocking supply. The state Constitution has a unique provision, by which the Legislative Council has the power to reject or amend any bill, except money bills dealing with the ‘ordinary annual services of the government’. This is a sensible provision for those who think an upper house should not have the power to usurp the electoral college role of the lower house by making it impossible for the government to continue, but who also think an upper house should have the power to amend or reject other supply bills.

Unfortunately the New South Wales scheme is not well worked out. The Legislative Council is confronted with a single bill covering not only the ‘ordinary annual services of the government’ but all other annual government expenditure. How does the Legislative Council amend such a bill to remove something it finds objectionable-a new capital-works project, for instance-without risking blocking supply? Unless the government was prepared to accept the Council’s amendment, or to split the bill, the very crisis which the provision is meant to overcome would strike. Besides, the expression ‘ordinary annual services’ is not defined. A definition was worked out between the federal government and the Senate in 1965, when it was agreed that certain government expenditures were not for ‘ordinary annual services’. Such exclusions covered the construction of public works and buildings; items of plant and equipment which are clearly definable as capital expenditure; certain grants to the states; and new policies not authorised by special legislation. To the list was later added the expenditure for Parliament, which is certainly not ordinary annual expenditure of the government.

There is nevertheless no guarantee that the New South Wales courts would use this definition. It seems desirable that the New South Wales Constitution should be amended to include a definition of the ordinary annual services of the government (preferably based on the federal model, which works), and also to require that the expenditure for the government’s ordinary annual services be submitted to Parliament in a separate appropriation bill. If New South Wales made these changes, it would have a model which other bicameral parliaments could well copy.

The frustration of a government by the upper house rejecting or unacceptably amending key government bills is a much less certain method of forcing an election, and in doing so an upper house destroys its credentials as a responsible legislature, and damages the community it represents in the process. In any case, the election initiative is in the hands of the government, and it certainly will not call an early election which it is likely to lose.


There are constant proposals for the reform of upper houses, particularly the non-elected House of Lords and the Canadian Senate, but all the reform proposals have started from the faulty premise that the role of an upper house has not changed since Bagehot’s day, a ‘revising and leisured legislature’ which is ‘extremely useful, if not quite necessary’. In fact, as most lower houses have effectively abandoned their legislative and critical public inquiry roles, and responsible government has become party government, the upper houses must take over the abandoned roles, for otherwise there will be an elective dictatorship.

The Australian Senate is the most effective of the eight upper houses being considered. It sends about a third of the bills it receives to legislation committees for input from the bureaucracy and the public. It has established a Scrutiny of Bills Committee, with independent legal advice, to examine the legal details of bills, and this committee has uncovered an astonishing number of flaws. It has established an efficient system for control of delegated legislation, a subject in which the House of Representatives takes no interest. The Senate has taken action to restrict the government’s use of ‘legislation by press release’ and also to force the government to proclaim within a reasonable time the bills duly passed by the Parliament-or to come back to the Parliament for permission not to. estimates committees examine all the appropriation bills, and force public servants to answer questions about their details. In the investigative field, the Senate committees have conducted many useful inquiries, and have shown themselves willing to venture into fields the government would have preferred to keep out of sight.

This is all very well, but there are many deficiencies. First, the Senate is the only upper house in the past 50 years which has forced a premature election by blocking, or threatening to block, supply. Nothing has been done to restrain this power.

Second, committee consideration of some controversial bills has often developed into electioneering slanging matches, a far cry from the days when the committees considering controversial bills nearly always produced unanimous reports. The Senate also often tamely acquiesced in absurdly short time limits imposed by the government, though the situation was greatly improved in 1993 when a motion was moved by a minor party to require bills to be automatically adjourned if insufficient time was available for scrutiny, unless the government could provide good reason for urgency. The motion was adopted by the Senate, although Prime Minister Keating described it as a ‘constitutional impertinence’. The responsible ministers, if they are in the House of Representatives, do not appear before the committees considering the bills. If the bills are controversial on matters on which government and opposition have taken opposing stances, the attitude of committee members is dictated by the decisions made at party meetings, at which senators are outnumbered two-to-one by representatives.

Third, the chairs of all the legislation committees are government party senators, and as the proportion of bills being reviewed by committees increases they are coming to see their role (and their prospects of promotion to the ministry) as getting government bills through with no delay and no amendments.

Fourth, the estimates committees are only temporary, and are poorly staffed. The Senate has not been prepared to stand up to the government to insist on adequate resources being made available.

Fifth, the Senate has taken no action to persuade the government that ratification of treaties and some aspects of the use of the defence power should be subject to approval by Parliament.

Sixth, the scrutiny of non-departmental government activities-business enterprises and so on-is derisory.

Finally, and this is the most serious of all, the Senate meets far too briefly, typically for only half as many days as the House of Lords, and much necessary work is rushed or neglected as a consequence. It is absurd that the Senate’s sitting pattern is largely determined by the government it is supposed to be watching. Ministers tend to regard parliamentary sittings as irritating distractions from their other work. Provided they can get their legislation passed, they feel that the less the Senate sits the better, and the Senate tends to oblige.

What can the Australian Senate learn from the other upper houses, and what will it have to solve for itself?

There is nothing to be said for the Senate usurping the electoral college role of the House of Representatives by blocking supply in order to force an election. If the House of Representatives had a fixed term this danger would be largely removed, though it is just possible that the Senate might try to force an election by demanding that the government party pass a vote of no confidence in itself. If that is thought to be a real risk, or a fixed term is not implemented, then the provision in the New South Wales Constitution denying the upper house any power to reject appropriations for the ordinary annual services of the government should be adopted.

In order to make the best use of the relatively small number of senators available for committee work, the Senate should concentrate on controversial issues. Committee examination of bills should concentrate on controversial bills-about 10 per cent of the whole-leaving the House of Representatives to handle the careful consideration of non-controversial bills.

It would also be important that the responsible minister should give evidence to legislative committees. If ministers from the House of Representatives showed reluctance to give such evidence, and be subject to cross-examination, the answer of the Senate would be simple. The bill will not be proceeded with until the responsible minister has given evidence. It would also be worth considering the Canadian Senate arrangement by which the responsible minister from the lower house is allowed to handle a bill in the Senate at the committee stage.

The budget should be accepted as a package, but to ensure that this understanding is not abused, the Senate should insist that any new programs of expenditure or taxation should be debated and agreed by the Senate before being incorporated in the budget.

The Senate general references committees should be encouraged to tighten their examination of non-departmental government agencies. A greatly increased annual number of sitting days-up to 150, say, to make the Senate comparable in this respect with the House of Lords-would facilitate this.

The Senate should look to the Canadian legislation which gives the Parliament control over the government’s use of its defence power, without preventing an immediate response by the government in an emergency. The Senate should press for a similar act in Australia. None of the four national parliaments has control over the ratification of treaties negotiated by the government, so the Senate will have to mount its own campaign there.

If the Senate functions as an effective legislature, deadlocks between the two houses over legislation would be inevitable. Many could be resolved by conferences between delegates from each house, but some will be intractable. Special general elections have proved a hopeless method of resolving such deadlocks, and joint sittings of the two houses would be little better. The only Parliament with a potentially satisfactory solution is that of the New South Wales, where a deadlock may be resolved by a referendum of voters. Unfortunately the New South Wales referendum procedure is so cumbersome as to be virtually unusable, but it would not be difficult to produce a workable scheme. A referendum would certainly be better than any of the alternative ways of resolving an intractable legislative deadlock.

The acceptance of the Australian Senate as effectively the sole legislature would require the modification of its method of election. Proportional representation results in a Senate which certainly represents the balance of political views in the community, but there are two defects in the electoral system. The first is caused by the equal representation of the six states, regardless of population. Fortunately voting patterns are remarkably consistent across the country, if one treats the Liberals and Nationals as a single party. This consistency of voting is quite unlike that in Canada and the UK, where major parties may have no representation in significant areas, and local parties proliferate. It would be almost impossible to alter the provision in the Australian Constitution for equal representation of the states, for the less populous states strongly (though wrongly) believe that the Senate protects their interests.

The second problem with Senate representation is caused by the election of half the senators every three years. This continuing nature of the Senate may be appropriate for a conservative house of review, but is not suitable for a legislature, which should be as reflective of current community opinion as is the government whose proposed laws it is reviewing. The answer is to make the terms of both houses the same (as has been done by some of the state parliaments), hopefully a fixed four year term for both.

The status and financial rewards of the chairs of major Senate committees should be raised so that they approximate those of ministers. To avoid abuse of these rewards, the number of such major committees should be limited to half the number of ministers there are in the House of Representatives. If the chairs were fairly divided among the parties in the Senate one might expect more unbiased chairs, for they would owe their positions not to who was in government, but to their personal standing in the Senate. These rewards to prominent senators would compensate them for the loss of the possibility of ministerial office, which is an essential change.

For the presence of ministers in the Senate is the greatest obstacle to reform. Upper houses have been regarded in most parliaments as providing a useful pool of ministerial talent, but there are other and better ways of providing such a pool if it is needed, which indeed it often is because of the shortage of talent in the lower house. While the aspiration of most senators remains to become ministers, there will be little pressure for reforms which will make the government more accountable to the Parliament, and the Senate more effective as a legislature. Those in power will resist such moves, while those out of power will not wish to see any new constraints on their power when their turn comes.

Previous Page | Contents | Next Page

Back to top