![]() ![]() ![]() |
|||
|
|
Parliament |
Term |
Date of change to 4 years |
Fixed term component |
Reserve power to dissolve the parliament? |
|---|---|---|---|---|
|
Commonwealth |
3 years |
|
Nil |
Yes |
|
NSW |
4 years |
1981 |
4 years |
Yes |
|
Victoria |
4 years |
1984 |
3 years |
Yes-restrictions apply during fixed term period |
|
Queensland |
3 years |
|
Nil |
Yes |
|
Western Australia |
4 years |
1987 |
Nil |
Yes |
|
South Australia |
4 years |
1985 |
3 years |
Yes-restrictions apply during fixed term period |
|
Tasmania |
4 years |
1972 |
Nil |
Yes |
|
ACT |
3 years |
|
3 years |
Yes-but only if deemed incapable of effectively performing its functions or is acting in a grossly improper manner |
|
Northern Territory |
4 years |
Always 4 years |
Nil |
Yes |
Source: Adapted from Queensland Parliament, Legal, Constitutional and Administrative Review Committee, 'Four Year Parliamentary Terms', Background paper, April 2000, p. 7.
What is the 'best' term for a national lower house?
A bipartisan issue of some longevity in Australia has been the suggestion that the House of Representatives maximum term of three years ought to be increased by one year, and recent calls for such a change suggest that the question is worth reopening. Before one can assess this realistically, it is relevant to ask whether there is some optimum length for the term of a national lower house. What is the situation in other nations? Is Australia very unusual in having a three-year term?
According to a study published by the Inter-Parliamentary Union in 1993, the overwhelming number of national lower houses have terms of four or five years. Relatively few have a three-year term, and the United States House of Representatives is the only lower house with a two-year term. Table 2 gives the figures, and where possible includes examples of liberal democracies in each category.
Table 2: National lower house terms
|
Length of Term |
Number |
Comments |
|---|---|---|
|
2 years |
1 |
USA |
|
3 years |
13 |
Including Australia, New Zealand, Sweden |
|
4 years |
55 |
Including Austria, Belgium, Denmark, Germany, Japan |
|
5 years |
76 |
Including Canada, France, Ireland, Italy, United Kingdom |
|
6 years |
3 |
Morocco, Nicaragua, Sri Lanka |
Source: Inter-Parliamentary Union, Electoral Systems. A World-Wide Comparative Study, Geneva, 1993, see also Blackwell Encyclopaedia of Political Science, Blackwell, Oxford, 1991, p. 332.
What is the best length for a parliamentary term? Despite figures showing that four or five year terms are most preferred, the answer is far from clear cut, and a surprising lack of research data does not help in the search for an answer.(1) The best guidance that the Inter-Parliamentary Union can give poses the problem to be solved:
In theory, elections to a Parliament should not be so infrequent that they fail to reflect the opinions of the electorate, nor be so frequent that they are likely to produce excessive discontinuities in the process of government.(2)
As shown in this paper, the question is how Australia can balance a preference for stable government with the countervailing need to maintain democratic practices.
At the Constitutional Conventions
Three-year terms were generally favoured throughout the process of constitution-writing in the 1890s. They were provided for in Tasmanian delegate A. I. Clark's draft constitution circulated to members of the 1891 Convention, in the draft Bill that emerged from that Convention, as well as in the final 1897-98 Convention draft. This clear preference presumably reflected the fact that five of the colonies had three-year terms; only Western Australia had four-year terms. The initial draft at the 1897-98 Convention had provided for the longer term, but this was reduced to the shorter term while the draft was in committee. By comparison with other issues, however, this was not a matter which stirred much debate among the constitution founders.
The Commonwealth Constitution (section 28) therefore stated that:
Every House of Representatives shall continue for three years from the first meeting of the House, and no longer, but may be sooner dissolved by the Governor-General.
Despite the near-unanimity of the Constitution writers, over the years there have been many calls to increase the House of Representatives term, and although some have called for five-year terms (see below), most have expressed a preference for terms of a maximum of four years.
In 1983 the Commonwealth Parliament passed the necessary legislation for five constitutional referenda that were to be held on 25 February 1984. Although there were differences of opinion among Senators and Members concerning the different proposals, most were in favour of the second of the five-that to extend the maximum life of the House of Representatives from three to four years. The Leader of the Opposition, Andrew Peacock (LP), referred to the existence in the community of 'wide support' for this change.(16) Perhaps unfortunately, the Hawke Government's response to the Senate's difference of opinion with it over the issue of funding for the YES case, was to defer the five referenda indefinitely.(17) Two of the proposals (Simultaneous Elections and Interchange of Powers) eventually were put to referendum at the same time as the 1984 Commonwealth election, but the others, including the four-year term proposal, were never put to the people.
Five years later, in September 1988, voters were presented with a constitutional amendment which proposed replacing the current term with a maximum term of four years. Frustratingly for advocates of the longer term, however, the Hawke Government confused the issue by including in the proposed change a reduction of Senate terms to four years as well as a provision for simultaneous elections, the latter of which had been defeated on three previous occasions. Voters could not pick which of the three aspects to support or reject, for they were required to vote YES or NO for the entire package. Voter opposition to changes to the Senate term, and to simultaneous elections meant that fewer than one-third of the electorate was prepared to support the change. The ploy of linking a proposal that had quite a deal of bipartisan support to matters that were extremely contentious had failed, pushing the four-year term model aside for an indefinite period.(18)
Length of parliaments since 1901
Because the current House of Representatives term is for maximum terms only, and because of the convention that Prime Ministers can call elections virtually whenever they choose, the 38 completed parliaments have had terms of greatly varying length. It is not always appreciated that House of Representatives elections can actually be held further apart than three years. The three-year clock begins at the first post-election meeting of the House, which need not be held for up to 30 days from the day appointed for the return of the electoral writs. At the expiry of a House, the electoral writs need not be issued for up to 10 days, and their return is required within a 100 day period. All of this means that there is no constitutional barrier to an election being held more than three years after the previous election. There have, in fact, been nine such occasions, most recently in 1972 (Table 3).
Table 3: Elapsed time between House of Representatives elections (years)
|
Elapsed time |
Date of election |
|---|---|
|
Over 3 years |
1910, 1913, 1922, 1928, 1937, 1946, 1949, 1954, 1972 |
|
2 years 9 months-3 years |
1906, 1925, 1940, 1943, 1958, 1961, 1966, 1969, 1980, 1993, 1996 |
|
2 years 6 months-2 years 9 months |
1903, 1917, 1919, 1934, 1987*, 1990, 1998 |
|
2 years 3 months-2 years 6 months |
1983* |
|
2 years-2 years 3 months |
1931 |
|
1 year 6 months-2 years |
1955, 1963, 1975*, 1977, 1984 |
|
1 year-1 year 6 months |
1914*, 1951*, 1974* |
|
Under 1 year |
1929 |
Source: Australian Electoral Commission, Electoral Pocket Book, Canberra, 1999, pp. 66-71.
* Double dissolution elections
The 1910 election was held three years and 122 days after the 1906 election, while the break between the 1946 and 1949 elections was three years and 73 days. At the other end of the scale, eight elections have been held less than two years after the previous election, though four of these (1914, 1951, 1974, 1975) were double dissolution elections which cannot be called within six months of the expiry of the House, and are therefore far more likely to be announced early in a parliament than are 'normal' elections. The remaining four parliaments (ending in 1929, 1955, 1963, 1984) averaged 18 months, with the shortest period being the 10 months and 25 days between the 1928 and 1929 elections. Since 1901 the average elapsed time between elections has been 30.3 months, though if the six double dissolution elections are not counted, this figure climbs to 32.1 months.
If we look at specific periods over the years, we note that there has been a marked reduction in term length during the past 25 years. The average for all elections during this time is only 27.5 months, though the holding of four double dissolution elections no doubt distorts the figures. Even with these four elections removed from the figures, the average parliament (ignoring the current parliament) has lasted barely 30 months (see Table 4).
Table
4: Elapsed time between House of Representatives elections,
by period (months)
|
Dates |
All elections |
Double dissolution elections excluded |
|---|---|---|
|
1901-98 |
30.3 |
32.1 |
|
1901-49 |
32.1 |
33.1 |
|
1951-98 |
28.6 |
30.9 |
|
1901-28 |
32.8 |
34.8 |
|
1929-49 |
32.1 |
33.1 |
|
1951-72 |
29.9 |
31.6 |
|
1974-98 |
27.5 |
30.0 |
Source: Australian Electoral Commission, Electoral Pocket Book, Canberra, 1999, pp. 66-71.
These figures suggest that parliamentary terms have been shortening over the long haul. There is, however, no reason why this should be so. Since the 1972 election Australia had a period (1974-87) during which House terms averaged only 24.3 months-25.7 without the four double dissolution elections-yet the average length for the nineties has been 33.7 months. As noted later in the paper, a key factor in the whole issue of House of Representatives terms is the political reality that Prime Ministers will always be seeking to call elections at times of maximum benefit to their party or coalition. The fact that more than one quarter of all elections have been held after an interval greater than three years, suggests that if successive Prime Ministers were of a mind to do so, all parliamentary terms could be of three years or more.
Despite this, even if all parliaments were to run full term, the question still remains: is a 'three-year' term too short for a modern national lower house?
Among those concerned about the operation of government in Australia, by far the preferred option is to introduce four-year maximum terms for the House of Representatives.
The case for four-year terms(19)
Over the years a number of the points once made in favour of four-year maximum terms have fallen into disuse, as can be seen in the report of the 1927-29 Royal Commission into the Constitution. At that time the Commissioners believed that the three-year period was inadequate for Australia in view of:
Only the deleterious effect of a short period between elections remains an important argument.
Modern critics of the status quo tend to focus on at least seven specific benefits they claim will flow from an extension of House of Representatives terms to four years.
A long-standing claim holds that longer terms would encourage governments to introduce policies that were long-term rather than merely politically expedient. There is a widespread view that increasing the term for the House would enable governments to enjoy the luxury of being able to take 'more responsible, long-term views' than is possible when the next election is quite likely to be less than two years away. The commonly-heard view of the typical three-year term is that governments:
... tend to spend their first year settling in; begin taking tough and far-sighted decisions in the second year; and then effectively shut up shop in the third year because it is getting too close to the next election.(21)
this aspect of the governmental system can cause frustration within the bureaucracy; as the Queensland Constitutional Review Commission recently put it:
It has been said that under a three-year term, the first budget is devoted to paying off the promises made at the previous election and the third budget in anticipating the promises to be made at the forthcoming election. Consequently, only one budget out of three, the second, it likely to address important, long term policies without the contamination of short-term political considerations.(22)
Quite clearly, there is 'little time to engage simply in good government'.(23)
This point is said to be strengthened by the distortion caused by the uncertainties of the electoral 'game'. Too often, it is claimed, all the important players in the game find themselves distracted from the day-to-day slog of government. The electorate is too often in election mode:
There is ... an unholy preoccupation among politicians, Press and public with political manoeuvring at the cost of attention to policy issues. The real business of government suffers.(24)
Associated with this, it is argued that governments would have longer to weather the community response to the implementation of policies seen as necessary for the community good. It is said that this would be especially valuable in the area of economic management.
It is claimed that longer terms would enhance business confidence. The private sector has long complained that national elections disrupt their long-term planning, with deleterious effects upon the national economy. It has been noted, for instance, that retail sales drop in the period before a Commonwealth election. Calls for longer terms in New Zealand have also tended to come from business organisations.(25)
The Business Council of Australia has been a keen advocate of a longer term. Writing in the late 1980s, the President of the Council, Sir Roderick Carnegie, stated:
The uncertainties created by frequent elections and consequent shifts in Government policy in turn have an adverse effect on business confidence and business investment. Very few other democratic nations suffer this disadvantage as most have maximum parliamentary terms that are significantly longer than that of Australia.(26)
Ron Brunton of the Institute of Public Affairs is less certain of the importance of this argument. He suggests that business support for such a change might depend less on concerns about efficiency than 'the desire for a longer period of return on all the time and resources they spend in cultivating the party in power'.(27)
Elections cost money. The Australian Electoral Commission has reported that the 1998 election cost almost $95 million. Over time, a great deal would be saved by having fewer national elections. Since 1985 there have been four elections in South Australia for the four-year term House of Assembly; by comparison, there have been five Commonwealth elections for the three-year term House of Representatives. The longer the period between elections, the greater the saving for the taxpayers forced to foot the election bill.(28)
It is often said that Australians dislike the frequency with which they are required to vote, something that is believed to be linked to a distaste for the tough nature of our party and electoral politics. Fewer Commonwealth elections would reduce this to some extent. Former Professor of Politics at La Trobe University, Joan Rydon, has stated that any reduction in the number of elections should be seen as part of a process of 'reducing the adversary nature of our party politics'.(29)
Bringing the House of Representatives into line
A change to four-year terms would bring the House of Representatives term into line with most State and Territory lower house terms. In both Queensland and the ACT there has been recent discussion about the possibility of changing from three to four year terms. If this occurs, the House of Representatives will be the only Australian lower house retaining the shorter term.
Former MHR, Jim Snow (ALP), has claimed that at the electorate level the current system does little for the representative function that is so important a part of the MP's duties. He believes that three-year terms may make local members adept at campaigning, but they do not encourage them to work in a sustained way on long-term problems: 'Members are tempted to become show ponies rather than watch dogs'.(30)
A final, speculative assertion, is the possibility that longer periods between elections would raise the standard of political debate. It has been wondered if such a period might 'create more opportunity for genuinely bipartisan discussion of a wider range of issues [than is normally the case]'.(31)
In summary
Such are the benefits that it is claimed would flow from the introduction of four-year maximum terms for the House of Representatives. It can be noted that four of these are essentially unprovable-the reduction of election costs would be certain, and the House term would be another reflection of a general move to four-year terms for Australian lower houses.
It is also important to note that recent suggestions have referred to four-year maximum terms for the House of Representatives, and that making this change is not to suggest that each term would always last for four years. If the first century of our Parliament is any guide, most terms would be less than four years. If Prime Ministers followed previous practice, however, then the House usually would not be dissolved earlier than six months prior to its term expiring. The nett impact therefore would be that the usual length of a House term would have been extended by a year-and any term that did run for the full term would, obviously, be a bonus. In order to guarantee a full four-year term, fixed terms would be necessary (for fixed terms, see below).
Despite there being much support for four-year terms, it is by no means certain that a referendum to ratify such an amendment would pass. Although a referendum for four-year terms passed comfortably in New South Wales in 1981, a similar referendum was defeated in Queensland in 1991. It goes without saying that it is difficult to amend the Australian Constitution. It seems that if an amendment on any topic is to be accepted by the voters, it needs to appear as non-controversial as is possible to be. There certainly needs to be no division between the major parties.(32)
In regard to any proposal to increase House of Representatives terms, two obvious objections would be that:
On the first, there would seem to be little that could be done to lessen any possible impact this might have in a society that is said to be endemically suspicious of politicians.
In regard to the second objection it would probably be prudent to have the amendment worded so that the change could not be waved away simply as a change being made solely to benefit the government of the day. Perhaps the first extended term should not begin for the House of Representatives for which the election was being held, but for the next?
If an argument in favour of lengthening the House of Representatives term is that this would give government and business longer to plan and introduce policies, should the maximum term be increased from three to five years? The lower houses in Ireland, France, Canada and the United Kingdom all have terms of this length. Table 1 shows that more nations have five-year parliaments than any other term.
Some Australians have argued for a five-year term for the national upper house. In the 1920s, for instance, William Higgs MHR (ALP, Nationalist), chair of the Royal Commission on the Finances of Western Australian, As Affected by Federation (1925), noted that the average length of a parliament was two and one half years, and spoke of the difficulties this caused for members. The time spent on electioneering made the performance of parliamentary duties, particularly in such a large country, extremely difficult. Higgs called for an increase in House terms to five years and Senate terms to ten years, asserting that:
Trade, commerce, and industry would profit by the change, and members of Parliament would be able to give more time to the study of Commonwealth problems.(33)
Five-year terms have not been unknown in Australia. In the nineteenth century five of the colonies had five-year terms at some stage. Legislation passed in Tasmania in 1936 introduced such a term for the House of Assembly, and this remained in place until 1969. In 1937 South Australia also introduced five-year terms, though it reverted to the three-year term two years later, following public criticism.
Elaine Thompson of the University of New South Wales believes a move from three-to five-year terms would be too great a change for the electorate to accept, primarily because of concerns about a perceived reduction in the democratic elements of the political system. At the same time, she acknowledges that five-year terms might be expected to give greater stability and improved government 'efficiency'.(34) Although Thompson claims that examples of the UK and Canada suggest that in these systems with five-year terms there is no guarantee of such stability, it is noteworthy that there have been 21 elections in Australia in the past 50 years, compared with only 14 in the UK and 16 in Canada.
Five-year House terms would open up the question of what to do with Senate terms. Should they be ten years with half retiring every five years? Five-year terms in parallel with House terms? Equal to two House terms? Any of these would probably be seen as too long, creating a Senate that was said to be 'out of touch' with the electorate. If the present fixed Senate term of six years was retained, this would result in the holding of many more elections, something that would fly in the face of the aim of reducing the number of elections (for more on Senate elections, see below).
On balance, it is unlikely that many MPs would support an increase from three to five-year terms. Even in the UK there has been a significant amount of sentiment expressed that the House of Commons term of a maximum five years is too long.(35)
Should the three-year maximum term be retained?(36)
In a dissenting note to the 1929 Royal Commission on the Constitution, three members refused to accept the need for the increase in the House of Representatives term that some Australians called for. They claimed that the three-year term 'is quite long enough', and noted that 'it is in the control of members' to alter their fixation with preparing themselves for the next Commonwealth election. Their concern was with democracy:
The greater the control of Parliament by the electors the better for the people, and the lengthening of the term of Parliament tends to weaken this control.(37)
Former Prime Minister Keating (ALP) has also referred to the question of democracy, claiming that the Australian democratic system is 'very robust' because the voters have a chance to change the Parliament, and hence the Government, 'every three years or less'.(38) In the words of a 1981 Parliamentary Library paper, the arguments for longer terms 'water down our democratic standards'.(39) The historian, Geoffrey Blainey, has also seen this matter in terms of democracy, believing it would be a 'harsh penalty' to deprive the Australian people of the right after three years to dismiss an incompetent or lacklustre government.(40) Similarly, Senator Brian Harradine (Ind) has said that 'it is important that the people are given the chance regularly to audit what the government is doing by voting it in or out of office'. Harradine's solution would be to create a climate in which parliaments are allowed to run their full three-year term.(41) Brunton claims that although short-term parliaments can produce problems, they can also 'help foster more responsible and accountable governments'.(42)
Others have observed that the solution to this 'problem' lies with the Prime Ministers who have consistently reduced the term of Parliament by calling early elections. The current Clerk of the Senate, Harry Evans, has said that if Prime Ministers could restrain themselves, at a stroke we would have longer, and therefore, more stable parliaments. Nearly twenty years ago, three political scientists felt able to assert that it had 'never' been the case that Australians were bothered by elections every three years. The problem, they claimed, lay with Prime Ministers who did not allow the parliament to last the full term: 'It is the constant possibility of premature parliamentary elections that is so destructive of good government'.(43)
One difficulty with moving from the three-year term is the lack of evidence that such a change would actually bring the benefits that are claimed. Surprisingly, there appears to have been no research undertaken on the consequences of the change to four-year terms that occurred in four of the States in the 1970s and 1980s. There is, in fact, a lack of international evidence in regard to this aspect of legislative behaviour, no doubt because political scientists have regarded it as a settled question in most countries. Even were research to be done, however, the findings could only be speculative. Claims are made about the deleterious impact of three-year terms upon the Australian economy, but as far as can be ascertained there is no methodologically sound study that establishes, without doubt, that economic performance has been materially affected by a legislative term.
A second problem relates to the contention that the existing term has a deleterious impact upon the legislative performance. Critics point to the rush to legislate before the end of a parliament, but seem not to consider the possibility that the shorter term acts as a strong motivating instrument to get planning under way and legislation passed promptly. In addition, extending the House term to four years will not necessarily see the improved pursuit of medium- and long-term planning strategies. In many cases lengthy periods may be required after the passage of legislation before policies are seen to be producing results. The required lead-time may be far longer than four years and the difference between three and four year terms may therefore be quite marginal.
In summary, although there is a lot of sentiment in favour of the four-year term, it is all based on speculation rather than hard evidence. Some, at the least, might prefer to stick with what is known rather than take the punt on longer terms. Conversely, four-year terms seem to work satisfactorily in the five States that have them.
Modifying Westminster-four-year fixed terms?
Some argue that the power to choose the election date gives a substantial advantage to the Prime Minister, allowing 'arbitrary, partisan and capricious early elections'.(44) It is clear that while the House has a maximum term only, politics is going to remain in the equation, whether the term is three or four years. Three University of New South Wales academics have described a four-year non-fixed term as:
The worst of all possible worlds. It gives an extra year to a government without accountability to the people and yet the opportunity for a prime minister to call an early election at will still remains.(45)
The claim is therefore made that the only way to eradicate this is to introduce fixed terms. Even in Britain there have been occasional proposals to introduce fixed term parliaments to remove this Prime Ministerial prerogative.(46)
Most Australian debate of this type has focussed on four-year fixed terms, though three-year fixed terms would be quite possible. In her study of the length of lower house terms, Thompson saw merit in three-year fixed terms, primarily on the grounds of their retaining 'more accountability', with governments being 'sensitive to the need for re-election throughout their term, rather than just at the end'.(47)
There has, though, been more interest in four-year fixed terms. In a Legislative Council debate in 1984, the South Australian Attorney-General, Chris Sumner (ALP) identified what he claimed were the benefits that would flow from the introduction of such terms for the South Australian Parliament. His claims are summarised here:
At the national level, Harry Evans, has pointed out that a change to fixed terms would also help provide a solution if the Supply problems of 1975 were to be repeated. The insertion into the Constitution of a fixed House term, that could only be shortened by a motion constitutionally identified as a motion of no confidence, would withdraw the usefulness of blocking or rejecting Supply as a parliamentary tactic.(49)
The ALP developed an interest in fixed terms in the late 1970s and the New South Wales Labor Opposition supported the change to fixed terms in the 1995 referendum held in that State.(50) There is currently a four-year fixed term proposal before the Victorian Parliament,(51) and the former national leader, Gough Whitlam, still calls for fixed four-year terms for every Australian Parliament, Commonwealth and State.(52) Whitlam has warned that the rising costs of campaigning leaves the process open to corruption: 'The cost of campaigns is the greatest source of political corruption confronting the Western democracies'. He wonders if creating fixed terms of parliaments would lessen the possible danger.(53)
One Labor proposal for fixed terms was in fact passed by the Senate in November 1982. This particular model involved simultaneous House and half-Senate elections on the third Saturday in November every third year. The only exceptions to this would be (a) when a government was defeated in the House on a formal vote of no confidence, and (b) when a double dissolution was called following a deadlock between the houses. If an extraordinary election were forced by either (a) or (b), the incoming government would be able only to see out the term of its predecessor to the normal date for a general election.(54) Geoffrey Lindell, Professor of Law at Melbourne University, has described what he sees as the advantages of such a system:(55)
By contrast, some Coalition politicians, such as Malcolm Fraser and Peter Durack (LP), have been determined opponents of fixed terms, seeing it as 'inappropriate, ineffective and dangerous':
It is this final point about the modification of the Westminster model that is of great concern to some critics. Former Chief Justice Dixon has pointed out that in our system of government, we 'insist on the dependence of Cabinet upon Parliament'. Furthermore, our governmental arrangements provide that 'if a difficulty arises between the executive government and Parliament, it shall be resolved by an appeal to the people'.(58) This is not possible under fixed term arrangements, except when the constitutional change is written so as to allow for earlier elections if a substantive no-confidence motion is passed.(59) Former law lecturer, and later Labor Minister for Justice, Michael Tate, is another who has expressed doubts over this model, focussing on the conventions of responsible government. He has particularly wondered what such a change would do to the office of the Prime Minister. At present the Prime Minister can go to the people if circumstances warrant it, and Tate believes this power ought not be thrown away lightly. For example, if a Prime Minister is being frustrated by the Opposition, 'he ought to be able to go to the people and renew his mandate'-it is what Professor Don Aitkin of the University of Canberra has described as the 'flexibility' that exists in the Westminster model.(60) The calling of the 1974 election by Whitlam is an illustration of what Tate had in mind.
David Clune of the New South Wales Parliamentary Library has noted other consequences. He maintains that if fixed terms are introduced then the Vice-Regal reserve powers have to be excluded, or else their existence must be acknowledged in the relevant legislation and constitutional amendment. The New South Wales Parliament took the latter option when fixed terms were introduced in that State, arousing strong opposition. Clune also notes that under a fixed term 'a government could be fixed to an election in inauspicious or unfortunate circumstances that militate unfairly against its chances of re-election'.(61)
Not all on the Coalition side of politics are opposed to fixed terms. In New South Wales the change to fixed terms seems to have been accepted by the major parties, with no pressures to repeal the legislation. Former Liberal Senator David Hamer has suggested that the idea of a fixed three-year term-with the unchanged fixed Senate term of six years-would be a way of avoiding the criticism that would be heard if the Senate term were increased to eight years.(62)
An interesting point made by the English doyen of election studies, David Butler, that is usually not referred to in the Australian context, relates to the issue of the cost of elections. In an article in the Blackwell Encyclopaedia of Political Science, Butler notes that flexible election dates-as in Australia-tend to produce shorter, and therefore cheaper, campaigns. Uncertainty over the date of an election means that there is usually little to be gained by an Opposition beginning to campaign well before the Prime Minister's announcement.(63) The Labor 'mini-campaign' of 1971 remains the only example in Australian electoral history. By contrast, a well-known feature of US national elections-all of which are fixed term-is just how long the campaigning can take. This is not only because of fixed terms, but is certainly exacerbated by that aspect of the system.
Writing nearly twenty years ago, Uhr claimed that while few Commonwealth MPs opposed four-year terms, 'there is a considerable number of opponents to the notion of a fixed parliamentary term'.(64) The position is probably still as Uhr described it, despite the Joint Standing Committee on Electoral Matters receiving a number of submissions in favour of fixed terms.
In an effort to find a path through the thicket, some support has been heard for a mixed constitutional arrangement, which would combine elements of both maximum and fixed term arrangements. The model usually referred to would give the House of Representatives a maximum term of four years, with the House not dissolvable during the first three years after an election. This restriction on early dissolution would be subject to some exceptions.(65) This is the constitutional arrangement introduced in Victoria in 1984 and South Australia in 1985.
In Victoria the Legislative Assembly cannot be dissolved during the first three years unless:
Legislative Council terms are equivalent to two terms of the Assembly, with half of the Council resigning at each election.(66)
In South Australia the Legislative Assembly cannot be dissolved during the first three years unless:
As with Victoria, Legislative Council terms are equivalent to two terms of the Assembly, with half of the Council resigning at each election.(67)
This model has recently been recommended for introduction in Queensland by both the Queensland Constitutional Review Commission and the Parliament's Legal, Constitutional and Administrative Review Committee.(68)
It can be suggested that this constitutional arrangement brings together several major benefits. Longer terms would presumably have the effect of bringing more stability and predictability to the political system, but the Prime Minister would still have the flexibility of choosing the election day, albeit only in the fourth and final year of each parliament. The period of uncertainty and indecision that inevitably affects governments prior to an election could be restricted to just the last quarter of a parliamentary term.(69) The Melbourne Age has suggested that this model is a 'useful' guide to reform in Canberra, because:
It combines the benefits of certainty and consistency of tenure without denying political parties their craving for flexibility.(70)
Furthermore, Senator Alston (LP) has described this model as having 'considerable appeal', for he sees it as protecting the staggered electoral arrangement for the Senate which he describes as a safeguard for Senate independence.(71) Former Queensland Premier, Russ Cooper (NP) criticised the Goss Government's 1991 four-year term proposal for its failure to guarantee at least a minimum term of three years:
The failure of the Government to guarantee a minimum term of three years is the major fault in this Bill.(72)
Any discussion of longer House of Representatives terms raises the important question of how such terms would be co-ordinated with the terms of the members of the Senate. Simultaneous elections are not a constitutional requirement, but they save money, and only six of 39 House elections have been held alone-the last was nearly 30 years ago in 1972. The Australian experience has therefore been that the three-year/six-year model makes it relatively easy to hold both elections on the same day. However, if the House of Representatives maximum term became four years, with no alteration of Senate terms, simultaneous elections would be much rarer.
The question of Senate terms therefore cannot sensibly be avoided in any discussion of the possibility of extending House terms to four years.
Although there has been much consensus about the need to lengthen lower house terms, there is far less agreement about what should be done with the terms for the upper house:
as one works one's way down the list of proposed reforms, one inevitably enters the tangled thicket of Australian bicameralism from which ... few investigations emerge unscathed'.(73)
Possible options that have been given some publicity are listed in Table 5 and are discussed below.
Table 5: Possible models for House and Senate terms
|
Model |
House term |
Senate term |
Simultaneous election? |
Comments |
|---|---|---|---|---|
|
A |
4 year (maximum) |
8 year (fixed) |
No |
- |
|
B |
4 year (maximum) |
8 year (fixed) |
Yes |
This would introduce simultaneous elections |
|
C |
4 year (fixed) |
8 year (fixed) |
Yes |
Used in New South Wales |
|
D |
3 year (minimum) |
8 year (maximum) |
Yes |
Senate terms would equal two House terms. Used in Victoria and South Australia |
|
E |
4 year (maximum) |
6 year (fixed) |
No |
Used in Japan. Tasmania has six-year fixed terms for its Legislative Council, but with annual elections. |
|
F |
4 year (maximum) |
4 year (fixed) |
Yes |
Used in Western Australia |
|
G |
4 year (maximum) |
4 year (maximum) |
Yes |
Rejected by voters in 1988 |
|
H |
4 year (fixed) |
4 year (fixed) |
Yes |
Proposed for Victoria by Bracks Government. |
|
I |
4 year (maximum) |
4 year (fixed) |
Yes |
Half of Senate elected every two years. |
Model A (four-year maximum, eight-year fixed, no simultaneity)
This change would increase the House term by a year and the Senate term by two years, with the co-ordination of election dates for the houses being optional, as has always been the case. The major doubt about this change is whether extending Senators' terms to eight years would be accepted by voters. Might it, asks Kathryn Cole, formerly of the Parliamentary Library, be seen as 'cynical and self-serving by a somewhat jaded electorate'?(74) Is an eight-year term just too long? While praising the idea of four-year House terms, the Age described eight years between elections as 'a strange concept of democracy'.(75) Professor James Crawford has noted that:
An eight year term for Senators is a very long one, which stretches any notion that parliamentary office is the result of a more-or-less current mandate from the electorate.(76)
During a debate in the upper house, former Senator Reg Withers (LP) asserted that if Senators gained a longer term it would 'almost be as good as being a tenured academic or a tenured public servant'.(77) Major parties might also reject the idea of some minor party Senators holding a seat for such a length of time. This has been an issue in New South Wales since the 1999 election following the election of some MLCs with very small shares of the vote.(78)
Despite such views, this would be the change that would see the maintenance of a system very much like the current arrangements, and might, therefore, be the easiest to sell to defenders of the Senate-and even to an electorate traditionally sceptical of constitutional change. There have been many politicians who have supported it. Tim Fischer (NP), for example, has expressed his support and, incidentally, the support of his party as well.(79) Peter McGauran (NP) has acknowledged the difficulty of selling eight-year Senate terms to a sceptical public, but sees it as a question of overall benefit to the Parliament and, hence, the nation. For him, the advantages of the four-year House term outweigh the disadvantages of longer Senate terms.(80) John Spender (LP) also worried about the longer Senate term, yet could see:
No sensible way of improving the system other than by having a four-year period for the House and an eight-year period for the Senate.(81)
It is perhaps relevant to note that eight-year upper house terms were introduced in four States during the period 1972 and 1987. It is sometimes stated that extending Senators' terms in this way would weaken party discipline, but there is no evidence from the States that this would be so. Senators seeking re-election would still need party pre-selection and an undisciplined performance would not be the way to guarantee continued party support.
The ALP has taken the view that it would be preferable to make this change if there were also the introduction of simultaneous elections (Model B), as well as a reduction in the power the Senate holds to destroy a government that controls the House of Representatives.(82) This view has been held despite the fact that it was Labor governments which introduced longer upper house terms into Victoria and South Australia, States with powerful Legislative Councils. The journalist, Paul Kelly, is another who has argued against such a change while ever Senate powers remain intact: 'Any proposal to marry the Senate's existing powers with even less frequent Senate elections is the worst [proposal] imaginable'.(83)
Model B (four-year maximum, eight-year fixed, simultaneity)
The only difference between Models A and B would be the introduction of simultaneous elections for the two houses under the latter arrangement. The 1974, 1977 and 1984 attempts to introduce simultaneous elections caused much controversy, being described by opponents of the 1984 change as 'the first stage of an attack on the Senate, the States' House, which provides the checks and balances of Australia's system of government'. If this model were attempted, the four-year House term alteration might be lost in the effort to defeat the simultaneous election provision.(84)
Model C (four-year fixed, eight-year fixed, simultaneity)
The model is now in use in New South Wales. This was introduced as the result of a memorandum of understanding between the Fahey Coalition Government and the independent members of the Legislative Assembly. There had already been legislative provision for the 1995 election to be held on a specific day (25 March), and the provision was made permanent as a consequence of a referendum held in conjunction with the 1995 State election. As mentioned earlier in this paper, there is some doubt-including that expressed by Prime Minister Howard-about the appropriateness of a fixed lower house term in a Westminster-style parliament. In the New South Wales Parliament similar reservations were expressed, though the Coalition Government proceeded to legislate for the 1995 referendum with the justification that the people ought to be given a chance to register their view on the question.(85)
Model D (first three-years fixed/four-year maximum, eight-year fixed, simultaneity)
This model, referred to earlier, which sets a maximum term of four years was recommended by a Standing Committee of the Australian Constitutional Convention (1982) and the Constitutional Commission (1988).(86) The actual length of Senate terms would probably be seven-eight years if, as claimed above, the House was usually not dissolved earlier than six months prior to the expiry of its term. Elections could be called in the first three years of a term only if the government lost control of the lower house, so that while a Prime Minister could not procure an election in under three years, nor could the Senate force an election in the first three years of a government's term. Some would argue that there would presumably need to be a concurrent alteration of section 57 of the Constitution so that a double dissolution could take place only in the fourth year of a House of Representatives term. The model therefore meets business leaders' concerns about the frequency of elections, though the fact that upper house terms would be twice lower house terms, would probably be seen as an attack upon the position of the Senate. Former Senator Peter Durack (LP), for instance, claimed that to make Senate terms equal to two House terms:
Would very much derogate from the Senate's position, its independence and its position under our Constitution as a States' House.(87)
Such has not been the case with the Legislative Councils in either Victoria or South Australia, both of which have retained their full range of powers since this alteration was made.
Model E (four-year maximum, six-year fixed, no simultaneity)
Writing in the late 1980s, Peter Reith (LP) suggested the introduction of this model largely, it seems, because other models which altered the Senate's term in some way had either been rejected in constitutional referenda or else were politically unacceptable. He suggested that this model was 'the only practical and sensible means of achieving the extended term for the House of Representatives'. He also spoke of the inevitable separate half-Senate elections that would occur as enhancing the power of the upper house by 'allowing the Senate a higher profile'.(88)
There are two main objections to this model. Reith himself has acknowledged that such a change might be a recipe for more, rather than fewer, elections. As referred to elsewhere, numerous elections can be disruptive to government planning; they would certainly mean a higher election bill over the long-term. The second problem harks back to the period between 1964 and 1970 when the last three separate half-Senate elections were held. Critics were concerned about the 'by-election' aspect of these elections that could distort the result of the previous House election. To objections that our elections are now unlikely to be put out of kilter, supporters would point to the period in 1989 when some political commentators spoke of the possibility of a half-Senate election being held in the last half of 1990. This would have helped the government postpone a House election until the last half of 1990.(89)
On the other hand, the Joint Committee on Constitutional Review (1959) believed separate Senate elections were to be avoided:
Separate elections for senators ... do not, in practice, affect the formation of a government, and while, on the one hand, it is possible that the Senate election results may give the government gains in the Senate, it is equally possible that the results may embarrass the government. It is not conducive to sound government that the future of a recently elected government should depend upon the eventualities of elections for senators which take place during the normal life of the House of Representatives.(90)
This is a model that is unlikely to receive much support, if only because of the likely long-term financial cost.
Model F (four-year maximum, four-year fixed, simultaneity)
and
Model G (four-year maximum, four-year maximum, simultaneity)
and
Model H (four-year fixed, four-year fixed, simultaneity)
Model F has been in use in Western Australia since 1987, Model G was one of the four constitutional changes sought by the Hawke Government in 1988, and Model H has recently been proposed by the Bracks Labor Government in Victoria.(91) They are close enough in their likely effect to be dealt with together.
If introduced at the Commonwealth level, the three models would result in fewer elections. In addition, it has been claimed that this type of change 'would reflect the views of the people more accurately than is the case under the present arrangements'. The example usually given is that of the Whitlam Government, elected in 1972, which had to face a Senate elected in 1967 and 1970.(92)
The contrary view would claim that the review role of the Senate is in fact aided by the system of different length terms and staggered elections-that there is little point in having a house of review that merely mirrors the makeup of the lower house. The existence of half-Senate elections are said to be a reminder that the Senate was intended to be (at least in part) a house of review, 'which acts as a brake on hasty and ill conceived legislation'.(93) Many years ago Professor Geoffrey Sawer of the Australian National University stated before a Senate select committee that:
There is no point in having a House of review unless you have some degree of difference between the points of view of the Houses, and you get that more with the staggered system of elections.(94)
The Liberal and National Parties have both been critical of any proposal to reduce Senators' terms to four years, largely because of a conviction that this would threaten an unacceptable reduction in the power and influence of the upper house.(95) In the Victorian case, the two parties have dismissed the Government's proposal as a 'political stunt'.(96)
Critics also have noted that such a change would make it possible for the Prime Minister to force the Senate to an election if the upper house were causing problems for his or her legislative programme. This, in turn, might make Senators warier of performing their review function. In 1988 the official NO case stated:
There is more to this proposal than meets the eye. It has hidden and dangerous consequences which would forever remove the essential checks and balances in our democratic system of government. It means more power for the Prime Minister and the Government, and less power for the States. It does not even guarantee fewer elections or better government.(97)
An additional and practical problem for the major parties (assuming the continuance of proportional representation for Senate elections), would be the effect of this change upon election results. At present, a quota of 14.3 per cent produces a Senate in which the major parties usually win about 85 per cent of the seats in any given half-Senate election. Given the decline in major party votes in recent years, it is quite possible that with a smaller quota of 7.7 per cent for a full Senate election, that the major party share of seats would decline by at least 10 per cent. Clearly, it would not be in their best interests to support such a change.
The fact that the 1988 amendment attempt produced the second-lowest YES vote of all constitutional referenda since Federation (32.9 per cent) suggests that such an alteration will not be attempted again.(98)
Model I (four-year maximum, four-year fixed and biennial, simultaneity)
This model was included in the Reith paper referred to earlier and was discussed by the Australian Constitutional Convention.(99) The unusual feature is that Senate elections would be held every second year. However, if an important justification for four-year House terms is a reduction in costly elections, it is highly unlikely that a system such as this would be considered, for it clearly would involve many more elections being held.
So, what are the main aspects of the debate on extended terms for the House of Representatives?
House of Representatives terms