This evening you have kindly invited me to speak about New Zealand’s experience of using proportional representation (PR) to elect the members of its Parliament. I am honoured to respond before such an important audience so familiar with the workings of parliamentary democracy.
Before I explain New Zealand’s experience I would say, as a background observation, that in my view there is no perfect electoral system. They all require trade-offs, and what is acceptable will vary between individuals, groups and countries. Further, as I will show, New Zealand has a history of seeking to improve its democratic structures—in which the decision to go to PR is but the most recent change.
The first moves to a constitutional democracy in New Zealand started in London when in 1839 Lord Normanby, the Secretary for War and the Colonies, instructed a Captain William Hobson to adopt the most effective measures for establishing amongst those living in New Zealand a settled form of civil government.
So the following year in 1840, with Hobson acting for the Crown, the Treaty of Waitangi was signed between the Crown and the Chiefs in New Zealand. This established New Zealand as a British Colony and gave Maori full rights as British subjects. After the Treaty was signed, the first move to involve the public in governing New Zealand was made in 1846 when the New Zealand Constitution Act was passed. It created a General Assembly consisting of the Governor, an appointed Legislative Council and an elected House of Representatives. Franchise was given to landholders able to read and write— a requirement that effectively excluded Maori from participation.
This initial burst of enthusiasm for elected government came to an abrupt halt the following year, 1847, when the 1846 Act was suspended for five years, as it was on reflection considered to have marginalised Maori unduly. The 1846 Act should be marked down as our first failure to put in place a fair electoral system.
There was then a pause until 1853 before the first elections for the House of Representatives were held. They were a little different than today because unless a poll was demanded voting was by a show of hands. Fourteen years later, in 1867, to meet the desire to include everyone in the political process, four seats were established on a temporary basis for Maori voters. As often happens in politics, the temporary seats became permanent (in 1876) and then remained unchanged for the next 120 years until the first mixed member proportional (MMP) election in 1996. I have introduced the question of Maori participation early because the issue of Maori gaining fairer representation in Parliament was an important argument put forward by those supporting the move to PR.
In 1893 New Zealand made its biggest breakthrough in electoral reform by being the first country in the world to grant women the vote. However, the ever cautious reformers did not move to allow women to stand for Parliament until 1919. Again, an important issue in the debate leading up to New Zealand’s agreeing to adopt PR was that the system would likely ensure more women became MPs.
In 1951 New Zealand decided to abolish its upper house, called the Legislative Council. The Council was an appointed body and was considered little more than a rubber stamp for the decision makers in the House of Representatives, so out it went. This lack of a Senate or Second Chamber of some kind was also put forward as a reason why New Zealand needed to change its electoral system.
Over the years, many other changes have been made to modernise and update New Zealand’s electoral procedures, including first reducing the voting age to 20 years in 1969 and then lowering it again to 18 years in 1974.
The restive search to find the right electoral formula for New Zealand was put back into the public arena with the decision taken in 1985 by the then Labour Government—which had been elected by a very large majority in 1984—to establish the Royal Commission on Electoral Reform. This was not an isolated irrational act but a considered response to public concern that New Zealand’s electoral system did not produce a Parliament that reasonably reflected the wide diversity of views in the community at large.
These claims were not new but had been traditionally advanced by a cross-section of academics and, of course, all small political parties, who knew that they could never make real progress or exercise real influence under the first-past-the-post (FPTP) system.
The demands reached a new level of insistence, however, when the Social Credit party, which had been around for many years, gained 20.7 per cent of the vote in the 1981 election but won only two seats in Parliament. As the Royal Commission’s report points out, in that election it took 186,000 Social Credit votes to elect a member whereas for the two major parties it took only 14,900 votes for National to win a seat while Labour won one seat for every 16,300 votes. The National Party was elected to government by a narrow margin in 1981. In the 1984 election, which the Labour Party won, Social Credit support fell considerably but it still took 73,600 votes to elect a Social Credit Member of Parliament whereas it took only 14,800 votes to elect a Labour MP and 18,700 votes to elect a National MP. Of further interest for those arguing for PR, a new party at the 1984 elections, the New Zealand Party, attracted 12.3 per cent of the electorate’s vote but did not succeed in gaining a single MP.
The 1981 and 1984 elections highlighted in a graphic way that there were valid arguments to be heard in support of a change in New Zealand’s voting system, and hence the establishment of the Royal Commission.
In its very impressive and comprehensive report, the Commission canvassed all options but it seems clear to me that their unanimous decision to recommend change was an almost foregone conclusion: As shown in 1981 and 1984, the arithmetic clearly supported change. That being so, it is not surprising that the Commission found that New Zealand’s voting system had “serious deficiencies.” What was more difficult to anticipate and very much more difficult to prove, was the degree of instability a change in the electoral process could cause. More on that later.
On the two serious options considered for a proportional system for New Zealand—the system of mixed member proportional, as used in Germany since 1949 and the single transferable vote (STV) as used in Ireland—the Commission came to the conclusion that MMP “is clearly superior,” hence their recommendation that MMP should be adopted and that it should be a two-vote system where one vote would be for the constituency member and one vote for the preferred party.
To enable the proposed system to operate effectively the Commission recommended increasing the size of the New Zealand House from 99 to 120 members. Its reasoning was that a 120member House was necessary for an effective system with MMP, and that the members should be split with 65 MPs representing single-member constituencies and 55 MPs coming from the party lists, which would be made public in ranked order before the election. Following the vote, the “list MPs” would be drawn on to ensure that each party had the correct number of MPs relative to the support it had received in the party vote.
Remember, the constituency vote is largely irrelevant in determining the number of MPs each party has, other than in the situation where a party wins a constituency seat but does not gain the required threshold support of five per cent necessary for gaining representation in Parliament. In these cases winning a constituency seat means the votes cast for a small party do count in the allocation of seats.
Returning to the important question of Maori representation, the Royal Commission was clear in its view that an MMP system with a common roll—in other words no separate seats reserved for Maori—offered the optimal conditions for the effective representation of Maori interests. However, the select committee, after listening to the views of senior Maori leaders, who did not agree with the proposal to abolish Maori seats, rejected the proposal to have a common roll, and instead recommended an increase in the number of Maori seats from four to five.
Finally, the Royal Commission recommended a referendum to test the acceptability of its proposals with the general voter.
There was then another pause in the process as the Labour Government, which had established the Royal Commission, seemed to lose interest in electoral reform and failed to hold the recommended referendum before the 1990 election. With the change of government in 1990 the National Party reconfirmed its intention to hold a referendum and in fact decided to hold two referenda, one to give the voters a choice as to which form of PR they preferred and then a second binding referendum at the time of the 1993 general election in which voters would choose between the preferred PR option and the status quo FPTP.
The first referendum resulted in a clear-cut victory for MMP over the other three PR systems put forward. The scene was thus set for the most far-reaching reform of New Zealand’s electoral system in the 20th century.
To have a sense of the political environment at the time all this was being debated, remember that the leading members of the two major political parties opposed the proposed change. The point that I as prime minister made on many occasions was that while it was true that the status quo did not produce a parliament that precisely reflected voting patterns in the country, the alternative of PR would mean that smaller parties in a coalition would exercise more power in defining policy than their support in the community justified. In the minds of many, that is the classic trade-off that is central to any debate on PR electoral systems.
No move for change takes place in a political vacuum and in that context the lead-up to the 1993 referendum was instructive in terms of what eventuated. After the 1990 election the government I led had to take some tough and unpopular measures to correct serious economic problems. During the early stages of this reform process the government was very unpopular and support for a change to PR reached very high levels. However, as the new economic approach began to pay dividends and public antagonism waned, support for a change waned as well.
Still, it is credible to argue that public support for a change was based more on a desire to punish the government and MPs in general for the economic changes that had been introduced, than on any deeply held philosophical views that the electoral system was seriously flawed and needed radical change. Nevertheless when the referendum was held the vote was 54 per cent to 46 per cent in favour of introducing MMP. This less than overwhelming support for such a major constitutional change explains, at least in part, why support for change went into rapid reverse when the public did not like what it saw in the practice of MMP both before and after the 1996 election.
For a start, the three-year transition period created inevitable tensions. When the implications of reducing 99 constituencies to only 65 became clear after the 1993 vote for MMP, what we saw in New Zealand can only be described as an unseemly scramble by sitting MPs to either find seats or gain a high place on the party lists. This scramble for survival was New Zealand voters’ first exposure to some of the realities of the multi-party environment that was just around the corner with MMP and they did not like the experience.
Nevertheless, the 1996 election met some of the important ambitions of those supporting PR, with a larger number of women, Maori and minority MPs being elected than ever before. These extra women MPs and representatives from minority groups in the main came from the party lists, showing that party lists can indeed be used to create a more representative parliament, as had been the hope of the Royal Commission.
Forming a government after the 1996 election was a protracted affair with coalition negotiations taking two months before a new government was formed. The largest party in the new Parliament, the New Zealand National Party, which I led at that time, could form a majority government only by entering into a coalition with the New Zealand First Party, which was led by the Honorable Winston Peters.
The Labour Party was the second largest party in terms of seats but to form a majority government it would have had to gain the support of both New Zealand First and the far-left Alliance Party. The media was clearly of the view that after negotiations Labour would indeed form the government. It was widely presumed that National and New Zealand First would not form a coalition because of past disagreement between myself and Winston Peters, who had formed his own party after I had sacked him from Cabinet in 1991.
Our newsletter about the public service.
Nominated for a Digital Publishing Award.
The public watched and waited and as time dragged on uncertainty grew as commentators speculated as to what deals were being hatched behind closed doors in order to seal a coalition agreement. It slowly dawned on the voters that their sovereign right to elect the government, which hitherto had been accepted as a given, had been severely eroded by the introduction of the new system. They now only had the power to determine the size of the various political parties but the government would be formed by negotiations held in secret. Rather than increase the voters’ influence, MMP had moved them back one step in the process of forming a government. This is a key point to remember in any debate on PR.
After two months, Winston Peters finally concluded the drama by announcing live on television that his New Zealand First Party would form a coalition with the National Party. It was a shock for many supporters of PR that a coalition government could be formed between parties which only recently had been outspoken in their condemnation of some of each other’s personalities and policies. Many other observers concluded that the smaller parties had indeed gained too much political power. For many supporters of MMP the lead-up to and formation of the first coalition government in over 60 years was not a happy event and support in the polls for the old system of FPP moved ahead of MMP and has remained there virtually ever since.
This was just the beginning of learning to live with the new electoral order. The ongoing debate concerning the formation of the coalition government and the impact that it had on opinion poll support for both parties in the coalition was causing considerable tensions in the National Party and I was replaced as leader and stepped down as Prime Minister in December 1997, to be replaced by Hon. Jenny Shipley.
That decision solved little in terms of creating a more stable government and following further headline-generating high political drama the Prime Minister sacked the leader of her junior coalition partner, who was holding the posts of Deputy Prime Minister and Treasurer, and the coalition formally collapsed on 26 August 1998. For those who thought they had seen it all there was more to come. Another unseemly scramble for survival occurred and the New Zealand First Party, led by the now-sacked Mr. Peters, fell apart. Mr. Peters retained the loyalty of only nine of his former 17 MPs. Five formed another new political group and three became independents. Three of the former New Zealand First Party retained their Cabinet posts and therefore remained in the executive. The government led by Jenny Shipley now relied not only on the votes of independents and the five members in the new Mauri Pacific Party but also the support of the eight MPs in the Act Party. To that extent, although the Act Party had played no part in the 1996 negotiations, it now became a core support group for the now-minority National Party Government.
With this interesting group in support of her, the Prime Minister on 8 September 1998 moved a confidence motion in the House, which the government won. New Zealand’s then-Governor-General, Sir Michael Hardie-Boys, gave a speech in December 1998 on life with MMP from a constitutional viewpoint and commented that “in initiating the vote of confidence, the Government acted responsibly and democratically. Its legitimacy—which might otherwise have been open to question—was established beyond doubt.”
From my perspective there never was any constitutional crisis in New Zealand’s first post-MMP government. But equally there can be no dispute that there was a major political crisis. The coalition government that had been formed after a lengthy negotiation only 20 months earlier had collapsed in disarray and the party of the junior coalition partner had splintered into many parts. That was not the end, however. Over the next few months the youngest person ever to be a Minister resigned from Parliament after only two years service in the House, and the far-left Alliance Group split three ways. The Parliament elected after the first MMP election started with six parties but by the end of the three-year term that number had grown to nine parties plus three independents. This was not the brave new world that PR was supposed to deliver.
Despite everything—and much to the surprise of political commentators and all of the government’s political opponents—the minority government managed to complete the allotted three-year term. Constitutionally, New Zealand had survived what can only be described as our turbulent experience preparing for and operating with PR and the MMP system of elections.
The lead-up to the second MMP election in 1999 was more stable and the outcome more predictable, given the large lead enjoyed in the opinion polls by the Labour Party, which for many months before the vote looked set to form the new government with the support of the Alliance Party.
When the votes finally were counted Labour had 38.7 per cent and 49 seats and the Alliance had 7.7 per cent of the votes and 10 seats. Between them they just failed to have a majority in the 120-member House, but in forming the second minority government under MMP they had the comfort of knowing that the Green Party, with five per cent of the vote and seven MPs, was very likely to support the new coalition on most of its policy agenda (some economic issues and international trade agreements apart). Where the Greens’ support could not be counted on the new government could with reasonable confidence count on the support of the two parties on the centre right.
There were other important differences between the formation of a coalition government after the first and second MMP elections. For example, there was widespread expectation that Labour and the Alliance would form a government in 1999 whereas there was little or no expectation that National and New Zealand First would form one after the 1996 election. Another big difference was that whereas in 1996 New Zealand First negotiated with both major parties before deciding which coalition to join, no such horse trading occurred in 1999. The lengthy negotiations in 1996 resulted in a long and detailed coalition policy agreement but in 1999 the coalition agreement was less than two pages in length and dealt with principles rather than details. Significantly, the Labour/Alliance agreement is more open to accepting disagreement on policy positions— where the issue is described as “party distinction.” It could be argued that Labour’s willingness to accept disagreement from the left is because, as noted earlier, it can with some comfort rely on support from non-coalition parties on the right on some key issues.
An overview of the experience thus far of the governments formed after the 1996 and 1999 elections is that after the 1996 poll protracted negotiations caused uncertainty and unwelcome surprises and produced a detailed agenda for the new government. By contrast, after the 1999 poll the government was formed quickly and as predicted, and the arrangements between the parties were more flexible.
The greater stability evident to date in the government formed after the 1999 election has restored some confidence in MMP, but the market research company that has conducted a tracking poll since the 1996 election reports that on a “head to head” choice between FPTP and MMP, FPTP has been preferred to MMP in every poll but one— in December 1999, immediately after the new Labour/Alliance government was installed.
A different breakdown of attitudes, on the overall question of retaining MMP as is or moving to another electoral system, shows that only 17 per cent want the system to stay as it is, 31 per cent want to change to a new system, and 47 per cent say the basic structure is acceptable but changes are needed in how it operates.
Not surprisingly, 53 per cent agreed and only 25 per cent disagreed that MMP had made politics a lot more messy and confusing. On a key question regarding PR, 52 per cent felt that the system had given minor parties too much power, while 29 per cent disagreed with that proposition.
Another key issue is accountability. Here 61 per cent said that list MPs are not as accountable to voters as electorate MPs, and only 15 per cent disagreed with that. In the lead-up to the introduction of MMP many argued the need to get the agreement of minor parties would mean that coalition governments formed under MMP would not be able to make hard decisions or take decisive action and the public agree with that view by 52 per cent to 21 per cent.
The research shows that seven and a half years after they voted for MMP and four and a half years after the first government under MMP was formed New Zealand’s voters want another say on the future of the electoral system. A commanding 76 per cent say they favour another binding referendum on whether to keep PR in its present form or move to something else.
These poll findings are instructive, especially in light of the fact that since the 1999 election most commentators would say that the turmoil experienced earlier has been replaced by a much calmer political environment— although peace is not yet complete: For example, a number of Cabinet ministers have been sacked or suspended for various reasons.
One of the major concerns in the first three years of the MMP system was the manner in which elected MPs switched parties or formed new ones. The current government has introduced legislation to try and restrict such party-hopping, but its proposals have run into strong opposition in Parliament on the grounds that such rules would give party leadership too much power and would limit the rights of MPs to decide issues on their merits. At this stage the coalition government is considering amendments to its proposal.
So where to from here? This time it is not for me to determine. Many of the benefits that had been hoped for by those who promoted PR have in fact emerged. The Parliaments elected in 1996 and 1999 did produce a more representative legislature in terms of ethnicity, gender and the representation of minorities. MMP has also delivered on the core issue of having the number of MPs gained by each party be more in line with the share of the vote they gained.
But despite this, research shows that considerable doubt and uncertainty remain. Only 17 per cent of poll respondents want MMP to stay as it is, and worries persist about the accountability of list MPs, the power of small parties and the overall impression that the new system is messy.
Those who want to see a silver lining would agree if you combine the 17 per cent who are satisfied and the 47 per cent who say the basic structure is all right but changes are needed in how it operates, then there is a comfortable majority in favour of retaining an amended form of PR. Unfortunately, a remedy is not that easy: Many of the concerns that voters have are inherent with virtually any form of PR.
The next challenge for New Zealand’s policy-makers is deciding what changes to the current PR system should be contemplated before another referendum on the electoral system is held—because the public definitely expect to get another chance to express their views.
In closing, I would observe that New Zealand embarked on a major change to its electoral system, not out of any crisis, but because we wanted a better or more fair system. The lack of crisis surrounding the FPTP system perhaps explains why a high level of ambiguity continues to exist regarding the acceptability of MMP.
There can be no doubt that New Zealand’s electoral system has come a long way since its first attempt at popular participation back in 1846, but I fear that there are many who would still agree with the observation of the American humorist, Will Rogers, who said, “the more you read and observe about this politics thing, you’ve got to admit that each party is worse than the other.”
Perhaps the same is true for political systems.