kicking the tyres Choosing a voting system for New Zealand

Transcription

1 kicking the tyres Choosing a voting system for New Zealand by steve thomas

2 contents Kicking the Tyres. Choosing a voting system for New Zealand 1 Evaluating Voting Systems 2 Mixed Member Proportional (MMP) 7 First-Past-the-Post (FPP) 11 Preferential Voting (PV) 15 Single Transferable Vote (STV) 19 Supplementary Member (SM) 23 Conclusion 27 Endnotes 29

3 acknowledgements We wish to acknowledge Ian McAllister, Graeme Edgeler and Malcolm Mackerras for offering advice and thorough and insightful comments on draft versions of this paper. Responsibility for the final paper and the views expressed within rests with us. about the author Steve Thomas is a Researcher at Maxim Institute. He was educated at the University of Canterbury, graduating with a Master of Arts with Distinction. Steve s background is in politics and history, and he has researched and written in a number of areas including New Zealand politics, education and taxation. He is the author of two of Maxim Institute s Tax Discussion Series papers, its Roll Play education report, which examined how access to Christchurch schools could be improved for more families, as well as two reports in the Institute s award-winning series of Parent Factor reports: Information for parents and Access to education. He also published his first book in 2008, Cotonou and Pacific Regionalism.

4 kicking the tyres Choosing a voting system for New Zealand New Zealanders have had to wait long enough for a chance to kick the tyres on MMP. So, National will give them that chance by holding a binding referendum on MMP by no later than And, if a majority of voters decide MMP is not their preferred electoral system, we will offer them a choice between a range of electoral systems to replace it. John Key, Speech to the Annual National Party Conference, Wellington, 3 August At the 2011 election in November, voters will get the chance to vote in a referendum on whether New Zealand will keep its Mixed Member Proportional (MMP) voting system. Voters will also get the chance to decide which voting system they would prefer if we do not keep MMP, out of four possible alternatives. If a majority of voters decide that MMP ought to be kept, then parliament has said that there will be a review of MMP to find ways of improving it. The referendum is an opportunity to kick the tyres on MMP, and see whether it should be kept or whether the voting system should change. This should not be done without reasoned and considered national debate of all the options. Every voting system operates within a specific political culture and party system and these factors can influence the way voters behave in using a particular voting system. 2 This paper contributes to the debate about which voting system New Zealand should have by evaluating the five different voting systems that voters will be asked to choose from at the 2011 referendum MMP; First-Past-the-Post (FPP); Preferential Voting (PV); Single Transferable Vote (STV); and Supplementary Member (SM). We shall evaluate each system according to what we believe are criteria of effective representation. We have limited ourselves to the questions which the referendum is asking about the voting system. We therefore do not consider elements of the voting systems debate that are outside of this scope. We hope that our evaluation will help voters to decide which system they will choose at November s referendum. kicking the tyres - 1

5 evaluating voting systems There is no perfect way of turning everyone s votes into 120 politicians in parliament. Different voting systems use different methods. This year, New Zealand is asking which system is best for our country. What we decide will tangibly affect which MPs are elected, what parties get representation and how governments are formed. 1 For example, some systems (majoritarian ones) make it easier for major parties to win most of the seats, while other systems (proportional ones) make it easier for minor parties to win seats and normally mean parties have to share power. 2 There are strengths and limitations to any system. Choosing a voting system involves making trade-offs between each system s strengths and limitations. So, how do we decide what s important? 3 Simply put, New Zealand should have a voting system that enables representative democracy to flourish. That is, the voting system should enable our MPs and the government to have the freedom to lead and to make decisions in the best interests of all New Zealanders, but it should also encourage them to listen and be responsive to the interests of New Zealanders and their local communities. 4 Evaluation criteria We have developed seven evaluation criteria and have organised them under two major headings: how voting systems enable quality representation; and how they affect the workings of parliament and government. 5 No system will get a tick against every criterion, and in fact, some of the criteria might exclude each other. For example, there is often a trade-off between the government being able to pass the laws that it wants, and making parliament more representative of society at large. Our criteria cannot be used to produce a tally to decide which voting system is best. Instead they help us to understand the systems, so that we can vote according to what we believe is most important for quality representation. 2 - evaluating voting systems

6 Representation Representation is the life-blood of democratic government. Instead of four million people trying to decide every law together, we have people acting on our behalf to do so those people are our MPs. When it comes to voting systems, the question becomes which system can produce a parliament that represents the different communities and interests in our country well? There are two common ways to think about how representation can be provided for in parliament through MPs who are from our local, geographic communities (called electorate representation); or through MPs who represent interest groups to which people belong or with whom people agree (called representation of interests). 1. Electorate representation how important are electorate MPs in the system? Electorate MPs are connected to, and held accountable by, a community of people who live in a particular place. A community knows that an electorate MP will have their concerns in mind in parliament. 6 This concern provides for a relational connection and loyalty between MPs and voters. When an MP represents an electorate, their responsibility is quite clear they represent the people who are from their area while they make decisions about what is good for the country as a whole. 7 Electorate representation means that all local areas, whether rural or urban, North Island or South, are heard in parliament. 2. Representation of interests how does the system provide for the representation of interest and identity groups in parliament? Identity is not just shaped by location. For example, people are also shaped by the social, ethnic and religious communities to which they belong or with whom they identify, which may not be located all within one electorate. 8 In a diverse and complex society, there is a range of other interests that need to be considered when deciding what is good for everyone. It is important that parliament can reflect the breadth of society so that minority voices have the chance to be heard in parliament. All voting systems provide the opportunity for the representation of interests in some way, but some systems make it easier for people from minority interest groups to be elected and for those views to be easily represented. Given that society is comprised of a diversity of communities it is difficult for two or three major political parties to represent all of the different identities and groups. Majority rule cannot always take into account the wide range of these perspectives in the decisions which affect the community. 9 This is why some countries use proportional systems. They are intended to elect a parliament that reflects a wide cross-section of the different interest and identity groups. It is impossible to measure exactly how a system does or does not cater for interest group representation. This is because an interest group can be represented by someone who is not part of that group themselves. For example, a non-pasifika MP can act on behalf of the Pasifika community in parliament. But systems that are proportional are typically seen to be best at evaluating voting systems - 3

7 providing for interest groups to be represented. We will use the proportionality of parliament as our measure for how well interest groups are represented. That is, whether the number of seats each party has in parliament is roughly in proportion to the number of votes that it receives. Proportionality is an imperfect measure because effective representation of interests relies on more than the sheer number of people from different interest groups being elected to parliament. However, it is the most helpful and concrete indicator available. When it comes to voting systems, there is usually a theoretical tension between giving priority to the representation of interests and giving priority to electorate representation. In practice, the two do not need to be seen in opposition. In fact, they tend to operate together to produce effective representation. Every MP comes from a particular ethnicity and might stand for that group s interests in parliament, but they can also represent the interests of their local electorate and the people who they know of other ethnic groups, whose wishes and needs they consider as they think about what is good for the country as a whole. The two styles of representation do not exclude each other they go hand-in-hand. Effects on parliament and government We also need to think about what the voting system means for how parliament functions in practice. Voting systems influence what sort of government is formed and how parliament does its job. Particular criteria to consider are: 1. Accountability does the voting system help voters hold the government and MPs to account for their performance? To quote the philosopher Karl Popper, throwing out the rascals ought to be easy if they are not performing. 10 In voting systems where people are chosen through party lists instead of an electorate, this accountability can be reduced as MPs are reliant on their party s success more than the support of local voters. 2. Legitimacy does the voting system deliver the electoral outcomes that voters, as a whole, want and expect? Voters should be confident that their vote counts towards the fair election of candidates and parties. The voting system should not be biased but should assign all votes the same value as much as possible. 11 Voters should also be confident that their vote contributes towards the outcome of an election and that they can accept the result as fair and reasonable even if the government which they did not prefer is elected Stable government does the voting system enable governments to form easily and do they last for their term of office? Stability helps the government to run smoothly, thereby helping people to have confidence 4 - evaluating voting systems

8 in the government and its work. The voting system can affect how stable the government is. For example, if a voting system makes coalition government more likely, so that a number of different parties will have to work together, then the government may not be very stable Effective government how easy or difficult is it for the government to carry out what it has promised? Governments need to be able to function effectively. If the voting system produces fractious parliaments then it can be difficult for governments to do what they pledged to do when they were elected Opposition and oversight does the system promote the formation of an opposition that can criticise and challenge the government? An effective opposition is required so that the government s work gets debated, considered and evaluated. 15 For this to happen, a viable opposition party, or parties, need to be elected. Having an opposition will not guarantee that these things happen well, but they do make it more likely. If parliament is fractious then this may affect how unified and therefore how effective the opposition is. It is important to remember that these criteria cannot all exist perfectly in any system. For example, the criteria of effective government and opposition and oversight tend to compete. The best systems will give some room to all the criteria and will balance these competing elements. We now turn to our evaluation of the five different systems that will be considered at the referendum in the order in which they will appear on the ballot paper, beginning with MMP. 16 evaluating voting systems - 5

9 6 - mixed member proportional

10 MMP mixed member proportional How does MMP work? New Zealand has been using MMP as its voting system for fifteen years since MMP blurs the usual categories that voting systems fall into, by combining electorate representation with a proportionally elected parliament. 1 This combination happens by giving people two votes instead of one. One vote is for a local candidate, and one vote is for a party. The second vote the party vote is the one that decides each party s total degree of representation in parliament. The idea is to make the allocation of seats through all of parliament proportionate to the share of the party vote that each party receives. The first candidates to be given seats in parliament are the ones who win the most votes in their local electorate. So each patch of New Zealand has their own MP who looks after their interests. Then parties are allocated however many top up seats they need to get the overall proportion right. These top-up seats are filled by MPs from a list of the party s candidates that is drawn up before the election. 2 So, if a party wins ten percent of the votes, it gets roughly ten percent of the seats. If it wins forty percent of the vote then it gets roughly forty percent of the seats, and so on. 3 But to get any sort of representation, a party has to achieve one of two things. They either need to have a candidate The illustration opposite shows the key features of how MMP works. mixed member proportional - 7

11 MMP voted into an electorate seat, or they have to win five percent or more of the party vote. This is called crossing the threshold. One interesting curiosity in the system is that sometimes a party can also win more electorate seats than it is entitled to according to its share of the party vote. So, theoretically, a party might only win two percent of the nationwide party vote, but they might get five people elected in local electorates. When this happens overhang seats are created in parliament. 4 The overhang seats increase the total number of MPs in parliament by however many extra electorate MPs there are. This is to ensure that the regular list seats are allocated proportionally. These overhang seats are the reason we currently have 122 MPs in parliament instead of 120. Like with every voting system, there are all sorts of details that can change under MMP. For example, there could be more or fewer electorate seats or there could be different requirements for how parties win seats. At this referendum we are being asked about MMP as it stands. If a majority decide to keep MMP then parliament has said that the Electoral Commission will review MMP. 5 Depending on whether the government acts on the review s recommendations, the system could change in the future. The Electoral Commission says that there would be seven Maori seats under MMP. 6 The continuum below shows that MMP produces more proportional electoral outcomes. Electoral outcomes (Gallagher least squares index of disproportionality)* MMP More proportional (the share of seats each party has in parliament is close to its share of the vote) More majoritarian (the share of seats each party has in parliament is not as proportionate to its share of the vote) * Based on disporportionality scores calculated using the Gallagher least squares index provided by D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 8, table Evaluating MMP Representation MMP combines electorate representation and the representation of interests. Electorate MPs are retained, but parties representation in parliament is proportional which provides room for more interest group representation. Local electorate MPs can represent the interests of specific geographic communities and provide a human point of contact, 8 while list MPs can concentrate on representing non-local communities of interest, like minority and ethnic groups. 9 In terms of sheer numbers, there are now more female MPs and a greater diversity of MPs drawn from a variety of ethnic and cultural groups in parliament under MMP than there were under 8 - mixed member proportional

12 MMP FPP. 10 This increase in diversity is partly because MMP has made it easier for major political parties to stand a more diverse range of minority candidates within their lists, when these candidates might have struggled to win an electorate seat. 11 It is also because the proportional nature of MMP has made it easier for the minor parties which represent different interest groups to be elected. New Zealand s experience concurs with other evidence that shows party list systems enhance interest group representation. 12 Overall, MMP provides well for both forms of representation, as both electorate and list MPs have represented local communities and communities of interests. However, because government is decided by the party vote rather than the electorate vote, MMP emphasises parties as the vehicle for representation, rather than electorates. Effects on parliament and government MMP has had both positive and negative effects on the quality of New Zealand s democracy. Against predictions, it has led to increasingly stable governments. After a period of initial instability in the mid-1990s, multi-party governance agreements have been created that have enabled minor parties to maintain their points of difference in their core policy areas while having a say in government. 13 They have also helped various governments to pass legislation with relative ease. 14 However, a serious problem with multi-party parliaments under MMP is that minor parties have an incentive to prioritise the good of particular interests and minority groups in order to be elected. 15 In this situation, it is more difficult for parties to work together to produce law that is in everyone s best interests because the focus can easily be on what is best for the particular group of people that each party represents. 16 This can easily take us into the murky ground of identity politics where groups compete to promote their particular interests and ideas. Another question-mark over MMP is to do with accountability. Those who originally supported MMP argued that it would produce governments that were more accountable to voters. 17 But voters do not vote directly for one political party to form a government under MMP; they vote for a range of parties which bargain after election day. 18 The indirect election of governments means that sometimes parties that were part of a government which was voted out can stay in power as part of a newly formed government. This is a threat to direct accountability. Another issue is that list MPs can owe more loyalty to their parties than to voters so that they can receive a high list ranking. 19 Under MMP we cannot conceive of the opposition as a collective entity that is entirely separate from the government. Non-governing parties can do deals with the government to get their legislative proposals on the agenda, and only oppose the government on certain issues. 20 Summary: MMP MMP provides well for electorate representation and the representation of interests, and can provide for reasonably stable government. 21 The strength of MMP is the flip-side of its drawbacks. It enables more parties to be elected to parliament, which is great for the breadth of representation, but it also gives parties a lot of power. It can also create bargaining instead of debate among parties, and a weakened accountability of the government to voters. It can also encourage interest groups to act in unhelpful ways. mixed member proportional - 9

13 10 - first-past-the-post

14 FPP first-past-the-post How does FPP work? FPP is the voting system that New Zealand used up until Under FPP, voting is very simple. Everyone casts one vote for one candidate from their local electorate and parliament is made up entirely of those electorate MPs. All a candidate has to do to be elected under FPP is to win more votes than any other candidate in their electorate. 1 Voters put a mark beside the name of their preferred candidate on the ballot paper. The votes can be counted quickly and by the end of election night it is usually certain which candidate has won the most votes in each electorate and which party can form a government. 2 Political parties win seats in parliament by their candiates winning electorate seats. The party that wins the most electorate seats has the mandate to form a government. Based on current census data, the Electoral Commission estimates that there would be twelve Maori seats under FPP. 3 The illustration opposite shows the key features of how FPP works. first-past-the-post - 11

15 The continuum below shows that FPP produces more majoritarian electoral outcomes. FPP Electoral outcomes (Gallagher least squares index of disproportionality)* FPP More proportional (the share of seats each party has in parliament is close to its share of the vote) More majoritarian (the share of seats each party has in parliament is not as proportionate to its share of the vote) * Based on disporportionality scores calculated using the Gallagher least squares index provided by D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 8, table Evaluating FPP Representation FPP originates from a time before political parties became a common way of organising political representation. Local MPs were responsible for representing the interests of everyone in a geographic community. 5 Historically, under FPP in New Zealand, there was a close relationship between MPs and their electorates. MPs returned to their electorates from parliament to meet with and live among their constituents. 6 The idea of electorate representation was that local MPs ought to exercise their judgement according to their conscience, with one eye to local interests and another to national interests. 7 If the situation warranted, national interests were put first. Electorate representation under FPP and other similar systems encourages MPs and candidates to form a deep connection with their local electorate. However, a drawback of electorate representation under FPP (and other systems which elect just one MP from each electorate), is that the system makes it harder for minority voices to be heard. 8 This is usually because even if minor parties which often represent minorities and specific interest groups win a high percentage of the vote from across the country they find it hard to win electorate seats because their share of the vote is not concentrated in one place. 9 Interest group representation under FPP mainly relies on the major parties representing a broad cross-section of interests in society, but there is not much incentive for them to do so. 10 The major parties tend to represent the interests of the average, median voter at the centre of the political spectrum. 11 It is difficult for minority interests to enjoy formal avenues of parliamentary representation under FPP. The quality of the representation under FPP is therefore lop-sided. Whether we get a good balance depends on whether major parties choose to stand a diverse range of candidates in local electorates, whether local MPs and major parties choose to act on behalf of different communities of interest, 12 and whether major parties pay heed to the good of everyone whom they represent not just those who voted for them first-past-the-post

16 Effects on parliament and government FPP is biased towards candidates standing for major parties, because minor parties supporters are likely to be spread nation-wide and not concentrated in one electorate. This means the most common outcome under FPP is a two-party system in parliament, where one party wins the largest number of the seats in parliament and forms a government, and an opposition holds the second largest number of seats. Marginal seats are crucial under FPP, as those seats, which swing back and forth between parties at different elections, are the ones that can decide the overall outcome. Parties and candidates have an incentive to focus their campaigns on these seats, and they end up being driven by what is important to voters in those electorates. 13 One big criticism of FPP is that election results are often disproportionate at the national level that is, there is a legitimacy problem because the share of seats that parties win in parliament does not always closely match their share of the vote. 14 This happened in New Zealand under FPP. Between 1935 and 1993, the winning party won an average of 58 percent of the seats in parliament with only an average of 46 percent of the vote, whereas minor parties averaged 12 percent of the vote but only 0.1 percent of the seats. 15 The last FPP election in 1993 is another example of the distortion that can occur with FPP. The National Party was elected with 35 percent of the vote, while the Labour Party, the Alliance, and the New Zealand First Party together won 61 percent. 16 These sorts of disproportionate electoral outcomes and acute instances of wrong winner elections were two important reasons why many New Zealanders grew disillusioned with FPP and voted against it in the 1992 referendum. 17 It can be hard for the government to claim that it has a strong mandate if it does not have the support of a large proportion of voters. The stability of a strong two-party system can be a double-edged sword. On the one hand, electoral outcomes are predictable one political party is likely to form a government and to implement its policies, and it is clear who the opposition is. 18 On the other hand, strong singleparty majority governments can pass legislation with little interference, which can leave voters with uncertainty about how responsive and accountable the government is. In New Zealand, FPP was blamed for manufacturing dictatorial single-party governments. 19 FPP Summary: FPP FPP is simple to understand and it usually produces clear results. FPP delivers strong, stable, singleparty majority government most of the time, and there is usually no confusion about which party can form a government. It is easy for voters to dump a government and elect a new one since parties generally do not negotiate together to form a government. 20 But, as New Zealand s experience indicates, instances of highly disproportionate election results weakened the legitimacy of electoral outcomes and the Cabinet s tight control over legislation and parliament weakened the public s trust in government. 21 It can also be difficult for minorities to be represented, either because safe seats make it difficult to dislodge a popular candidate or because it is difficult for minority candidates to win enough concentrated support in one electorate. first-past-the-post - 13

17 14 - preferential voting

18 PV preferential voting How does PV work? Under PV, parliament is only made up of electorate MPs. Everyone gets one vote to elect one candidate from their electorate. To be elected, a candidate has to win an absolute majority of the vote that is, more than 50 percent. 1 Instead of just ticking a box, voters rank the candidates which they prefer. Voters put a 1 next to the name of their preferred candidate, followed by a 2 next to their second choice, a 3 next to their third choice, and so on. 2 The purpose of rank-ordering candidates is to enable votes to be transferred from one candidate to another. 3 If one candidate fails to win an absolute majority during the first round of counting votes then the preferences of the least popular candidate are transferred to the other candidates. This process continues until a candidate wins an absolute majority. 4 PV is a bit like a knock-out system in this respect. Depending on the PV system s design, in some cases voters are required to indicate a preference for every candidate on the ballot paper; while in others voters only have to indicate as many preferences as they want. It is not clear yet which design New Zealand would choose if PV were used in New Zealand. 5 Political parties win their seats by their candidates winning in electorates. Typically, PV makes it easy for one major party to form a government. The illustration opposite shows the key features of how PV works. preferential voting - 15

19 Based on current census data, the Electoral Commission estimates that there would be twelve Maori seats under PV. 6 The continuum below shows that PV produces more majoritarian electoral outcomes. Electoral outcomes (Gallagher least squares index of disproportionality)* PV PV More proportional (the share of seats each party has in parliament is close to its share of the vote) More majoritarian (the share of seats each party has in parliament is not as proportionate to its share of the vote) * Based on disporportionality scores calculated using the Gallagher least squares index provided by D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 8, table Evaluating PV Representation Because PV s basis is electorate representation, it provides a strong connection point between local communities and parliament. Yet, because voters can indicate a list of preferences, those voters who could otherwise be discouraged from voting for a minor party candidate can do so in the knowledge that their latter preferences will count, even if their first choice is eliminated. This gives minor parties a fighting chance, unlike FPP. Some argue, however, that this is one of PV s most serious shortcomings since only the subsequent preferences of those who have supported the least popular candidate(s) are taken into account when first preferences are redistributed. 8 It is still difficult for minor party candidates to challenge strong candidates in safe seats, yet PV is certainly an improvement on simple majoritarian voting systems because voters can select their preferred candidate without worrying about wasting their vote as happens with FPP. Some also claim that PV can provide well for the representation of interests in divided societies because it produces more moderate outcomes compared to FPP. 9 Generally, however, PV tends to be limited at providing for the representation of interests in parliament. Effects on parliament and government Overall, PV s performance against the criterion of legitimacy is mixed. Generally, the fact that voters can allocate preferences is an advantage. But sometimes PV delivers results that are different to what voters expect. For instance, if second and subsequent preferences are required to elect an MP, then sometimes the highest polling candidate on first preferences may not be the one that is elected preferential voting

20 Whether or not this is a good thing is contestable. Like with other single-member electorate systems, sometimes a party s share of seats is also disproportionate to its vote share. 11 For example, Australian PV elections have produced high levels of distortion between the share of the vote and the share of the seats that parties have won. 12 In three recent federal elections the winning party has won more seats but fewer votes than its rivals according to final preferences (1969, 1990 and 1998). In two more instances this happened on the basis of first preferences (1987 and 2010). 13 This distortion drops PV s legitimacy score. The election of fewer political parties under PV ought to produce fairly predictable electoral outcomes. PV tends to provide for a clear government and opposition since PV makes it difficult for minor parties to be elected. 14 If electoral outcomes are predictable then generally parliament and government ought to be stable; although some of the stability observed in Australian elections is due to compulsory voting and because there are two strong, well-organised parties. A final issue with PV concerns the party deals that can take place regarding voters preferences. In Australia, for example, both the Liberal-National Coalition and the Labor Party try to beat each other by swapping preferences to arrange the order in which preferences flow. 15 Because voting is compulsory, the parties put a lot of energy into producing how to vote cards in each electorate to maximise their own advantage over the opposition. This can unhelpfully affect the way that parties campaign as political tactics can end up dominating discussion of the issues. Voters may stop thinking about which candidates they would like to be their MP which is the point of PV and accept a party s choices. PV Summary: PV PV provides for strong electorate representation, through the election of local MPs, which usually leads to the election of single-party majority governments. That said, PV gives minor party candidates a fighting chance of winning a seat when second and subsequent preferences are used to help elect a candidate. However, it is still harder for minority candidates and parties to be represented in parliament under PV because it is not a proportional system. Further, PV can sometimes produce electoral outcomes that might not be considered entirely legitimate if the most popular candidate on first preferences does not win although this point is debatable. While PV would enable voters to more clearly express their preferences for certain candidates it could also introduce some new ways for parties and candidates to engineer electoral outcomes, as parties would advise supportive voters how to vote to give them the best advantage. 16 preferential voting - 17

21 18 - single transferable vote

22 single transferable vote STV How does STV work? STV may be familiar to voters who have participated in local body elections. All candidates are elected from local electorates. But instead of electing just one candidate, each electorate typically elects between three and seven candidates because the electorates are bigger but there are fewer of them. If New Zealand used STV there could be between 24 and 30 electorates. Based on current census data, the Electoral Commission estimates that there would probably be about four Maori electorates and twelve Maori seats altogether. 1 Everybody ranks their favourite candidates, like they do in PV. A candidate secures a seat when they reach a particular quota of votes. 2 If a candidate is elected at the first round of counting votes, any votes that they have in excess of the quota are transferred often proportionally according to second preferences, to elect the next candidate. 3 If seats are still unfilled after the surplus votes have been transferred, then the candidate with the lowest number of votes is excluded and their votes are transferred according to voters second preferences. 4 This twostep process continues until all the seats in the electorate are filled. If STV were introduced in New Zealand, voters would also be able to vote for a ranking chosen by a party, instead of ranking all the candidates by themselves. 5 This is called voting above-theline. 6 Voting above-the-line is meant to make voting simpler for people. But voters who choose to The illustration opposite shows the key features of how STV works. single transferable vote - 19

23 vote this way would have to accept a default ranking of candidates decided by their preferred party (which would be published prior to the election). The continuum below shows that STV produces more proportional electoral outcomes. Electoral outcomes (Gallagher least squares index of disproportionality)* STV STV More proportional (the share of seats each party has in parliament is close to its share of the vote) More majoritarian (the share of seats each party has in parliament is not as proportionate to its share of the vote) * Based on disporportionality scores calculated using the Gallagher least squares index provided by D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 8, table Evaluating STV Representation Because all MPs come from electorates, local representation is strong under STV. 8 Every voter is represented by more than one MP. Theoretically, constituents can have access to a variety of MPs who can listen and act on their behalf, although given that electorates would have to increase in size it could be difficult for MPs to have much contact with all of the people in their electorate. For example, Christchurch or central Auckland might become single electorates electing five MPs each. On the other hand, electorate representation could be strengthened if some MPs served certain localised communities within their electorate to increase their chances of election. As the system remains untried nationally in New Zealand, it is difficult to know how this issue would play out. STV could help to increase the ethnic and social diversity of MPs in parliament as multi-member electorates would make it easier for minor parties to win seats. 9 Major parties might also choose to stand a wider cross-section of candidates who could represent different national interest or social groups. 10 Thus, STV could provide for a high level of representation of interests. Effects on parliament and government Sometimes STV can encourage candidates who represent the same party to compete against one another when they stand in the same electorate. 11 In Ireland, for instance, STV has contributed to a strong focus on local issues in elections as candidates tap into local political concerns to garner votes. However, national issues can be sidelined in local candidates election campaigns. 12 STV can produce coalition government because of how minor parties candidates stand a better chance of being elected. Strangely, one of the puzzles of how STV works in practice is that, in the countries where it is used, large numbers of minor parties have not been elected to parliament. 13 For 20 - single transferable vote

24 example, both Australia (at the federal senate level) and Malta use STV and have strong two-party systems. This would not necessarily be a negative outcome if STV were used in New Zealand, since a two-party system is likely to produce stable parliaments, governments which can more easily enact legislation, and a clearly defined opposition. Even though STV is not primarily designed to distribute seats in parliament proportionally, examination of STV election results worldwide has shown that it still produces proportional outcomes. 14 The degree of proportionality depends on such design factors as how low the electoral quota is and whether political parties stand a greater or lesser number of candidates than they can expect to get elected. 15 If STV were used in New Zealand the system would probably not produce disproportionate election results, and wrong winner elections could be avoided. Voting above-the-line however creates problematic issues for accountability and legitimacy. This is because voting above-the-line essentially turns STV into a party vote system. 16 By voting above-the-line, voters would not be choosing which candidates are elected they would be choosing to accept the parties choice of candidates and the order in which they would be elected. The latter point is important if parties make agreements to swap preferences to give an electoral advantage to a potential coalition partner. 17 In Australia, where voting above-the-line is used in federal senate and state STV elections, between 85 and 95 percent of voters have been found to use it. 18 The option is used by such a high proportion of voters because the alternative is the more time consuming and potentially laborious task of rank-ordering literally dozens of candidates on the ballot paper. 19 We do not know for certain whether New Zealand voters would prefer voting above-the-line to indicating their own preferences for candidates. 20 There is evidence that the majority of voters who voted in New Zealand District Health Board STV elections listed their preferences for candidates in alphabetical order. This behaviour might indicate that voters misunderstood that they had a single transferable vote and not seven individual votes, 21 or maybe that voters found voting under STV too time consuming. While the evidence from New Zealand local body elections is helpful for illustrating voter behaviour, it is still difficult to say definitely whether New Zealand would experience the same accountability and legitimacy issues as Australia has if STV were used for parliamentary elections because we cannot predict exactly how candidates, parties and voters would respond. STV Summary: STV STV is an attractive system in principle since it enables voters to indicate exactly which candidates they would like in multi-member electorates. STV enables voters to choose both between and within parties, meaning that parliament ought to reflect a wider diversity of opinions within society. 22 The use of multi-member electorates also means that electoral outcomes will be more proportional. The theoretical advantages of STV have to be weighed carefully against the practical issues with using it and the way voters tend to interact with this relatively complex system. For example, it could undermine the cohesiveness of political parties as candidates from the same party would compete against each other for election. The option of voting above-the-line can also give parties more control over which candidates are elected and in which order. In this case, many voters would not actually end up individually choosing their local MPs. In short, the advantages offered by STV could be eroded by measures to make it easier for voters to understand and use. single transferable vote - 21

25 22 - supplementary member

26 supplementary member SM How does SM work? SM is from the same family of voting systems as MMP. It is also a mixed system that tries to combine electorate representation with some proportionality. Under SM, like MMP, everyone has two votes one for a candidate from their local electorate and one for a party. However, under SM the party vote is not as important as under MMP because it is not used to allocate all of the seats in parliament proportionally. In New Zealand, it would be used to determine the allocation of just 30 of the 120 seats to candidates drawn from party lists that parties would decide and publish before the election. 1 The other 90 seats would go to the candidates who win the most votes in each electorate. 2 This is the key difference between SM and MMP. The electorate vote and the party vote are independent of each other. They work alongside each other, but are not fused as they are under MMP. 3 The election results produced by SM are more majoritarian. 4 If New Zealand chooses SM, then the way we vote would look similar to how we currently do, even though the results and the outcome of elections would change. The electorate MPs would be chosen in the same way as under MMP, while the party list MPs would be elected by everyone ticking a box for a party as we do now. 5 Each party s total number of seats in parliament would be determinedby its combined number of electorate seats and list seats. The illustration opposite shows the key features of how SM works. supplementary member - 23

27 Based on current census data, the Electoral Commission estimates that there would be nine Maori seats under SM. 6 The continuum below shows that SM produces more majoritarian outcomes. Electoral outcomes (Gallagher least squares index of disproportionality)* SM More proportional (the share of seats each party has in parliament is close to its share of the vote) More majoritarian (the share of seats each party has in parliament is not as proportionate to its share of the vote) * Based on disporportionality scores calculated using the Gallagher least squares index provided by D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 8, table SM Evaluating SM Representation SM would increase the number of local electorate MPs from the current number of 70 to MPs and candidates would have an incentive to represent their electorate s interests as well as their parties. 9 The problems which New Zealand s list MPs have encountered under MMP would still remain. The popular perception that they are faceless MPs who are selected by party bosses, 10 or the issue of split loyalty do they primarily represent a party or voters would not be fixed. However, because there would be more electorate MPs under SM, some of the punch would be taken out of those issues. Because three-quarters of the seats under SM would be electorate seats, the electorate seats will become more important to how parties and candidates campaign. Some evidence suggests that when political parties stand candidates in local electorates under mixed systems it helps them to boost their share of the list vote. 11 Hence, the prospect of receiving a boost in the list vote however great or small might encourage more party competition in local electorates under SM than under FPP, for example. 12 On the other hand, parties might choose to coordinate their electorate nominations as part of their electoral strategy, which could reduce the number of candidates contesting a seat. It is hard to predict exactly what would happen, but it is thought that the latter effect could be stronger in mixed systems in which electorate elections are more important like SM. 13 There also might be a stronger tendency for parties to coordinate their campaigns given New Zealand s history of voter choice being influenced strongly by parties and that candidates only seem to have a limited effect on the parties party vote because of the high instances of split voting. 14 SM also differs from pure single-member electorate systems like FPP because it provides specifically for the representation of interests through the list system. Like the MMP party list system, under SM party lists would encourage the major parties to put candidates on their list who do not 24 - supplementary member

28 necessarily have broad enough appeal to win an electorate seat, but who could represent minority interest groups. 15 The 30 list seats would make it possible for minor parties to win seats in parliament. The effect would be more muted than under MMP, since there would be fewer list seats, and because the party vote would only determine the share of the 30 list seats. 16 Like MMP, SM allows for elements of both electorate representation and representation of interests, but it is ultimately weighted towards electorate representation because the party vote does not determine the allocation of all the seats in parliament. Effects on parliament and government If New Zealand adopted SM, the outcome of elections would usually be more majoritarian because of the smaller number of list MPs that would be elected compared to electorate MPs. 17 New Zealand research has shown that SM would not have enough of an impact on the shape of parliament to make it highly proportional. 18 Single-party majority or majority coalition governments would be the most likely ones to form. The same legitimacy problems that exist with single member electorate systems would exist with SM, although to a slightly lesser extent. For example, marginal seats would be important for election results, meaning that some local areas might get prioritised over others. If a coalition government formed, it would be more difficult for voters to hold it directly to account than a singleparty majority government. 19 While there are potential legitimacy and accountability issues with SM, it can encourage an effective opposition. Being a more majoritarian system, it can be expected over time not to produce highly fragmented parliaments that have, say, more than five parties. 20 If SM were used in New Zealand, initially there would probably be about the same number of parties represented in parliament as there are now under MMP, due to the number of minor parties currently represented which hold one electorate seat. 21 But over time the smaller number of list seats would reduce minor parties number of MPs and it could be harder for minor parties to win electorate seats. Generally we could expect parliament would operate more like it did under FPP, with a clear government and opposition, and with the government perhaps supported by a handful of minor parties. SM Summary: SM In trying to blend two styles of voting system, SM has some of the benefits and some of the drawbacks of both. It is neither a completely proportional system, nor does it guarantee that one party will win a large enough majority to be able to govern alone. In terms of representation, SM has the potential to achieve a good balance between national and local representation of interests. 22 Electorate representation would be strong, creating good ties between parliament and voters, but a quarter of parliament would also be made up of list MPs who tend to be able to represent minority interest groups well. Because there would be more electorate MPs under SM than under MMP the major parties would benefit, but there is also a chance coalitions would be needed to form a government and that minor parties would have more representation than they typically do under single-member electorate systems, like FPP. supplementary member - 25

29

30 conclusion New Zealand should have a voting system that would enable our MPs and the government to have the freedom to lead and to make decisions in the best interests of all New Zealanders, but it should also require them to listen and be responsive to the interests of New Zealanders and their local communities. In November, we will have a chance to vote on which system we think could do this best. When voting, there are a range of factors to consider, including: Representation electorate representation; representation of interests; Effects on parliament and government accountability; legitimacy; stable government; effective government; and opposition and oversight. conclusion - 27

31 By looking at these criteria, we have been able to think about the implications of the different voting systems. We found that more proportional systems (such as MMP and STV) can contribute to healthy representative democracy by improving the representation of interests without unduly inhibiting either parliament s role or the government s freedom to govern. However, we also found that negatively they can give political parties a lot of influence, as candidates have to rely on their party s favour to be selected and then elected. Given the likelihood of multi-party parliaments under these systems, there are greater incentives for political parties to do pre-election deals with each other so that the successful ones can more easily form a coalition government. By giving political parties greater influence, these systems can weaken the degree of direct connection between parliament, the government, and voters in local communities, as parties focus more on serving sectional interests to win votes throughout the country. Inter-party bargaining can also mean it is difficult to know what you re voting for when you choose a party. On the positive side, interest groups get a voice. On the downside, that voice can be too loud if an interest group becomes politically important, it can dominate policy and law-making. Interest group politics can make it difficult for MPs to govern with what is in everyone s best interests in mind. Voting systems which come from the majoritarian family can also provide for healthy representative democracy. These systems provide well for electorate representation, which enables the interests and views of local communities to be represented in parliament, and mean that voters can hold MPs and the government more directly to account for their performance. Voters can also be confident that electoral outcomes will be what they expect. But majoritarian systems are not so good at providing for the representation of interests, because the composition of MPs in parliament tends not to reflect so closely different interest groups share of the population. Majoritarian voting systems can also have other negative effects. For example, electoral outcomes are not as proportionate. Some voters votes can count substantially more than the votes of others, depending on, for instance, whether they live in a marginal or a safe seat. There is also the strong possibility of wrong winner electoral outcomes where a party wins lots of seats but does not win a large majority of the vote across the whole country. Mixed voting systems, like SM or MMP use elements of both proportional and majoritarian systems. They represent a trade-off between proportional representation and the direct accountability of MPs to voters. Mixed systems do not eliminate the problems in either style of system but rather smooth some of the rough edges of each and try to combine their benefits. Our evaluation has shown that there is no such thing as a perfect voting system there are aspects of each system which would be positive for representative democracy and aspects which would be negative. When thinking about which system represents the best balance of positive and negative features, voters will have to trade-off the features which they like by prioritising them according to the kind of representative democracy that they believe is in New Zealand s best interests. We hope that each and every voter will decide their preferred system after thinking carefully about what provides for effective representation, and with an awareness of each system s implications conclusion

32 endnotes kicking the tyres. choosing a voting system for new zealand 1 J. Key, National s Blueprint for Change. Speech to the Annual National Party Conference, Wellington, 3 August (2008), pledge ten. 2 S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System (London: The British Academy Policy Centre, 2010), 20; International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook (Stockholm: 2005), 4-5; M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds? (Oxford, UK; New York: Oxford University Press, 2001), 592; R.S. Katz, Democracy and Elections (New York: Oxford University Press, 1997), 4-6. evaluating voting systems 1 See, for example, works such as Pippa Norris Electoral Engineering. Voting rules and political behaviour (Cambridge: Cambridge University Press, 2004), This work highlights how different voting systems affect the electoral incentives of candidates and parties and how voters choose to exercise their vote depending on the options available to them. 2 The following sources discuss the concepts of majoritarian and proportional representation. S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System (London: The British Academy Policy Centre, 2010), 21; M.B. Vieira and D. Runciman, Representation (Cambridge, United Kingdom: Polity Press, 2008), 74-79; G. Newman and S. Bennett, Electoral Systems, Research Brief, 10 (Canberra: Department of Parliamentary Services, Parliament of Australia, 2006), 15; G.B. Powell, Elections as Instruments of Democracy: Majoritarian and proportional visions (New Haven, CT: Yale Univeristy Press, 2000), 7, 9; M. Pinto-Duschinsky, Send the Rascals Packing: Defects of Proportional Representation and the Virtues of the Westminster Model, Representation 36 (1999): S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, Previous Maxim Institute publications which discuss the value of parliament s deliberative role and representative government s role in protecting the common good include: R. Ekins, A Government for the People. The value of representative democracy, Guest Paper (Auckland: Maxim Institute, 2009); and J. Waldron, Parliamentary Recklessness: Why we need to legislate more carefully, Annual John Graham Lecture (Auckland: Maxim Institute, 2008). 5 It is important to note that while the research which we consider indicates general trends in terms of the electoral outcomes associated with each system, they are predictions, not an exact science. It is a best guess at what might happen with each system, especially where some systems are untried in New Zealand. S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, 15. Cf. E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, Regional and Federal Studies 17, no. 1 (2007): 72; L. Malpass and O.M. Hartwich, Superseding MMP: Real electoral reform for New Zealand, Policy Monograph, 109 (St Leonard s, NSW: Centre for Independent Studies (CIS), 2010), Cf. H.F. Pitkin, The Concept of Representation (Berkeley: University of California Press, 1967), We value the community link because we believe that the role of the state, which includes institutions like our parliament, is to serve and protect the good of its members. Elsewhere we have described what is good for the members of society as the common good. We think that what is good for people, this common good, is revealed and realised as it is lived out in community. This common good is indicated by the community s history, customs and traditions, ranging from its laws, music, art or literature, to the patterns of everyday life, such as work, study or play. The common good is also indicated by reasoned deliberation about what the intrinsic basic ends of a good life are, such as life, play, friendship, and knowledge. These goods are not mutually exclusive; each can be considered necessary for living a good life. A representative s job is to defend the common good. How we conceive of the common good is discussed in more detail in S. Thomas, Governing for the Good: What does it really mean? Tax Discussion Series, 1 (Auckland: Maxim Institute, 2008), Also see R. Ekins, A Government for the People. The value of representative democracy, 2. 8 M.B. Vieira and D. Runciman, Representation, See, for example, A. Lijphart, Patterns of Democracy: endnotes - 29

33 evaluating voting systems continued Goverment forms and performance in thirty-six countries (New Haven: Yale University Press, 1999). 10 J. Vowles, S.A. Banducci and J.A. Karp, Forecasting and Evaluating the Consequences of Electoral Change in New Zealand, Acta Politica 41 (2006): 270, citing K. Popper, On the Theory of Democracy, in All Life is Problemsolving (London: Routledge, 1987); and K. Popper, The Open Society and Its Enemies Revisited, The Economist 28, no. 25 (1988): 28. Also see G.B. Powell, Elections as Instruments of Democracy: Majoritarian and proportional visions, 9; M. Pinto-Duschinsky, Send the Rascals Packing: Defects of proportional representation and the virtues of the Westminster model, P. McCarvill, Devising an Electoral System for the 21st Century: The case for AMS (London: Institute for Public Policy Research (IPPR), 2010), New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy (Wellington: Government Printer, 1986), 12. The legitimacy of election results is therefore closely associated with the performance of our democratic institutions, such as parliament or list MPs. If our democratic institutions are believed to be failing to provide the kind of representative democracy that voters expect, then it can reflect badly on the voting system. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook (Stockholm: 2005), Cf. J. Boston, S. Church and T. Bale, The Impact of Proportional Representation on Government Effectiveness: The New Zealand experience, Australian Journal of Public Administration 62, no. 4 (2003): 10, citing M. Gallagher, M. Laver and P. Mair, Representative Government in Modern Europe (Boston: McGraw-Hill, 2001), 365, who found that of 14 European countries during , single-party majority governments lasted an average of 953 days, coalitions 638 days and minority governments 505 days (or less than one-and-a-half years). Boston, Church and Bale also cite W. Muller and K. Strøm, Coalition Governments in Western Europe (Oxford: Oxford University Press, 2000); A. Lijphart, Patterns of Democracy: Goverment forms and performance in thirty-six countries; K. Strøm, Minority Government and Majority Rule (New York: Cambridge University Press, 1990); and M. Taylor and V. Herman, Party Systems and Government Stability, American Political Science Review 65, no. 1 (1971): in support of the point that coalition or minority governments are not as durable as single-party majority governments. Also see J. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, (Wellington: Government Printer, 1986), J. Boston, S. Church and T. Bale, The Impact of Proportional Representation on Government Effectiveness: The New Zealand experience, 9, citing R. Weaver and B. Rockman, Assessing the Effects of Institutions, in Do Institutions Matter? Government capabilities in the United States and abroad, eds. R. Weaver and B. Rockman (Washinton D.C.: The Brookings Institute, 1993), 6; J. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, 12. Cf. S. Levine, N.S. Roberts and R. Salmond, A Wider View: MMP ten years on, in The Baubles of Office. The New Zealand General Election of 2005, eds. S. Levine and N. S. Roberts (Wellington: Victoria University Press, 2007), Cf. A. Kaiser, MMP, Minority Governments and Parliamentary Opposition, New Zealand Journal of Public and International Law 7, no. 1 (2009): 86. On the deliberative role of parliament, see: J. Waldron, Parliamentary Recklessness: Why we need to legislate more carefully, 13, 29, 33; and E. Burke, Speech to the Electors of Bristol on Being Elected. November 1774, in The Political Philosophy of Edmund Burke, ed. I. Hampsher Monk (London: Longman, 1987), Cf. Electoral Referendum Act 2010, sch (1). Mixed Member Proportional (MMP) 1 A more formal, if broad, definition of such mixed voting systems is that an electoral system is mixed if more than one formula is employed to distribute legislative seats. F. Ferrara, E.S. Herron and M. Nishikawa, Mixed Electoral Systems. Contamination and its consequences (New York: Palgrave MacMillan, 2005), 17. Two political scientists who have studied mixed voting systems, Matthew Shugart and Martin Wattenberg, explain why they think mixed systems represent the best of both worlds in the following way: Our general point is that MM systems permit myriad variations that can suit a specific political context, while still holding out the promise of providing the best of both worlds i.e. the best of both identifiable governing blocs and proportionality, and the best of both local accountability and cohesive and programmatic national parties. M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds? (Oxford, UK; New York: Oxford University Press, 2001), 1, This feature of MMP illustrates how the two votes are linked together and why the system is proportional. S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, (London: The British Academy Policy Centre, 2010), 87; F. Ferrara, E.S. Herron and M. Nishikawa, Mixed Electoral Systems. Contamination and its consequences, 19; M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds?, Once the votes are counted and it is known which parties have crossed either or both of the thresholds, the 120 seats in parliament are allocated to parties roughly in proportion to their vote share according to a mathematical formula called 30 - endnotes

34 Mixed Member Proportional (MMP) cont. the Sainte Laguë electoral formula. The system allows for a whole number of seats to be allocated to parties. The system also makes it harder for parties to win each additional seat. G. Newman and S. Bennett, Electoral Systems, Research Brief, 10 (Canberra: Department of Parliamentary Services, Parliament of Australia, 2006), Scotland and Wales use a form of MMP but their systems do not provide for overhang seats. 5 Electoral Referendum Act 2010, Part 4, 75(2). 6 Electoral Commission, The MMP Voting System. Mixed Member Proportional, Fact Sheet (Wellington: 2011), 1. 7 Also see disproportionality scores for 82 elections in 23 countries between 1979 and Source: M. Gallagher, Proportionality, Disproportionality and Electoral Systems, Electoral Studies 10, no. 1 (1991): The disproportionality index measures the difference between parties shares of the votes and their shares of the seats. M. Gallagher and P. Mitchell, The Politics of Electoral Systems, (New York: Oxford University Press, 2005), 602. Levine and Roberts also cite an average disproportionality figure of 2.9 for the five New Zealand MMP elections between 1996 and S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, New Zealand Journal of Public and International Law 7, no. 1 (2009): E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, Regional and Federal Studies 17, no. 1 (2007): 72, citing J.T. Anagnoson, Home Style in New Zealand, Legislative Studies Quarterly 8, no. 2 (1983): 160. Anagnoson theorised that the disciplined two-party system and the small size of constituencies (between 20,000 to 35,000 residents) contributed to local MPs devoting a large share of their time to electorate work. 9 List MPs who belong to minor parties, like ACT or the Green Party, for example, have a greater incentive to represent interest groups, national issues or their party s policy programme because contesting electorates is not a viable electoral strategy for them. E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, 87. New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy (Wellington: Government Printer, 1986), 53. Also see R. Miller and J. Vowles, Public Attitudes Towards MMP and Coalition Government, New Zealand Journal of Public and International Law 7, no. 1 (2009): 97; J. Vowles, S.A. Banducci and J.A. Karp, Forecasting and Evaluating the Consequences of Electoral Change in New Zealand, Acta Politica 41 (2006): 275; and P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand (Wellington: New Zealand Business Roundtable, 1992), MMP s supporters even argue that MMP is better than other electorate systems at providing strong local representation because voters do not have to be concerned about how their electorate vote will influence which government is formed. This means that they can choose who they think will be the best person for their electorate and the best party for the nation. Under FPP, PV, SM and STV, voters who want to elect a particular party to form a government have an incentive to support that party s local candidate(s). This motivation may outweigh their desire for strong electorate representation. Graeme Edgeler, Personal Communication, 6 May In the case of women, the list seats have contributed to the overall increase in women s representation. For example, in 1987, 14 percent of MPs were women. In 1996, the total share increased to 29 percent, with 45 percent of list MPs being women. By 2005, the total share of women MPs increased moderately to 33 percent and the list share had recovered to 1996 levels, at 46 percent. Forty-one women were elected to parliament at the 2008 election. Four women have resigned since then, so the total share of women in parliament is currently about the same as in 2005, at 32 percent. However, we should also bear in mind that women s representation was improving before MMP was introduced due to the Labour Party standing more women in winnable electorates. We should therefore not neglect to acknowledge how electorate seats and parties candidate selection strategies have contributed to the increase in women s representation. E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, 88. Cf. S. Levine, N.S. Roberts and R. Salmond, A Wider View: MMP ten years on, in The Baubles of Office. The New Zealand General Election of 2005, eds. S. Levine and N. S. Roberts (Wellington: Victoria University Press, 2007), Parliamentary Library, Final Results for the 2008 New Zealand General Election, Parliamentary Library Research Paper (Wellington: 2008), 9. J. Vowles, S.A. Banducci and J.A. Karp, Forecasting and Evaluating the Consequences of Electoral Change in New Zealand, 274, citing New Zealand Electoral Commission data. In terms of ethnic group representation, Maori representation benefited from the switch from FPP to MMP. In 1987, Maori made up about 5 percent of the MPs in parliament. At the first MMP election in 1996 their share increased to about 13 percent and the proportion of MPs claiming Maori descent more than doubled over the period from 1993 to Today, their share sits at 17 percent of the seats (that is, there were 20 Maori MPs elected at the 2008 election). After the 2005 election, Maori representation had improved to the point that for the first time Maori were now slightly over-represented in parliament in relation to their share of the population. Removing the seven Maori seats from the equation, Maori were represented almost exactly in proportion to their population share. The improvement is partly due to higher constituency representation, with the increase in the number of Maori seats from four to seven. However, as the figures illustrate, the proportional list vote has also made a big difference. An exception to the rule that the list system improved representation for ethnic minority groups is that it did not make a very big difference to the representation of Pacific Islanders. Their overall share of parliamentary seats did not endnotes - 31

35 Mixed Member Proportional (MMP) cont. increase between 1993 and In 2008, five Pacific Island MPs were elected to parliament, whereas none were in P.A. Joseph, The Maori Seats in Parliament, Te Oranga o te Iwi Maori: A study of Maori economic and social progress (Wellington: New Zealand Business Roundtable (NZBRT), 2008). Parliamentary Library, The Origins of the Maori Seats, Parliamentary Library Research Paper (Wellington: 2009), 29. S. Levine, N.S. Roberts, and R. Salmond, A Wider View: MMP ten years on, For instance, nearly all of the Asian MPs in parliament have been elected from party lists. E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, See J. Vowles, S.A. Banducci and J.A. Karp, Forecasting and Evaluating the Consequences of Electoral Change in New Zealand, J. Boston, Innovative Political Management: Multi-party governance in New Zealand, Policy Quarterly 5, no. 2 (2009): 56. A longer treatment of New Zealand s experience of multiparty governance may be found in J. Boston and D. Bullock, Experiments in Executive Government Under MMP in New Zealand: Contrasting approaches to multi-party governance, New Zealand Journal of Public and International Law 7, no. 1 (2009): S. Levine, N.S. Roberts, and R. Salmond, A Wider View: MMP ten years on, Political and legal analysts Penelope Brooke Cowen, Tyler Cowen and Alexander Tabarrok note that government can be influenced by special interest and minority groups under mixed voting systems like MMP. This is easier when those interest and minority groups can appeal to a minor party that can speak for them rather than having to work through the channels of a major party which will tend to reflect the views of the majority. As Cowen, Cowen and Tabarrok say, Partial list systems channel interest group pressures through multiparty coalitions instead. P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand, American economist and social scientist Mancur Olsen s work on collective action and group theory is also relevant to this discussion. Olsen argued that small, concentrated special interest groups can be more influential in politics than groups that are more numerous but less cohesive and organised together. Further, the government or in our case the major governing party under MMP has an incentive to listen to what minority interest groups, lobby groups and minority parties want and the policies which they demand in return for their support in government. Olsen argued that minorities policies are not necessarily good for the nation as a whole. See, for example, M. Olson, The Logic of Collective Action and the Theory of Groups (New York: Shocken Books, 1971); M. Olson, The Rise and Decline of Nations: Economic growth, stagflation and social rigidities (New Haven: Yale University Press, 1982), cited in P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand, 1.5. While this will not be the case for every interest or issue supported by minorities, sometimes their interests can be harmful because they prioritise some goods over others. For example, a redistributive policy that favours one group could, if it were expensive, be a big cost for every taxpayer and therefore a cost for the entire productive economy. 17 J. Vowles, S.A. Banducci and J.A. Karp, Forecasting and Evaluating the Consequences of Electoral Change in New Zealand, 270; J. Boston, S. Church and T. Bale, The Impact of Proportional Representation on Government Effectiveness: The New Zealand experience, Australian Journal of Public Administration 62, no. 4 (2003): P.A. Joseph, MMP and the Constitution, New Zealand Journal of Public and International Law 7, no. 1 (2009): Having said that, party selection rules might also be a factor in whether candidates are nominated to stand for parties in other voting systems. For example, MPs may not get re-selected for an FPP electorate contest if they challenge the party hierarchy. Further, in some cases under MMP, electorate MPs might stand a better chance of winning or holding an electorate seat because voters do not have to vote for their favourite party when they vote for their preferred candidate. Most of the other alternative systems that are being considered at the referendum require voters to vote for their party s preferred candidate to vote for their favourite party. We should also remember that even under New Zealand s old FPP voting system there was an equally, if not more, tightly-disciplined party system than there is under MMP. For example, voters choice of electoral candidate tended to be influenced by which parties voters supported than by strong preferences for a particular candidates (although the effect of candidate preferences have still been detected). J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, Electoral Studies 28 (2009): 42, citing J. Vowles et al., Toward s Consensus? The 1993 general election in New Zealand and the transition to proportional representation (Auckland: Auckland University Press, 1995), 161; D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, in Mixed- Member Electoral Systems. The best of both worlds? eds. M. S. Shugart and M. P. Wattenberg (Oxford: Oxford University Press, 2001), A. Kaiser, MMP, Minority Governments and Parliamentary Opposition, New Zealand Journal of Public and International Law 7, no. 1 (2009): 79, 90. P.A. Joseph, MMP and the Constitution, 121. Political analysts Claudia Geiringer, Polly Higbee and Elizabeth McLeay note that the confidence and supply agreements entered into by the minor parties which support the National Party have clauses that require the minor parties to offer procedural support, such as for urgency motions. C. Geiringer, P. Higbee and E. McLeay, Standing Orders Review 49th Parliament. Submission to Standing Orders Committee (Victoria University Wellington: The Urgency Project, 2011), 8. These clauses could make it more difficult for a minor party to oppose the major governing party. 21 E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, endnotes

36 First-Past-the-Post (FPP) 1 S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System (London: The British Academy Policy Centre, 2010), 37; International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook (Stockholm: 2005), P.A. Joseph, MMP and the Constitution, New Zealand Journal of Public and International Law 7, no. 1 (2009): Electoral Commission, The FPP Voting System. First Past the Post, Fact Sheet (Wellington: 2011), 1. 4 Also see disproportionality scores for 82 elections in 23 countries between 1979 and Source: M. Gallagher, Proportionality, Disproportionality and Electoral Systems, Electoral Studies 10, no. 1 (1991): The disproportionality index measures the difference between parties shares of the votes and their shares of the seats. M. Gallagher and P. Mitchell, The Politics of Electoral Systems (New York: Oxford University Press, 2005), 602. According to Gallagher, New Zealand s average score was 14.0 for three FPP elections between 1979 and B. Manin, The Principles of Representative Government (Cambridge: Cambridge University Press, 1997), 95-98, 193ff; H.F. Pitkin, The Concept of Representation (Berkeley: University of California Press, 1967), 168ff. 6 One 1983 study showed that MPs on average spent about 30 percent of their time on electorate work. Voters therefore came to see MPs local work officiating at public functions, dealing with housing and welfare issues, attending sporting events, and more as one of their main jobs. E. McLeay and J. Vowles, Redefining Constituency Representation: The roles of New Zealand MPs under MMP, Regional and Federal Studies 17, no. 1 (2007): 72, citing J.T. Anagnoson, Home Style in New Zealand, Legislative Studies Quarterly 8, no. 2 (1983): M.B. Vieira and D. Runciman, Representation (Cambridge, United Kingdom: Polity Press, 2008), x; B. Manin, The Principles of Representative Government, ; J. Coniff, Burke, Bristol and the Concept of Representation, Western Political Quarterly 30 (1977): 339; H.F. Pitkin, The Concept of Representation, 168, 175, Cf. D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, in Mixed-Member Electoral Systems. The best of both worlds? eds. M. S. Shugart and M. P. Wattenberg (Oxford: Oxford University Press, 2001), International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, L. Malpass and O.M. Hartwich, Superseding MMP: Real electoral reform for New Zealand, Policy Monograph, 109 (St Leonard s, NSW: Centre for Independent Studies (CIS), 2010), L. Malpass and O.M. Hartwich, Superseding MMP: Real electoral reform for New Zealand, 5. P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand (Wellington: New Zealand Business Roundtable, 1992), , 3.4, citing D. Mueller, Public Choice II (Cambridge University Press, 1989) and A. Downs, An Economic Theory of Democracy (New York: Harper and Row, 1957). 12 P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand, , 3.4. This point is made in the 1986 Royal Commission s discussion of the Supplementary Member (SM) voting system where they comment that SM would enable political parties to represent minority interest groups by standing a more diverse range of candidates. New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy (Wellington: Government Printer, 1986), Also see H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand (Palmerston North: The Dunmore Press, 1992), S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, 45-46; P. Norris, Electoral Engineering. Voting rules and political behaviour (Cambridge: Cambridge University Press, 2004), 12. A similar point is made that FPP can encourage ethnic cleavages, too. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, 15-16; H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, 19-23; New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, The result in each electorate seat is a snapshot of a particular community s preferences. Provided that a community elects the candidate which they prefer it may not matter very much if election results are disproportionate. 15 D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, Electoral Commission, General Elections Seats Won by Party (2011), resultsdata/fpp-seats-won.html (accessed 10 May 2011). 17 D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, On the ease and clarity of policy implementation, see: L. Malpass and O.M. Hartwich, Superseding MMP: Real electoral reform for New Zealand, 5; New Zealand Business Roundtable, Submission to the MMP Review Committee (Wellington: 2000), ii, 3-5, 8; P. Brook Cowen, T. Cowen and A. Tabarrok, An Analysis of Proposals for Constitutional Change in New Zealand, The high level of retrospective accountability of governments to voters, which FPP was supposed to provide at each election, appeared to have disappeared by the 1980s with the growing disproportionality of election results and acute wrong-winner elections. D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, Strictly speaking, under FPP voters can only dump that is directly vote out their local MP, not the government. This is because under FPP voters elect candidates to parliament and the government is formed by the party which has a majority of MPs. But, because voters have direct control over which endnotes - 33

37 First-Past-the-Post (FPP) cont. candidates are elected from each electorate, it is still true to say that it is easy for them to elect or dump a government at the election. Graeme Edgeler, Personal Communication, 6 May D. Denemark, Choosing MMP in New Zealand: Explaining the 1993 electoral reform, 70. Preferential Voting (PV) 1 D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences (Sydney: University of New South Wales Press, 2006), 3-4; International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook (Stockholm: 2005), H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand (Palmerston North: The Dunmore Press, 1992), STV Taskforce, Choosing Electoral Systems in Local Government. A resource document (Wellington: The Department of Internal Affairs, 2002), PV therefore creates certain electoral incentives for voters, candidates and political parties with implications for election results and the legitimacy of electoral outcomes. Political scientists David Farrell and Ian McAlister argue that PV ought to be cursorily categorised as a non-proportional voting system. However, their findings suggest that preferential voting produces different sorts of electoral behaviour among candidates and voters, such as plumping for one candidate and preference swapping among political parties. This means that election results, and sometimes electoral outcomes, can be different to what is expected of most single-member electorate majoritarian voting systems. On this basis, they argue that PV and its cousin STV ought to be thought of differently as examples of a family of ordinal voting systems. D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, Electoral Referendum Act 2010, sch 2(5). It should be noted that New Zealand already has a preferential system in STV, which operates the same way as PV, for mayoral elections. The form of STV used does not require voters to indicate a preference for all candidates. Given that ranking every candidate has not been a requirement at local body level, it is probable that voters would not have to rank every candidate if PV were used in general elections. 6 Electoral Commission, The PV Voting System. Preferential Voting, Fact Sheet (Wellington: 2011), 1. 7 Also see disproportionality scores calculated using the Gallagher least squares index for 82 elections in 23 countries between 1979 and Source: The disproportionality index measures the difference between parties shares of the votes and their shares of the seats. M. Gallagher and P. Mitchell, The Politics of Electoral Systems (New York: Oxford University Press, 2005), Lord Alexander, of Britain s Jenkins Commission on the Electoral System, found this scenario wholly illogical, as he thought that it made more sense for the preferences of those who supported the strongest candidates to be considered when subsequent preferences were allocated to candidates. McCarvill, citing Lord Alexander s dissenting position in the 1998 Jenkins Commission Report. P. McCarvill, Devising an Electoral System for the 21st Century: The case for AMS (London: Institute for Public Policy Research (IPPR), 2010), 8. 9 Political scientist Ben Reilly writes that, Political scientists have long theorized that the use of preferential election systems can help promote successful conflict management in divided societies. As it turns out, evidence supports this conclusion. B. Reilly, Electoral Systems for Divided Societies, Journal of Democracy 13, no. 2 (2002): Two divided societies which are nearby to New Zealand that have used PV are Papua New Guinea, which uses a limited preferential voting system based on the Alternative Vote, and Fiji, where the 1998 constitution introduced instant run-off voting. PV has worked with a mixture of success in Papua New Guinea, such as in terms of participation, but more attention is required to improve voters education levels about PV s implications. Fiji has experienced instances of preference swapping, with parties attempting to exploit the rules to gain an electoral advantage and it has in fact decided to change from its instant run-off PV system to a proportional representation system. R. McIlveen, The Alternative Vote - The system that no-one wants, Research Note (London: Policy Exchange, 2010), 1, 5; A. Ladley and J. Williams, Electoral Education in PNG: A survey of existing literature and reports (Wellington: Institute of Policy Studies, School of Government, Victoria University of Wellington, 2007), 57; J. Fraenkel, A Note on the Fiji Electoral System, in From Election to Coup in Fiji. The 2006 campaign and its aftermath, ed. Jon and Stewart Firth (Canberra: ANU E-Press; Asia Pacific Press, 2007), xxiv-xxviii. 10 New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy (Wellington: Government Printer, 1986), Cf. D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, 81; S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System (London: The British Academy Policy Centre, 2010), D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, 81. In Australia, compulsory voting and the prevailing use of full preferential voting where voters must mark a preference for every candidate bar one on the ballot paper undoubtedly contribute to the strength of the major parties and the difficulty that minor parties face in winning electorate seats (and therefore the high distortion between parties share of the vote 34 - endnotes

38 Preferential Voting (PV) continued and their share of parliamentary seats). Only New South Wales and Queensland use optional preferential voting along the lines of what New Zealand might use if PV were introduced. 13 R. McIlveen, The Alternative Vote - The system that no-one wants, 9. This outcome also occurred in the 1940, 1954 and 1961 elections. 14 New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, 82-83, and more generally chapter The 1986 Royal Commission on the Electoral System was also hesitant to recommend PV as a voting system for New Zealand. It thought that PV might represent some improvement over plurality [that is, FPP] in singlemember constituencies; however, we do not consider this improvement would be significant and do not regard it as the best alternative to our present system [FPP]. New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, 31. Further, Farrell and McAlister remark that Its [PV s] preferential nature has [electoral] consequences, and there is evidence that this is of increasing significance in recent elections. D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, 171. Single Transferable Vote (STV) 1 Electoral Commission, The STV Voting System. Single Transferable Vote, Fact Sheet (Wellington: 2011), 1. 2 The electoral quota is determined by dividing the total number of valid votes cast by the number of seats to be filled in each electorate, plus one and then adding a whole number, often one, or a fraction to the quotient. That is: the electoral quota = (valid votes/(available seats + 1)) + 1. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, 71, 76; H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, The votes are often distributed on a pro rata basis. D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, (Sydney: University of New South Wales Press, 2006), 63ff, 94ff; H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, (Palmerston North: The Dunmore Press, 1992), 49. In New Zealand local body elections, the Meek method is used to distribute the votes. This method ensures that the count, as far as possible, reflects the voting preferences of each voter and the number of wasted votes is kept to a minimum. STV Taskforce, Choosing Electoral Systems in Local Government. A resource document, (Wellington: The Department of Internal Affairs, 2002), 15, 35-39; B. Meek, A New Approach to the Single Transferable Vote, Voting Matters 1 (1994): H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, Electoral Referendum Act 2010, sch 2(6)(3). 6 H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, Also see disproportionality scores for 82 elections in 23 countries between 1979 and Source: M. Gallagher, Proportionality, Disproportionality and Electoral Systems, Electoral Studies 10, no. 1 (1991): The disproportionality index measures the difference between parties shares of the votes and their shares of the seats. M. Gallagher and P. Mitchell, The Politics of Electoral Systems (New York: Oxford University Press, 2005), New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy (Wellington: Government Printer, 1986), New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, S. Bowler and B. Grofman, Elections in Australia, Ireland and Malta Under the Single Transferable Vote. Reflections on an embedded institution (Ann Arbor: The University of Michigan Press, 2000), 9; New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, M. Marsh, Candidate-centred but Party Wrapped: Campaigning in Ireland under STV, in Elections in Australia, Ireland, Malta under the Single Transferable Vote. Reflections on an embedded institution, eds. S. Bowler and B. Grofman (Ann Arbor: University of Michigan Press, 2000), S. Bowler and B. Grofman, Elections in Australia, Ireland and Malta Under the Single Transferable Vote. Reflections on an embedded institution, Political scientist David Farrell cites a Gallagher least squares index of disproportionality score for of 5.4 for STV and 3.8 for MMP. D.M. Farrell, Electoral Systems: A comparative introduction (New York: Palgrave, 2001), xiii, 241, cited in D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, 8. The proportionality of STV in the case of Australian state elections is discussed in D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, 48. endnotes - 35

39 Single Transferable Vote (STV) cont. 16 Cf. D.M. Farrell and I. McAllister, The Australian Electoral System. Origins, variations and consequences, P. Brent, Time to Scrap the Ticket Vote for the Senate, 11 (Canberra: Democratic Audit of Australia; Political Science Programme, Research School of Social Sciences, Australia National University, 2004), P. Brent, Time to Scrap the Ticket Vote for the Senate, 1. Political scientists Ben Reilly and Michael Maley also provide percentage figures for ticket voting use in the Australian states for elections from 1984 to They show figures of between 85 and 95 percent in total from elections in the different Australian states. The share since 1990 has been between 90 and 95 percent, indicating ticket voting s high popularity. B. Reilly and M. Maley, The Single Transferable Vote and the Alternative Vote Compared, in Elections in Australia, Ireland and Malta Under the Single Transferable Vote. Reflections on an embedded institution, eds. S. Bowler and B. Grofman (Ann Arbor: The University of Michigan Press, 2000), B. Reilly and M. Maley, The Single Transferable Vote and the Alternative Vote Compared, New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, Evidence cited in S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, New Zealand Journal of Public and International Law 7, no. 1 (2009): S. Bowler and B. Grofman, Elections in Australia, Ireland and Malta Under the Single Transferable Vote. Reflections on an embedded institution, 1ff. Supplementary Member (SM) 1 The Electoral Referendum Act 2010 does not state either which formula would be used to distribute the 30 list seats among the political parties which cross the party vote threshold or what the party vote threshold would be if SM were used. The 1986 Royal Commission proposed that a modified version of the Sainte Laguë electoral formula ought to be used to distribute the list seats under SM so that it would be more difficult for the major parties to win their first list seat in parliament. This formula or the regular Sainte Laguë system that is used with MMP could be used to allocate the list seats under SM. Further, the 1986 Royal Commission did not specify a threshold for the list seats but noted that an effective threshold of about five percent of the party vote would exist in an election of about two million voters voting for a parliament of 120 MPs with 30 supplementary seats. Without a legal threshold the effective threshold will of course vary according to how many votes are cast at each election. A statutory five percent threshold party vote could be used with SM to solve this problem. Electoral Referendum Act 2010, sch 2(7); New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, (Wellington: Government Printer, 1986), 38-39, Electoral Referendum Act 2010, 10, sch 2(7)(2). 3 The terminology used to describe mixed voting systems like this varies from Supplementary Member, to Mixed-Member Majoritarian (MMM), to parallel systems (because the two votes operate side-by-side, much how like parallel lines do not touch). SM is often called Mixed-Member Majoritarian outside of New Zealand because under such a system it is likely that a single party would win an absolute majority of seats in parliament. S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System (London: The British Academy Policy Centre, 2010), ; International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook (Stockholm: 2005), International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, 104, Note, the Electoral Referendum Act 2010 does not say how the ballot paper would be designed that is, whether there would still be one ballot paper, just like with MMP, or whether there would be two separate ballot papers. Ballot paper design can have an important effect on electoral outcomes. As political scientist Jeffrey Karp indicates, A number of studies have found that the position of candidates on a ballot can influence electoral outcomes though the magnitude of these effects varies. Little is known, however, whether these effects would extend from voting for candidates to parties. For instance, see J.G.S. Koppel and J. Steen, The Effects of Ballot Position on Election Outcomes, Journal of Politics 66, no. 267 (2004): 281 and J.M. Miller and J.A. Krosnick, The Impact of Candidate Name Order on Election Outcomes, Public Opinion Quarterly 62 (1998): , cited in J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, Electoral Studies 28 (2009): 42. In New Zealand, a minor party s party vote in an MMP election can benefit from being associated with a local candidate s name if candidates names are printed side-byside with parties names on the ballot paper. This is important because if no candidate stands in an electorate then the party s name is printed in alphabetical order below the names of the parties which do stand candidates. Thus, minor parties that contest fewer electorates have an electoral disadvantage since their party name is printed further to the bottom of the ballot paper, where some voters might not read. J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, endnotes

40 Supplementary Member (SM) cont. 6 Electoral Commission, The SM Voting System. Supplementary Member, Fact Sheet (Wellington: 2011), 1. 7 Also see disproportionality scores for 82 elections in 23 countries between 1979 and Source: M. Gallagher, Proportionality, Disproportionality and Electoral Systems, Electoral Studies 10, no. 1 (1991): The disproportionality index measures the difference between parties shares of the votes and their shares of the seats. M. Gallagher and P. Mitchell, The Politics of Electoral Systems (New York: Oxford University Press, 2005), 602. Levine and Roberts also cite an average disproportionality figure of 9.5 for five New Zealand MMP elections between 1996 and 2008 re-run under SM rules. S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, New Zealand Journal of Public and International Law 7, no. 1 (2009): H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand (Palmerston North: The Dunmore Press, 1992), New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, R. Miller and J. Vowles, Public Attitudes Towards MMP and Coalition Government, New Zealand Journal of Public and International Law 7, no. 1 (2009): 97. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, 112; M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds? (Oxford, UK; New York: Oxford University Press, 2001), ; cf. New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, F. Ferrara, E.S. Herron and M. Nishikawa, Mixed Electoral Systems. Contamination and its consequences (New York: Palgrave MacMillan, 2005), 48, 63, ; J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, In this respect, the style of politics and campaigning at the electorate level might be more similar to the situation under MMP. The assumption is that there is contamination between the two electoral tiers; that is, parties which stand electorate candidates will receive a boost to their party vote even if they do not win an electorate seat. J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, 42, F. Ferrara, E.S. Herron and M. Nishikawa, Mixed Electoral Systems. Contamination and its consequences, 140. Some analysis produced by political scientist Jeffrey Karp nonetheless offers some clues. Using data from the 2005 and 2002 MMP elections, Karp finds that the impact of electorate candidates on parties list votes is quite small. According to Karp, parties appear to matter more in New Zealand politics even though districts are relatively small and MPs are likely to invest a great deal of effort in constituency service. Karp s findings indicate that the influence of electorate elections over the party vote election might not be as important in New Zealand under MMP, and in other mixed systems, as Ferrara and others have stated. J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, 49. However, we do not know for certain whether the result would be the same with SM. 14 Jeffrey Karp finds only modest evidence that in New Zealand electorate candidates have helped to increase their parties list vote. New Zealand s high incidence of split voting between 29 and 39 percent of voters have split their vote between the electorate and the list vote in MMP elections might explain some of the modest results. J.A. Karp, Candidate Effects and Spill-over in Mixed Systems: Evidence from New Zealand, New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, 40. Cf. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, 112; H. Catt, P. Harris and N.S. Roberts, Voter s Choice. Electoral change in New Zealand, New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, S. Hix, R. Johnston and A. Cummine, Choosing an Electoral System, 97; M.S. Shugart and M.P. Wattenberg, Mixedmember Electoral Systems: The best of both worlds? 13ff. International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, New Zealand political scientists Stephen Levine and Nigel Roberts re-ran New Zealand s five MMP election results, between 1996 and 2008, using SM rules. Their findings showed that at three of the elections a single-party majority government would have been elected and that overall the election results would have been three times more disproportionate than they were with MMP. S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, 135, S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, 145; New Zealand Royal Commission on the Electoral System and J.H. Wallace, Report of the Royal Commission on the Electoral System: Towards a better democracy, F. Ferrara, E.S. Herron and M. Nishikawa, Mixed Electoral Systems. Contamination and its consequences, ; International Institute for Democracy and Electoral Assistance (IDEA), Electoral System Design: The new international IDEA handbook, See tables 2 to 6 in S. Levine and N.S. Roberts, MMP and the Future: Political challenges and proposed reforms, Cf. M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds? M.S. Shugart and M.P. Wattenberg, Mixed-member Electoral Systems: The best of both worlds? endnotes - 37

41 First web edition published in September 2011 by Maxim Institute 49 Cape Horn Road, Hillsborough, Auckland 1041, New Zealand PO Box , Roskill South, Auckland 1445, New Zealand Ph (+64) Fax (+64) Copyright 2011 Maxim Institute ISBN This publication is copyright. Except for the purpose of fair review, no part may be stored or transmitted in any form or by any means, electronic or mechanical, including recording or storage in any information retrieval system, without permission in writing from the publisher. No reproduction may be made, whether by photocopying or by any other means, unless a license has been obtained from the publisher or its agent.

42 PO Box , Roskill South Auckland 1445, New Zealand 49 Cape Horn Road Hillsborough, Auckland t f