Voting in New Zealand local government elections: the need to encourage greater voter turnout

Within a global context, local government in New Zealand occupies an enviable position: it enjoys both a statutorily-defined ‘power of general competence’ and financial autonomy from central government. However, despite this, voter turnout rates in New Zealand local elections continue to fall as ever fewer New Zealanders engage in this fundamental act of civic engagement. This review article examines the decline in voting over the last four New Zealand local government elections (2010/13/16/19). It aims to do three things: plot the decline; identify and analyse the causes of this decline; and suggest ways in which the decline might be countered. The authors reach the conclusion that local government in New Zealand is at a crossroads – it will either be rejuvenated as a source of local democracy and prosper, or decline into an administrative arm of central government.


Introduction
Every three years New Zealanders participate in local government elections for 78 local authorities.
Within most of those local authorities, councillors are elected based on wards (subdivisions), orin a minority of areas -'at large'; that is, by the whole electorate. Mayors, however, are in every case directly elected by the whole electorate. Across New Zealand there is a decline in the number of citizens participating in local government elections. Turnout has decreased over the last four decades from a national average of 57% in 1989, to a low of 38.2% in 2019 (Beattie 2011;Local Government New Zealand 2019), raising issues of democratic legitimacy in local government (Bush 1995;Cox 2003;Gronlund and Setala 2007). This decline in voting is not reflected at the central government level, which shows voter turnout of 79.8% (Electoral Commission 2018) and recently 82.5% in 2020 (Electoral Commission 2020).
It has been argued by Asquith (2012) that local government in New Zealand occupies an enviable international position. At the time of its enactment, the country's Local Government Act (LGA) 2002 formally recognised the role of local government in New Zealand and empowered 'local bodies' (the term for local authorities) to pursue the 'four well-beings'. This is a broad power to act as they see fit within the law to enhance the well-being of their citizens economically, socially, environmentally and culturally. The pursuit of the well-beings (repealed in 2012 but reinstated in 2019) is equated with what has been described as a 'power of general competence' (Perriam 2002;McKinlay 2003;Shand 2019).
Alongside the power of general competence, New Zealand local bodies also enjoy unusually high levels of financial autonomy (Asquith 2012;Shand 2019). Unlike local authorities in jurisdictions such as England, where local governments have also been granted the power of general competence, and Scotland, where they have a power of general well-being, local bodies in New Zealand have also largely been freed from dependence on finance delegated from a higher tier of government. An English local authority, for example, is dependent on grants of various kinds from central government for in excess of 80% of its income; by contrast, a New Zealand local body will raise over 85% of its income locally through property taxes -both general and targeted 'rates' (Department of Internal Affairs 2015). Yet despite the importance of these local taxes, local body elections attract declining numbers of citizens and engagement in the democratic process is decreasing.
In this review the authors draw on extant literature to explain the influences on voter turnout in New Zealand. We begin with a review of conceptual frameworks concerning voting behaviour internationally and relevant features of New Zealand local government. We then plot the decline in turnout and analyse the causes, before identifying seven specific factors which discourage New Zealand citizens from voting. We conclude with suggestions to strengthen local democracy in New Zealand.

Conceptual frameworks in a New Zealand context
The New Zealand non-voter mirrors the international non-voter. According to Foster and Taylor (2019), they are more likely to be from non-European ethnic groups. In national elections, for example, those enrolled in Maori electorates were 14% less likely to vote (Electoral Commission 2015). General electorates with a high Maori and Pasifika population showed a 9% lower turnout than average, and New Zealanders of Asian descent had the lowest turnout per capita in the 2008 and 2011 elections. Nonvoters also tended to be younger, with 42% of 18-24-year-olds not voting in the 2011 election (Statistics New Zealand 2014). In 2017, non-voting of this age cohort decreased by 6.5%, preceded by decreases in 25-29-year-olds and 30-34-year-olds of 5.5% and 3.5% respectively (Electoral Commission 2018).
Alongside this, New Zealand non-voters were more likely to have low incomes and fewer assets, and to be unemployed and to live in rural areas. Foster and Taylor (2019) highlighted that many of these characteristics were intersectionaleg young Maori with a low standard of living, or living rurally, were less likely to vote (Electoral Commission 2015; Foster and Taylor 2019).
The international literature highlights a range of concepts that help to explain these observations and influence voter turnout at local elections. They include in particular: • 'rational choice' theorywhether elections are seen to matter • 'youth footprint' theory -the experience of young voters' first three elections • 'social group' theorythe demographic and social nature of electorates • 'firstand second-order elections'the perceived significance of the election within the national framework of democracy.
Rational choice theory (Riker and Ordeshook 1968) suggests that elections must matter in order for voters to participate. It focuses on the costs of participation, generally established by institutional features in the country or region; the perceived benefits, generally arising from the political context and the state of electoral party competition; and the probability of one's vote making a difference. Further, an individual's level of political engagement is determined by their feeling of 'political efficacy'ie whether they believe that individual political action has an impact on the political process (Campbell et al. 1954). The advocates of rational choice theory assert that the higher the degree of efficacy, the more likely it is that an individual will vote.
Youth footprint theory was advanced by Franklin (2004). He argued that a significant change in established democracies was the swelling of electorates in most Western countries from the late 1960s, as a result of the 'baby boomer' generation reaching voting age. Franklin also observed that newer cohorts were "… more responsive to changes in the character of elections, and the relative size of this group strongly affects turnout by changing the proportion of the electorate that responds to changes in the elections ' character" (p. 210). This theory has been supported by multi-country longitudinal data showing increasing cohorts of new voters as the electoral age lowered (Franklin 2004) and a tendency CJLG June 2021 82 for "incoming cohorts to retain the habit of higher or lower turnout gained in their first elections" (p. 60). Such a finding tends to support anecdotal evidence that establishing voting habits early is a successful strategy to increase turnout. The disenfranchisement of younger voters leading to a trend in declining turnout was attributed to a socialising effect at a time when they were establishing themselves in society, and not yet in a position to perceive political benefits. This age effect has influenced voting patterns over a long period of time, and over 20-30 years could account for the trend of declining turnout (Ellis 2006).
In New Zealand's case, the findings of Vowles (2010), showing turnout among youth has not declined between successive generations, cast some doubt on the applicability of 'youth footprint' theory in The concept of first-and second-order elections was advanced by Reif and Schmitt (1980) in their analysis of European elections. First-order elections are those offering voters a critical choice of who should govern a country. Second-order elections are seen as less important, because, while open to influence by national party politics, they determine the outcome for lesser offices such as the roles of regional and local officials. Reif and Schmitt suggest that lower levels of voter turnout in municipal elections are explained by less being at stake. If we accept that the concept of first-and second-order elections advanced by Reif and Schmitt (1980) is a valid one, it seems logical that the rational choice theory and its corollary concern of political efficacy (Franklin 2004;Vowles 2010), will heavily influence local government election turnouts. highlighted elector apathy and lack of effort as reasons for non-voting. They note as frequently cited factors electors claiming they are too busy or that they ran out of time to vote. Institutional arrangements (see footnote 1), the nature and scale of elections and the resulting complexity were also cited as barriers to voting. In addition, the timing of elections with respect to school holidays and the three-week election period (which was seen as too long), were identified as problematic (Department of Internal Affairs 2010).

CJLG June 2021 83
The next section of this paper will discuss the relationship between central party politics at the national level and in local government in New Zealand's main centres.

Significant features of New Zealand local government
Until 1989, New Zealand had around 850 multipurpose regional and local authorities (Derby 2016).
Local government reforms in 1989 significantly reduced the number of local bodies to 86. By 2021 further reforms had reduced the number to 78. However, it is the 61 local authorities (district and city councils) and six unitary authorities (councils holding both local and regional powers) that are of interest here. The direct election of mayors in those bodies has been observed to have a significant impact on the media profile of an electoral cycle, and hence voter turnout, compared with the 11 regional councils which appoint a chairperson from amongst the elected councillors.
A common rationale for voting in local elections is that, as the tier of government closest to the populace, local government has great meaning and relevance to the citizenry (Wilson and Game 2012; Drage and Cheyne 2016). Local government development in New Zealand has been described as one of 'stakeholder democracy', in that it was landowners who were first granted the vote in local government (Hill 2003). According to Hill (2003), the long-running impression created among the voting public was one of local government in New Zealand being for the 'protection of the ratepayers' (ie those people directly paying property taxes). This impression was reinforced for decades by continued reference to 'ratepayers' in speeches by local government politicians in New Zealand, notwithstanding a law change in the 1890s that opened up voting rights to all residents of a local government district. Even in 1950, the prime minister of the day, Sidney Holland, proclaimed in parliament that New Zealand was a property-owning democracy. Hence the enduring view subsists that property owners are the primary stakeholders in local government in New Zealand.
Another relevant consideration is the high ratio of citizens to elected representatives (councillors and mayors) in local government. Put in an international context, New Zealand presents as a significant statistical outlier, and the high ratio increases the distance between voters and their representatives. Table 1 below illustrates this position.

The role of party politics
Party politics undoubtedly has a socio-psychological effect (Greene 2004;Hornung et al. 2019) on voter behaviour in local government. Bush (1995) commented upon the long-standing popular view that "politics contaminates the pure air of local government" (p. 260), and the effect it has had on many voters in New Zealand over the decades. In New Zealand, citizens have since the 1980s flocked to the banner 'Keep Politics out of Local Government', an irrational slogan that nonetheless expresses repugnance for the idea of central government party politics influencing local governance (Bush 1980, p. 178). Bush (1995) sees the irrationality of this view, as he describes politics as "the process by which resources are allocated… who gets what, when, how and why" (p. 260). It follows that politics, in this sense, is as prevalent in local government as it is in central government; but, as Bush argues, the object of disapproval in New Zealand has been party politics of a polemic nature as seen in central government elections. However, Bush offers no rationale for why political parties have been unwanted in local government. In fact, in recent times party affiliation at the local level has been on the rise in New Zealand's two largest cities (Webster et al. 2019a(Webster et al. , 2019b. Between 2007 and 2016, the number of candidates affiliating overtly to a political party doubled in Auckland (from 15% to 38%) and Christchurch (11% to 26%) (Webster et al. 2019b).
Party politics at the local level simplifies the voting process by assisting electors to sort out policy pledges (Bush 1980), and he notes that "voters appreciate the cues given by candidate affiliations" (Bush 1995, p. 264). It seems clear that political parties can offer a broader choice of candidates and, if they are elected in sufficient numbers, provide some accountability on the delivery of policy pledges.
It is also well acknowledged in New Zealand and internationally that local politics can serve as a 'stepping-stone' for candidates seeking experience in grass-roots politics and campaigning before contesting a central government seat (Copus 1997;Copus and Erlingsson 2012;Webster et al. 2019b

Patterns of declining voter participation
The recent decline in voter participation in local body elections is summarised in Table 2 below, which also highlights the number of uncontested mayoral elections out of the 67 eligible local authorities.  Table 2 may be seen to cast some doubt on this proposition, it is supported by other evidence on the importance of contestability, as discussed below.
Secondly, the majority of New Zealand local elections since 1989 have been conducted by postal voting.
Twenty-five years ago, Bush (1995) argued the past decline in voter participation levels was due to the requirement to visit a polling station to cast a vote, with the adoption of postal voting being seen as the solution. It was hoped the downward trend could be reversed. However, Table 2  trend is reported by Reid (2016) and clearly demonstrated by Local Government New Zealand's data from the last four electoral rounds, whereby there is a marked difference of at least 10% between voter engagement in rural authorities and that in bigger metropolitan councils. This is shown in Table 3.
CJLG June 2021 86 Source: Local Government New Zealand (2019) There is also evidence that the ratio of councillors to number of voters, and the total population and geographical size of constituencies, are important factors. Drage and Webster (2016) observe that the New Zealand Department of Internal Affairs statistics have consistently shown an increase in voter participation where councillors represent fewer citizens. This picture is amply illustrated by Table 4, which shows 2019 data from large and small councils.  Reid (2016) notes that councils with smaller representation ratios tend to have higher levels of voter turnout -something which is reflected in Local Government New Zealand's own statistics. These institutional facets of New Zealand local government support the rational choice argument discussed earlier, namely that elections must be seen as relevant and meaningful to voters' daily lives in order for them to turn outa point reinforced by the data contained in Table 4.

Opportunities for improvement
Acknowledging lack of council and candidate information; (6) the visibility of incumbent councillors; and (7) the cost of participating in local elections. The authors suggest that, taken together, these may significantly impede and discourage citizen engagement and participation in local elections in New Zealand.

Lack of contestability
Elections are contests and are a fundamental expression of local democracy (Beattie 2011). No one should be 'elected' unopposed. However, in New Zealand in 2016 nine incumbent mayors were elected unopposed (see Table 2)including one who had not had a challenger since first being elected in 2004 (ie over a cycle of four triennial elections). This speaks to the rational choice argument that elections need to matter to mobilise voters: if a mayor is popularly elected and recognised as the leader and spokesperson for an area, and is a focus of media and campaign activity, then a mayoral election is likely to attract voter participation (Department of Internal Affairs 2006). While the New Zealand data presented in Table 2 does raise questions about this reasoning, the theory is supported by a large body of international research which has found that failure to have a mayoral election has a negative influence

Political affiliation
The second influencing factor identified by this paper relates to candidates' reticence to declare a 'political' affiliation, a significant issue affecting local body elections in New Zealand's largest centres (Webster et al. 2019b). The country has an image of being at the forefront of environmental recycling.
Equally apparent is the constant 'recycling' back and forth of politicians between central and local government. While there are a small number of local politicians who make the 'step up' into parliament, there have been many former central government politicians fashioning a second political career in local government (Hehar 2016; Hurley 2019). While in parliament, these candidates have all been representatives of a major political party, but in local elections they claim to be 'independent', even though it appears that few, if any, relinquish their political party membership. This undermines perceptions of the transparency and credibility of local government. The idea of being able to shed one's political values and beliefs simply because of the desire to be seen as independent, is something we, the authors, find hard to accept.

Multiple electoral systems
A third factor is the use of multiple electoral systems. The present research suggests central government has inadvertently complicated the electoral system, with the unintended outcome of alienating voters.
The whole process of elections in New Zealand is complicated and confusingwith electors having to be familiar with, understand, and participate in three different systems simultaneously ( (Hayward 2016). There has, however, been a minor reversal recently, with 11 councils using STV in 2019, and in 2020 three additional councils voting to adopt STV for the 2022 local elections, bringing the total number of district and city councils using this method to 14 out of 67.
To coincide with the local body elections, every three years New Zealanders also elect members to their district health boards (DHBs). These elections all use the STV system. To be eligible to stand in a local government election, a candidate must meet three core criteria: (i) be a New Zealand citizen; (ii) secure nominations from two people on the electoral roll for the local authority in which they want to stand; and (iii) pay a deposit of NZ$200. This low threshold for entry has the effect of encouraging many candidates unlikely to be successful to stand. For example, the 2010 election for Auckland mayor attracted 21 candidates, of whom only two had a serious chance of victory something acknowledged by the prime minister at the time, who noted he could "work with either of the two candidates" (Asquith 2015). Meanwhile in Palmerston North, it appears that one recent candidate could have been barred from holding mayoral office due to long-standing mental health issues 4 (Galuzka 2018). The lack of formal political party involvement in local government elections in New Zealand exacerbates this issue by, in effect, removing a 'quality-control' mechanism in candidate selection such as is found, for instance, in local government elections in the UK (Widdecombe 1986;Bottom and Reiser 2014) and Europe (Aars and Ringkjobt 2005;Bottom and Copus 2011;Copus and Erlingsson 2012).

Council and candidate information
Fifth, there is a general lack of voter information, both about the councils themselves and about those seeking election to council. Although each council's chief executive is required to produce a preelection report, there is scant public knowledge of either the existence of the report or its content. Often the only way a voter has knowledge of a candidate is if they read the statutorily mandated 'candidate information booklets', which describe in 150 words each candidate's 'platform' and political affiliation.
As noted earlier, in New Zealand many candidates describe themselves as an 'independent' or simply ignore identifying their status (Webster et al. 2019a), regardless of previous political affiliation. During the 2016 election campaign the then prime minister even observed that the value to be gained from CJLG June 2021 90 reading the candidate profile booklet was so limited that he never read it, simply voting for those candidates with whom he was familiar (New Zealand Politics Daily 2016).
Finally on this point of candidate 'familiarity', considerable evidence suggests that the way people get to 'know' a candidate in local body elections is if he or she is a well-known figure or has 'been in the job' for a number of years (Bush 1980(Bush , 1981Drage 2008). Once again, with high councillor to population ratios, the lack of political party involvement in local elections is all the more intriguing given the finding of Webster et al. (2019b;2019a) that clear party affiliation can have a significant beneficial impact on a candidate's chances of electoral success.

Visibility of incumbent councillors
The sixth factor identified relates to the visibility or otherwise of incumbent councillors. Early work by Bush (1981) highlighted the high success rate amongst incumbent candidates in New Zealand. This pattern was further replicated in the 2008 local elections in the state of Victoria, Australia (Economou 2010) and in local elections in Canada (Sancton 2018). The low turnout of voters in New Zealand local body elections may be attributed in part to the relative invisibility of a councillor once they have been elected. In New Zealand there is no requirement for the elected councillors to hold 'clinics' or 'surgeries', and unlike the position in relation to their member of parliament, a voter may not see an office in their locality for the local councillor/s. Thus the elected members of councils in New Zealand can become virtually invisible to anyone not actively involved in the business of local government. Drage (2008) illustrates this in some detail, arguing that councillors have dual rolesas an elected representative, providing local voice; and alongside this a governance role, determining plans and policies and making some operational decisions. It is worthy of note that since the election of the fourth Labour government in 1984, New Zealand has embraced a neo-liberal 'new public management' model much more enthusiastically than any other country (Kelsey 1995;Martin et al. 1996;Pollitt 2000;Scott 2001;Asquith 2016). As a result, the governance role of a councillor, as opposed to their representative role, is dominant, and is similar to that of a non-executive director in a private sector business (Asquith 2012). As Drage (2008) points out, councillors who challenge this by arguing for a more balanced role, giving greater emphasis to their representative responsibilities, tend to find themselves 'frozen out'.

Cost of participation
The authors' seventh and final point relates to the cost for candidates of participating in local elections.
While the bar for standing for election is low, the ability to run a credible campaign which has impact is a different proposition altogether. The lack of overt party-political activity means that there are no nationally organised and funded election campaigns for local elections. The net effect of this is that candidates are either self-reliant for campaign funding or dependent upon a group of supporters or wealthy benefactors. This applies particularly to mayoral elections in the larger cities. There are nationally-set caps on the amount a candidate may spend in Auckland, and to a lesser extent the other CJLG June 2021 91 major urban areas, but candidates' expenses can still be considerable (Webster et al. 2019a (Niall 2016). It was also suggested by Orsman (2016) that Goff had raised NZ$250,000 in a single fundraising dinner for his 2016 campaign.
As such, all except those able to raise substantial funds (in the New Zealand local government context), are excluded from standing and coordinating a large-scale campaign (Orsman 2016

Conclusions: a way forward
The international literature highlights the importance of the rational voter theory (Franklin 2004): elections must matter and be perceived as important to voters. Age and generational effects, youth footprint theory (Franklin 2004;Vowles 2010), the concept of first-and second-order elections (Riker and Ordeshook 1968;Reif and Schmitt 1980), and social group or network theory may all also have an influence on voter participation in local government elections Pendakur 2009, 2011;Bhatti and Hansen 2012;Blais and Rubenson 2012). Both the literature and the present study's analysis highlight deterrents to voting arising from the way local elections are conducted in New Zealand. The latest decline in electoral turnout in the 2019 local government elections suggests a need for immediate reform that recognises these influences. Without significant reform, local democracy in New Zealand, specifically citizen engagement in the local electoral process, looks set to continue to decline. The authors propose that the various issues, influences and specific factors highlighted in this paper be addressed across four areas of reform, as follows: (1)  This is a challenge which the New Zealand Political Studies Association (2020) is embracing, alongside the Ministry of Education's own initiatives in Wellington (Ministry of Education 2020). While specific civics education would be beneficial, the authors acknowledge the potential for tensions between the neo-liberal elite within New Zealand and a 'civics-focused' population, which would be able to assess and critique the dismantling of traditional institutions of governance from an informed perspective (Kelsey 1995).
Finally, New Zealand needs to re-evaluate the current approach to 'second-order' elections: making both local elections and candidates more visible, taking note of the observations of Reif and Schmitt (1980) cited earlier, and recognising the arguments for political efficacy. To avoid the scenario of declining voter turnout translating into an inexorable decline in local democracy, councillors and mayors must be much more visible and their responsibilities better understood by the populace. They need to escape the shackles of the town hall and reinforce the important role of local government, which CJLG June 2021 93 arguably has more impact upon our daily lives than any other level or institution of the state. This reorientation would be about providing voters with needed information, not electoral propaganda. Central government needs to step up and take the lead, in partnership with Local Government New Zealand.
Taking Auckland as an example, it would be immensely influential if government ministers were to be constructive about the role of Auckland Council and hence about the importance of local elections.
If significant cultural change does not occur at both the local and national levels of government, engagement and trust in local government will continue to languish and election turnout will be in terminal decline. This risks moving local government from a functioning sphere of democracy to an administrative arm or agency of central government. Clearly, traditional methods of getting political messages across no longer have the same impact. For many citizens, social media has replaced the evening news. Yet, for example, the Scottish independence referendum in the UK in 2014 showed that when young people are engaged with an issue they see as important and relevant to them, they participate in record numbers with, in this case, over 80% of those eligible to vote doing so (McSmith 2014). Galvanising this engagement is the key task we face in New Zealand.

Given this background, the words of Jim Hacker from the British television comedy Yes Prime Minister
talking about English local governmentseem most apt: Only about 25% of the electorate vote in local elections. And all they do is treat it as a popularity poll on the political leaders [in central government]… Nobody knows who their local councillor is. And the councillors know nobody knows who they are. Or what they do. So they spend four totally unaccountable years on a publicly-subsidised ego trip, handing out ratepayers' hard-earned money... (Lynn and Jay 1989, p. 134).
As this paper was being finalised, New Zealand's Minister of Local Government, the Honourable Na Nanaia Mahuta, announced a review entitled 'The Future of Local Government'. This is the first fundamental, root-and-branch examination of the scope and scale of local government since the reforms of the late 1980s. Central to this review will be many of the issues identified in this paper: citizen engagement; participation; and how to achieve reinvigorated, vibrant local democracy. The review panel is scheduled to present its report in April 2023. The authors await its findings with great trepidation and hope this paper will be of interest to its members.

Declaration of conflicting interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The authors received no financial support for the research, authorship, and/or publication of this article.