Previous Next

A Populist Exception?

4

Populism, Authoritarianism, Vote Choice and Democracy

Jack Vowles

Jacinda Ardern’s responses to the terrorist attack on Islamic worshippers in Christchurch, on 15 March 2019, coined a phrase that was used repeatedly in the following days: ‘they are us’ (Ardern, 2019). She also articulated a vision of a nation united in its support for those who had died and those who survived to mourn them: ‘we are one’. This presented an exemplary rejection of exclusionary political rhetoric (see also Chapter 6). In contrast, Winston Peters’ words emphasised what he, and many others, now consider to be the core values of a New Zealander, the defining characteristics of the New Zealand people and a consequent source of national pride: giving people a ‘fair go’, practicality and tolerance (Peters, 2019). Unlike Ardern, he did not acknowledge difference by way of a collective ‘they’, other than through the principle of freedom of religion: an individual freedom. Arden used the word ‘nation’, whereas Peters used ‘New Zealand people’—a key signifier of populism. These represent subtle but significant differences.

At the end of 2018, New Zealand signed the United Nations Migration Compact, an agreement on common principles to apply to immigration policies. The United States and Australia refused to sign, claiming that the compact could abrogate national sovereignty. The New Zealand Government received legal advice confirming that the compact would have no binding effect on immigration and foreign policy. The Deputy Prime Minister and Minister of Foreign Affairs, Winston Peters, agreed; however, the National Party and its leader, Simon Bridges, opposed the pact on sovereignty grounds, despite the previous National-led government’s apparent support for the agreement (Bridges, 2018). A petition for withdrawal was launched, only to be removed in somewhat mysterious circumstances from the National Party’s website in the aftermath of the Christchurch mosque attack. A few weeks before the mosque attack, death threats had been made against Winston Peters at a far-right rally in Christchurch (Gower, 2019). Elements associated with the various varieties of populism, as defined in the literature, are found in virtually all New Zealand political parties—and not always in the expected places.

First, this chapter examines the association between populism and authoritarianism in the language of New Zealand’s political parties. Next, it examines the relationship between populist and authoritarian attitudes and left and right policy dimensions, to determine whether populism is predominantly a left-wing or right-wing phenomenon in New Zealand. In terms of vote choices, it is expected that both populism and authoritarianism will be associated with votes for New Zealand First. Considering further implications for the condition of democracy in New Zealand, the next step is analysis of the extent to which populism and authoritarianism can be linked to both satisfaction with, and support for, democracy. Initial expectations follow from theory: populists will be unsatisfied with, but supportive of, democracy. Because of the potential overlap between low political efficacy and populist attitudes as operationalised in our data, a combination of low political efficacy and anti-pluralism might shift populists towards apparently lower support for democracy. In theory, authoritarians should be more likely to be both unsatisfied and unsupportive—in the New Zealand case in particular, this is because they are likely to be less enamoured of proportional representation than liberals. However, the New Zealand Government continues to be based on concentrated power in a unitary state and still lacks constitutional restraints on legislative and executive authority—an institutional framework that should be appealing both to authoritarians and populists.

Party Positions and Discourses

Party positions are best estimated from the statements they make with the widest public currency—the kind of statements that were used to introduce this chapter. The extent to which parties across the globe employ populist discourse in their manifestos has formed the focus of much recent research. An excellent source of data on political party discourse regarding populism is the Chapel Hill Expert Survey (see Polk et al., 2017); however, this is largely confined to European countries. A 2019 paper includes analysis of recent New Zealand political party manifestos among 119 others from the United States, Canada, the United Kingdom, Ireland, South Africa and Australia (Dai, 2018). Consistent with interpretations of the populist nature of New Zealand political culture, Dai found that the most populist example was that of New Zealand First in 2011 and the fourth that of the New Zealand Labour Party in 2011. The 2008 New Zealand National Party manifesto ranked eighth. Across different election years, however, indicators of populism have waxed and waned, even among parties (e.g. New Zealand First) that are widely understood to be populist.

Dai employs a sophisticated methodology that matches phrases and combinations of words but also operationalises populism according to the Mudde model; that is, it is assumed to be anti-pluralist, moralist and the antithesis of liberal pluralism—the conflation of populism with authoritarianism discussed in previous chapters. It is worth noting that the election years 2008 and 2011 took place during, and immediately after, the global financial crisis, providing ample ammunition for anti-elite discourse; further, the three-party programmes in question were from opposition parties. Dai’s dataset does not include 2017 New Zealand political party statements. We now focus our attention on these. Like Dai, we searched for key words and phrases reflecting possible populism and authoritarianism; however, the quantity of text to be analysed is small enough not to require electronic processing.

Table 4.1 comprises several sections. The first contains populist words and phrases drawn from New Zealand First’s 2011 leader’s speech, probably the most populist document in New Zealand politics in recent history (Peters, 2011). Following this are similar phrases drawn from the 2017 New Zealand First, Labour, National and Green Party policy statements.

In 2017, the central focus of New Zealand First policy was ‘the regions’—that is, provincial areas falling behind in growth, living standards and infrastructure development. However, beneath this overall theme, populist language emerged, mostly directed at foreign ownership of New Zealand assets and New Zealand First’s perception of excessive recent immigration to New Zealand (Peters, 2017).1

Table 4.1: Populist words and phrases in New Zealand politics

A: New Zealand First 2011

the few

the favoured

special interests

all New Zealanders

special treatment

the many

stand for people

welfare of all

a fair go

rich people

bosses

mates

fat cats

foreign

rich person

multi-millionaires

government’s mates

all the people

select few

secret deals

people no say

foreign ownership

New Zealand ... great again

unite the nation

closed doors

crooks

bankers

financial wheeler-dealers

ordinary people

New Zealand ownership

one law for all

voice of the people

people power

not just the few

Source: Peters, 2011.

B: New Zealand First 2017

corrupt

all of New Zealand

open door immigration

queen street farmers

financial speculators

overseas owners

sell-off of our country

foreign buyers

foreign ownership

foreign companies

banana republic

record net immigration

Source: Peters, 2017.

C: Labour 2017

speculators’ unfair tax advantages

gap between rich and poor

what New Zealand meant to be

ban foreign buyers/speculators

take a breather on immigration

Source: Ardern, 2017.

D: National 2017

back New Zealanders

Kiwi character

all New Zealanders

Source: English, 2017.

E: Green 2017

policies decided by our members

a movement of New Zealanders

choose whose side we are on

take our country back

people that really represent them

a country that works for and includes everyone, that excludes no one

we are only great, when we are great together

Source: Shaw (2017); Turei (2017).

The central theme of the Labour Party’s main campaign speech was its new leader—Jacinda Ardern. Her immediate background before becoming a member of parliament was as a party policy advisor, including working for the Blair government in the United Kingdom. However, she drew on her childhood and youth to successfully present herself as a small-town person with deep roots in everyday New Zealand. Labour language approached populism via use of ‘rich and poor’. Labour expressed intentions to reduce the tax advantages of speculators in the housing market, particularly those from overseas, and to ‘take a breather’ on immigration. Meanwhile, significant sections of Labour policy also promoted multiculturalism (New Zealand Labour Party, 2017).

National’s principal policy statement, delivered in its campaign opening speech, stressed the importance of governing in the interests of those who ‘work hard and back themselves’ with ‘ambition for the future’. National proposed to back New Zealanders in those entrepreneurial terms and praised the ‘Kiwi character’, as so defined. However, the discourse is otherwise lacking in populist rhetoric (English, 2017). The two Green Party speeches could be described as containing ‘populism-lite’, referring to internal party democracy, our country, representation of all and New Zealanders being ‘great together’—this slogan had to be replaced during the campaign when co-leader Metiria Turei was forced to resign following her disastrous speech and two members of parliament left the party due to their premature demand for her departure (Shaw, 2017; Turei, 2017). We conclude that populist ‘frames’ were present in New Zealand politics in 2017; however, these were not dominant or central. Concerns regarding foreign ownership and immigration were present but expressed in relatively moderate language. Labour’s immigration policies were motivated by economic and social factors, most notably in the context of a highly inflated housing market combined with a housing shortage, for which high levels of recent immigration, foreign buyers and speculation in general were claimed to be responsible (a claim for which there was some evidence).

Data and Operationalisation

The authoritarian and populist attitudinal scales (defined in Chapter 3), as applied to New Zealand voters, correlate quite strongly with how people rate themselves on the left and right scale. Figure 4.1 demonstrates this using simple regressions of authoritarianism and populism together, against left–right position (details of this can be found in Table A4.1). New Zealand authoritarians tend to the right and populists to the left. A control for external political efficacy in Table A4.1 makes little or no difference to the strength of the relationship, for either populism or authoritarianism. Populism, in the New Zealand context, is a phenomenon generally found on the left rather than the right.

Figure 4.1

Figure 4.1: Populism, authoritarianism and the left–right scale.

Source: New Zealand Election Study (2017) (see Table A4.1 [Model 1]).

Figure 4.2

Figure 4.2: Populism, authoritarianism and vote choice in 2017.

Source: New Zealand Election Study (2017) (see Table A4.2 [Model 1]).

The next factor for consideration is party vote choice. Figure 4.2 shows that the two major parties—National and Labour—sit on different sides of the coin regarding both dimensions (see the estimates in Table A4.2). The Labour vote is associated with populism and liberal social attitudes and the National vote with elitism and authoritarianism; these associations are quite strong. The Green vote has a weak relationship with populism but a very strong relationship with liberal social attitudes. As expected, those who vote for New Zealand First tend to combine populism with authoritarianism.2 A second model, augmented with social structure controls and an estimate of respondents’ political efficacy, shows that the association between these dimensions and vote choice is almost completely unaffected by accounting for these factors (see Table A4.3). The association between populism and a desire to reduce immigration is partly explained by party policy positions. The Labour Party stated that New Zealand should ‘take a breather’ on immigration, whereas National made no statements on immigration in its major policy presentations.

A third model somewhat qualifies these findings. Populism might be associated with voting for the left in New Zealand in 2017 simply because the left had not occupied office since 2008; therefore, left-leaning voters may have felt less efficacious than under different circumstances. Although we employ a control for political efficacy, one’s party being out of government may still encourage apparent populist attitudes. The sample data enable investigation of this possibility, to some extent. The election was quite close; therefore, it was unknown for some time which major party would form the government. As the National Party won the most votes, until 17 October, the most probable outcome was that National would win negotiations with New Zealand First. However, post-October, it became clear that Labour would govern. Two-thirds of the sample responded before the announcement and one-third after. We created a dichotomous variable based on this distribution and further weighted the sample to ensure party vote distributions were the same in each set of respondents;3 the coalition formation dummy was then interacted with the populism scale (see Figure 4.3).

In Table A4.3, no interactions appear significant; however, the ‘after government formation’ variable is significant for both Labour and the Green Party. Plotting the interaction effects, a more robust approach, the two slope lines for the National Party were almost identical (and, for this reason, not displayed in Figure 4.3). The interaction is not significant for the Green Party vote and confidence intervals also overlap; however, we note that the weak association between populist attitudes reverses between pre- and post-government formation. Populism becomes a little less associated with the Labour vote following the government announcement. The vote probability gap is substantial and just outside confidence intervals. For New Zealand First, there emerges no difference. A change of government in favour of the left may reduce agreement with left populist statements; however, there still exists a strong association between populism and Labour vote choice. Models 3 and 4 in Table A4.1 confirm the same effect for the relationship between populism and left–right placements; it was a little weaker, although still strong, after the change of government was announced. Again, this was well within confidence intervals; however, in this case, the interaction effect is statistically significant.

Figure 4.3

Figure 4.3: Populism conditioned by time of government formation.

Source: New Zealand Election Study (2017) (see Table A4.2 [Model 3]).

Democracy: Satisfaction and Support?

The main current of populist-inspired voting in New Zealand tends towards Labour, a party of the mainstream centre-left, while authoritarianism is more strongly associated with voting for the mainstream centre-right National Party. New Zealand First, the only party whose voters combine populism and authoritarianism, is a minor (albeit pivotal) party. However, further analysis is required to probe more deeply into this issue—both attitudinal dispositions are relatively prevalent in New Zealand and there may be additional consequences.

Concern regarding the future of democracy is currently a major theme in comparative political science. Examples of creeping authoritarianism in countries such as Poland, Hungary and Turkey garner much attention. Of even greater concern is a claim that support for democracy is declining among mass publics in the most apparently secure and stable democracies, such as New Zealand (Ferrin & Kriesi, 2016). Populist and authoritarian attitudes could underpin or at least reinforce this trend. Indeed, there is strong behavioural evidence that political participation of most kinds is declining in established democracies. Until recently, such fears were confined to electoral turnout; however, they have now moved further afield and are found in, for example, so-called unconventional forms of participation such as protest (Grasso, 2016). A long-term trend of turnout decline has been well documented in New Zealand (Vowles, 2014), although turnout recovered somewhat in 2014 and again in 2017, albeit from a low base. The greatest concern is that these trends are shaped by age differences that are generational, rather than simply reflecting change over the life cycle, as seems to be the case for turnout in New Zealand and many other countries (Franklin, 2004; Vowles, 2010).

Behavioural change does not necessarily signify wholesale attitudinal change. People may continue to support democracy without feeling the need to participate themselves—a phenomenon labelled ‘stealth democracy’ (Hibbing & Theiss-Morse, 2003). Two instruments are available from the 2017 New Zealand Election Study (NZES) to test the relevant attitudes. First is a standard question eliciting satisfaction with ‘how democracy works’ in New Zealand (with options of ‘very satisfied’, ‘fairly satisfied’, ‘not very satisfied’, ‘not at all satisfied’ and ‘don’t know’). ‘Satisfaction with democracy’ is one of the most widely used estimates in the literature to assess both the accountability and responsiveness of democracies. However, this does not escape criticism—it taps into a wide range of sentiments and may confuse evaluations of democracies in principle and in practice (Thomas, 2016). Nonetheless, it has value as a summary measure; its wide use in the literature attests to its worth, subject to caution (Anderson, 2002). Democratic dissatisfaction may represent ‘a felt discrepancy between democratic norms and the actual democratic process’ (Thomassen, 1995, p. 383).

Democratic satisfaction has been measured in New Zealand since 1996. Unfortunately, we lack a time series prior to electoral system change. Two mid-term election datasets illustrate the picture in both 1998 and 2001. Data collection in 1998 was fortunately timed to capture the collapse of the first coalition government under the mixed member proportional (MMP) system and the following fallout (Karp & Bowler, 2001). Figure 4.4 shows that, at the first MMP election, the level of democratic satisfaction was only a little below 70 per cent. Disillusion followed but satisfaction had returned almost to the 1996 level by 2002. Since 2005, democratic satisfaction has remained steady at approximately 65 per cent, which is somewhat better than average, in international terms, but not outstanding (Aarts & Thomassen, 2008, p. 12; Thomas, 2016, p. 219). In 2017, a slight drop may be observed; however, this is within confidence intervals when compared to 2014. We elucidate the possible reason for this apparent change in Chapter 8.

Figure 4.4

Figure 4.4: Satisfaction with democracy in New Zealand (1996–2017).

Source: New Zealand Election Study (2017).

Unless New Zealand democracy fully meets their expectations, which is unlikely, one would expect populists to be less satisfied with democracy than non-populists. For authoritarians, more ambiguity is likely—dissatisfaction with the performance of democracy does not necessarily imply dissatisfaction with democracy, as such, or that people would prefer some authoritarian alternative. Alternatively, an authoritarian might be satisfied with democracy when things are going well but prefer authoritarian government when things go badly (Linde & Ekman, 2003).

The second question available in the 2017 NZES is a five-point scale measuring agreement or disagreement with the statement: ‘Democracy may have problems but it’s better than any other form of government’. This question implicitly references Winston Churchill’s famous words, a rueful comment on his rejection by the British electorate in 1945. The same question was also asked in the 2002 NZES, thereby providing a useful comparison across 15 years. Concern regarding declining support for democracy in established democracies has been widely expressed and younger generations have been identified as those most susceptible (Foa & Mounk, 2017). However, this finding appears driven largely by the United States. Elsewhere, including New Zealand, over the last 20 years, across a range of questions, overall support for democracy remains high, evidencing little change (Voeten, 2017). NZES comparison between 2002 and 2007, setting the five-point scale at minimum 0 and maximum 1 confirms this, demonstrating no significant difference in agreement with the question—the mean is 0.8, indicating a relatively high level of support. Nonetheless, changes within the electorate, particularly among and across generations, may still constitute cause for concern.

Figure 4.5 displays two sets of estimates derived from two linear regression models, one on satisfaction with democracy, the other on ‘democracy is better’; each use five-point scales with minimum set at 0 and maximum at 1 (see Tables A4.3 and A4.4). The figure shows the probability shifts of the categorical variables from their minimum to maximum values. Figures for the continuous variables—populism, authoritarianism, income and political efficacy—are found below. The models were also run using ordinal logit, which returned almost identical results. Our use of linear regression provides results that are easier to interpret.

Figure 4.5

Figure 4.5: Satisfaction with and support for democracy in New Zealand.

Note: Dots and 95 per cent confidence intervals indicate the comparative positions of each group on the two scales, with satisfaction/support at a maximum of 1 and dissatisfaction/opposition at a maximum of 0.

Source: New Zealand Election Study (2017) (see Table A4.3 [Model 4] and Table A4.4 [Model 4]).

The model contains controls for ‘winner’ and ‘loser’ to estimate short-term effects. It is expected that winners will be more satisfied and supportive, losers less so. The election result and the delay in deciding which parties would govern complicate the coding of this question. ‘Winners’ are defined as government parties; however, prior to Winston Peters’ announcement, most people expected that National and New Zealand First would govern. Therefore, ‘winners’ were coded as National and New Zealand First before the announcement of the government, and Labour, New Zealand First and Green afterward. ‘Losers’ were defined as parties who moved out of government—this applied to Māori Party and ACT voters throughout, because a National–New Zealand First coalition would have been very unlikely to include them, and also to National following the formation of the Labour-led coalition (and Labour before it). The group in the middle are non-voters and those voting for parties not in the government either before or after the election. Figure 4.5 shows the expected effects for winners and losers under satisfaction; however, little of note emerges under ‘democracy is best’. The questions appear to effectively separate short-term and long-term perspectives.

We defined generational cohorts as in Chapter 3. Compared to war and pre-war generations, successive generations become less satisfied with democracy until generation Z, whose score reverts in the other direction. The same pattern, albeit stronger, emerges for support. Younger generations, or age groups, show declining levels of support and satisfaction, consistent with fears expressed concerning generational decline in support in established democracies.

The use of the words ‘generations’ or ‘age groups’ as alternatives highlights a key point. The differences may not be generational but instead reflect the life cycle; support for and satisfaction with democracy is lower among the young but rises as people age and become more satisfied and supportive. In the wider international debate (in contrast to the debate regarding electoral turnout), the life cycle interpretation has the best evidence (Norris, 2017; Voeten, 2017). Shorn of all controls, Figure 4.6 displays the ‘generational’ probabilities of supporting democracy in 2002 and 2017, with the generations pinned to birth years, rather than age at any one time. The older generation becomes significantly more likely to support democracy. In 2017, this is a smaller group, excluding those who have died in the intervening years. Boomers remain in the same position; however, generation X shifts towards a higher level of democratic support. Millennials (only slightly represented in 2002 compared to 2017) may also have shifted upward; however, the confidence intervals overlap. Generation Z could not vote in 2002 and their slightly higher level than that of millennials in 2017 is not statistically significant. On balance, the evidence from this analysis best supports the life cycle interpretation.

Returning to Figure 4.5, Pākehā are, surprisingly, the least satisfied with democracy. Māori are more satisfied, but still within confidence intervals. Pasifika stand out as the most satisfied group. The Māori finding is surprising and presumably results from other variables in the model. With ethnic groups alone, in an alternative model, the picture changes: Māori are the least satisfied (although, again, they are not statistically different from Pākehā: the difference is only 0.03). All immigrant minority groups are more satisfied than the rest; however, according to the confidence intervals, the difference is only robust between Pākehā and Pasifika. Regarding the rest of the variables, confidence intervals tend to overlap or are quite close.

Figure 4.6

Figure 4.6: Generational comparisons of support for democracy.

Note: Lines and 95 per cent confidence intervals indicate the comparative positions of each group on the two scales, with satisfaction/support at a maximum of 1 and dissatisfaction/opposition at a maximum of 0.

Source: New Zealand Election Study (2002, 2017).

As expected, the ethnic groups divide more on ‘democracy is best’. Pākehā and Pasifika are most likely to agree; Māori, Asians and ‘others’ are less likely to agree. Given their experience as a colonised minority, the Māori position is to be expected. Asian and ‘other’ ethnic groups, most of whom are likely to be recent immigrants, are also more likely to have been born in countries less democratic than New Zealand. There exist more differences between the remaining groups for ‘democracy is best’. Men, those with university degrees and major urban dwellers are somewhat more pro-democracy than women, those without degrees and those living outside major urban areas.

Figure 4.7 shows that authoritarians are less satisfied with democracy, as expected, but are only marginally less likely to support democracy than liberals—the difference is, statistically, nothing to speak of. This is an unexpected but notable finding—it appears that New Zealand authoritarians do not desire non-democratic alternatives.

Figure 4.7

Figure 4.7: Satisfaction with democracy, support for democracy and authoritarianism.

Source: New Zealand Election Study (2017) (see Table A4.3 [Model 4] and Table A4.4 [Model 4]).

Figure 4.8

Figure 4.8: Satisfaction with democracy, support for democracy and income.

Source: New Zealand Election Study (2017) (see Table A4.4 [Model 3] and Table A4.4 [Model 3]).

As noted earlier, despite a proportional electoral system, governmental authority in New Zealand is concentrated in a unitary state, with no fundamental law to constrain the power of the legislature and executive. However, we might expect authoritarians to resist proposals for the introduction of binding constitutional law (see e.g. Palmer & Butler, 2019) or for devolution of central government authority to regional and local authorities. Figure 4.8 shows that income has no effect on satisfaction with democracy but demonstrates a significant association with democratic support—those with higher incomes are more likely to support democracy against alternatives.

Figure 4.9 shows that external political efficacy is strongly associated with both satisfaction with and support for democracy (but most strongly with the latter). Figure 4.10 displays the relationship between populist attitudes, satisfaction with democracy and support for democracy. As expected, populists are much less satisfied with democracy than non-populists. By contrast, regarding democratic support, although the slope of the probability estimate is in the same direction, the confidence intervals overlap and the coefficient is insignificant in Model 3 (Table A4.4). In the first two models of that table, populism appears strongly negatively associated with democratic support; the addition of efficacy to Model 3 reduces populism to statistical insignificance. If efficacy is a short-term perception of government responsiveness and populism is representative of more deep-seated preferences regarding government, we may conclude that populists in New Zealand are not anti-democratic. However, the relationship may not be so straightforward—populists may be generally prone to feelings of low external efficacy.

Figure 4.9

Figure 4.9: Democratic satisfaction, support for democracy and external efficacy.

Source: New Zealand Election Study (2017) (see Table A4.4 [Model 3] and Table A4.4 [Model 3]).

Figure 4.11 returns the focus to age, this time included by year rather than by generational cohort. The figure is derived from Model 5 on democratic support. To give a more robust estimate, the points at which populism is measured are somewhat estimated slightly short of the extreme values of the scale (which are 0 or 1). It shows that a significant proportion of age difference in support for democracy can be attributed to populists, who become more supportive of democracy as they grow older, perhaps as their expectations of democracy become more modest. This model includes the control for efficacy—without this, there exists a somewhat stronger effect, indicating the effect of increasing efficacy as people age. However, this figure controls for this effect.

Figure 4.10

Figure 4.10: Democratic satisfaction, support for democracy and populism.

Source: New Zealand Election Study (2017) (Table A4.3 [Model 4] and Table A4.4 [Model 4]).

Figure 4.11

Figure 4.11: How populism and age affect democratic support.

Source: New Zealand Election Study (2017) (see Table A4.5 [Model 5]).

Discussion and Conclusions

In New Zealand, populism is predominantly an attribute of those who lean left, rather than right. In part, this may be attributed to the effect of a centre-right government who have been in office since 2008. Left populist sentiments may be less associated with vote choice when the left is in power—populism did become somewhat less associated with Labour vote choice after the outcome of the election gave power to their party; consequently, some Labour voters became a little less populist. Over the longer term, an even stronger effect may be expected. Therefore, replication of this analysis with 2020 election data should be a strong priority. The separation between the two dimensions of populism and authoritarianism means that Labour may appeal to populists, but not so much to authoritarians, and National may appeal to authoritarians, but not so much to populists. The combination of populism and authoritarianism occurs among those who vote for New Zealand First; otherwise populists tend to vote Labour, authoritarians National and left–right positions display the same pattern.

As might be expected, populists are less satisfied with New Zealand democracy than non-populists. After controlling for political efficacy, populists demonstrate a high level of support for democracy; however, the level is not significantly higher than that of the population in general and is, perhaps, marginally lower. This is more likely a reflection of disappointment with, rather than a rejection of, democracy. Authoritarians are also no less satisfied with or supportive of New Zealand democracy than liberals. These findings are reassuring for democrats, as are the indications that support for democracy in New Zealand is not declining over time. If the young are less supportive of democracy than the old, this is most likely a life cycle rather than a generational effect. As people age, and tend to become more secure and integrated into society, they become more supportive of democracy. Higher incomes, higher education and, to a lesser extent, accumulation of assets also generate greater support. Populists, more so than non-populists, appear to be most responsible for this ageing effect.

There are several key implications for New Zealand politics. A greater degree of both populism and authoritarianism is clearly possible in New Zealand politics. Social and economic inequalities remain relatively high and demonstrate a strong association with social and political cleavage structures (Vowles, Coffé & Curtin, 2017). Māori, Pasifika and all those on low incomes and with insecure employment are generally less satisfied and less supportive of the democratic status quo. New Zealanders were lucky that their experiences of the global financial crisis and the following recession were mild, compared to those in many other countries. Immigration surged significantly after 2013; however, this occurred in a context of economic recovery and labour shortages. New Zealand is advantaged by moderate mainstream party elites and a conservative, but not politically manipulative, traditional media—extremist voices receive little traction; however, social media give them more opportunities than in the past. In the aftermath of the Christchurch attack on Islamic worshippers, political elites and traditional media uniformly broadcasted a message of social inclusion and cultural tolerance.

A historian coined the phrase ‘a lucky country’ to describe Australia in the 1960s (Horne, 1964). The description stuck, despite being ironic. Australians have demonstrated a habit of co-opting New Zealand’s achievements—perhaps it is time to return the favour? Crucially, this luck is only relative—in comparison to the darker pathways being followed in other countries. Despite widely acknowledged contemporary flaws in its politics and society, New Zealand’s moderately populist democracy better fits the ‘lucky’ label in the early 21st century. However, given historical experience, social inequalities and injustice, the ambitions of opposition politicians, unstable global politics and the possibility of a domestic spillover in relation to an external shock or global recession, no one can be sure that this relatively happy state of affairs will last.

References

Aarts, K. & Thomassen, J. (2008). Satisfaction with democracy: Do institutions matter? Electoral Studies, 7, 5–18. doi.org/10.1016/j.electstud.2007.11.005

Anderson, C. J. (2002). Good questions, dubious inferences and bad solutions: Some further thoughts on satisfaction with democracy (Unpublished manuscript). Retrieved from www.researchgate.net/publication/251498295_Good_Questions_Dubious_Inferences_and_Bad_Solutions_Some_Further_Thoughts_on_Satisfaction_with_Democracy/link/02e7e53199893ba764000000/download

Ardern, J. (2017). Jacinda’s speech to campaign launch. New Zealand Labour Party. Retrieved from www.labour.org.nz/jacindas_speech_to_campaign_launch

Ardern, J. (2019). Full statement: Jacinda Ardern addresses Parliament on Christchurch terror attack. Retrieved from www.tvnz.co.nz/one-news/new-zealand/full-statement-jacinda-ardern-addresses-parliament-christchurch-terror-attack

Bridges, S. (2018). New Zealand National party: Pull NZ out of the UN migration pact. Facebook. Retrieved from www.national.org.nz/national_would_pull_nz_out_of_un_migration_pact

Cook, H. & Manch, T. (2019). Major leak of NZ First membership database exposes personal details, Winston Peters refers to police. Stuff. Retrieved from www.stuff.co.nz/national/politics/116409225/major-leak-of-nz-first-membership-database-exposes-personal-details

Dai, Y. (2018). Measuring populism in context: A supervised approach with word embedding models. Retrieved from sites.psu.edu/yaoyaodai/files/2018/09/Dai_paper_populism-2bj779b.pdf

English, B. (2017). Speech to the National Party campaign launch 2017. New Zealand National Party. Retrieved from www.national.org.nz/speech_to_the_national_party_campaign_launch

Ferrin, M. & Kriesi, H. (2016). How Europeans view and evaluate democracy. Oxford, United Kingdom: Oxford University Press.

Foa, R. S. & Mounk, Y. (2017). The signs of deconsolidation. Journal of Democracy, 28(1), 5–15.

Franklin, M. N. (2004). Voter turnout and the dynamics of electoral competition in established democracies since 1945. Cambridge, United Kingdom: Cambridge University Press.

Gower, P. (2019). Police investigating white supremacists’ death threats against Winston Peters. Newshub. Retrieved from www.newshub.co.nz/home/new-zealand/2019/06/exclusive-police-investigating-white-supremacists-death-threats-against-winston-peters.html

Grasso, M. T. (2016). Generations, political participation and social change in Western Europe. London, United Kingdom: Routledge.

Hibbing, J. & Theiss-Morse, E. (2003). Stealth democracy: Americans’ beliefs about how government should work. Cambridge, United Kingdom: Cambridge University Press.

Horne, D. (1964). The lucky country? Hawthorne, Australia: Penguin Australia.

Karp, J. A. & Bowler, S. (2001). Coalition politics and satisfaction with democracy: Explaining New Zealand’s reaction to proportional representation. European Journal of Political Research, 40(1), 57–79. doi.org/10.1111/1475-6765.00589

Linde, J. & Ekman, J. (2003). Satisfaction with democracy: A note on a frequently used indicator in comparative politics. European Journal of Political Research, 42, 391–408. doi.org/10.1111/1475-6765.00089

Norris, P. (2017). Is Western democracy backsliding? Diagnosing the risks. Retrieved from www.journalofdemocracy.org/wp-content/uploads/2018/12/Journal-of-Democracy-Web-Exchange-Norris_0.pdf

New Zealand Election Study. (2002). New Zealand Election Study [dataset]. Retrieved from www.nzes.org/exec/show/data

New Zealand Election Study. (2017). New Zealand Election Study [dataset]. Retrieved from www.nzes.org/exec/show/data

New Zealand Labour Party. (2017). Think Labour’s got no policies? Think again. Retrieved from www.labour.org.nz/thelonglist

Palmer, G. & Butler, A. (2019). Towards democratic renewal: Ideas for constitutional change in New Zealand. Wellington, New Zealand: Victoria University Press.

Peters, W. (2011). For a fair go. Scoop. Retrieved from www.scoop.co.nz/stories/PO1110/S00543/rt-hon-winston-peters-for-a-fair-go.htm

Peters, W. (2017). NZ First campaign launch speech. New Zealand First. Retrieved from yournz.org/2017/06/26/nz-first-campaign-launch-speech/

Peters, W. (2019). Winston Peters speech to commemorate the Christchurch Mosque. Scoop. retrieved from www.scoop.co.nz/stories/PA1903/S00133/winston-peters-speech-to-commemorate-the-christchurch-mosque.htm

Polk, J., Rovny, J., Bakker, R., Edwards, E., Hooghe, L., Jolly, S., … Zilovic, M. (2017). Explaining the salience of anti-elitism and reducing political corruption for political parties in Europe with the 2014 Chapel Hill expert survey data. Research and Politics, 4(1), 1–9. doi.org/10.1177/2053168016686915

Shaw, J. (2017). Green Party campaign launch—Nelson 2017. Green Party. Retrieved from www.greens.org.nz/news/speech/james-shaw-green-party-campaign-launch-nelson-2017

Thomas, K. (2016). Democratic support and globalisation. In J. Vowles & G. Xezonakis (Eds), Globalisation and domestic politics: Parties, elections and public opinion (pp. 209–230). Oxford, United Kingdom: Oxford University Press. doi.org/10.1093/acprof:oso/9780198757986.003.0011

Thomassen, J. (1995). Support for democratic values. In H-D Klingemann & D. Fuchs (Eds), Citizens and the state (pp. 383–415). Oxford, United Kingdom: Oxford University Press. doi.org/10.1093/0198294735.003.0013

Turei, M. (2017). Green Party campaign launch—Nelson 2017. Green Party. Retrieved from www.greens.org.nz/news/speech/metiria-turei-green-party-campaign-launch-nelson-2017

Voeten, E. (2017). Are people really turning away from democracy? Retrieved from www.journalofdemocracy.org/online-exchange-democratic-deconsolidation/

Vowles, J. (2010). Electoral system change, generations, competitiveness and turnout in New Zealand, 1963–2005. British Journal of Political Science, 40(4), 875–895. doi.org/10.1017/S0007123409990342

Vowles, J. (2014). Down, down, down: Turnout from 1946 to 2011. In J. Vowles (Ed.), The new electoral politics in New Zealand: The significance of the 2011 Election (pp. 53–73). Wellington, New Zealand: Institute for Governance and Policy Studies.

Vowles, J., Coffé, H. & Curtin, J. (2017). A bark but no bite: Inequality and the 2014 New Zealand general election. Canberra, Australia: ANU Press. doi.org/10.22459/BBNB.08.2017

Appendices

Table A4.1: Populism, authoritarianism and the self-assigned left (0)–right (10) position

Ordinary Least Squares

Model 1

Model 2

Model 3

Model 4

Populism

–3.053***

–2.978***

–3.478***

–3.408***

(0.326)

(0.320)

(0.434)

(0.426)

After government (AG)

–0.660**

–0.658**

(0.298)

(0.298)

Populism* AG

1.232**

1.253**

(0.547)

(0.548)

Authoritarianism

4.576***

4.546***

4.520***

4.486***

(0.293)

(0.290)

(0.293)

(0.291)

Efficacy

0.342*

0.369*

(0.206)

(0.205)

Constant

4.566***

4.279***

4.803***

4.494***

(0.154)

(0.209)

(0.206)

(0.239)

Observations

3,455.000

3,455.000

3,455.000

3,455.000

R-squared

0.192

0.193

0.192

0.193

Note: Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1.

Source: New Zealand Election Study (2017).

Table A4.2: Populism, authoritarianism and party vote in 2017

Model 1

Multinomial Logit

(1)

Non-vote

(2)

Labour

(4)

Green

(5)

New Zealand First

(6)

Other

Populism

4.065***

5.293***

3.835***

4.974***

3.659***

(0.702)

(0.400)

(0.597)

(0.604)

(0.532)

Authoritarianism

–1.668***

–3.980***

–8.071***

–0.313

–3.092***

(0.547)

(0.365)

(0.553)

(0.584)

(0.430)

Constant

–1.15***

–0.846***

–0.416

–4.200***

–1.695***

(0.324)

(0.190)

(0.264)

(0.357)

(0.268)

Pseudo R-squared

0.075

Observations

3,438.000

3,438.000

3,438.000

3,438.000

3,438.000

Model 2

Multinomial Logit

(1)

Nonvote

(2)

Labour

(4)

Green

(5)

New Zealand First

(6)

Other

Populism

2.530***

4.573***

2.955***

4.306***

3.125***

(0.784)

(0.403)

(0.658)

(0.633)

(0.594)

Authoritarianism

–1.375**

–3.687***

–7.158***

–0.780

–3.154***

(0.649)

(0.346)

(0.612)

(0.632)

(0.541)

Age

–0.032***

–0.004

–0.033***

0.008

–0.012*

(0.008)

(0.004)

(0.007)

(0.006)

(0.006)

Female (male)

–0.614***

–0.103

–0.387**

–0.635***

–0.635***

(0.220)

(0.123)

(0.195)

(0.186)

(0.200)

Māori (European)

1.427***

1.363***

0.887***

1.221***

1.788***

(0.293)

(0.231)

(0.338)

(0.300)

(0.269)

Pasifika (European)

0.460

0.819*

–1.151

0.270

0.251

(0.816)

(0.482)

(1.109)

(0.871)

(0.831)

Asian (European)

–0.088

–0.320

–1.062**

–15.182***

0.148

(0.441)

(0.276)

(0.474)

(0.238)

(0.500)

Other (European)

–14.536***

1.852*

–0.416

–14.115***

1.378

(1.131)

(0.950)

(1.512)

(1.115)

(1.346)

University

–0.081

0.464***

0.674***

0.087

0.340

(0.303)

(0.143)

(0.220)

(0.266)

(0.222)

Household income

–0.181**

–0.153***

–0.311***

–0.211***

–0.142**

(0.092)

(0.047)

(0.078)

(0.079)

(0.072)

Low-risk assets

–1.042*

–0.899***

–0.473

–0.558

–1.493***

(0.537)

(0.314)

(0.610)

(0.523)

(0.498)

High-risk assets

–1.175**

–1.323***

–0.601

–0.525

–1.267**

(0.545)

(0.319)

(0.621)

(0.538)

(0.509)

Religious

–0.002

–0.260**

–0.621***

0.064

0.076

(0.240)

(0.129)

(0.205)

(0.203)

(0.205)

Major urban

–0.181

0.249**

0.254

–0.242

0.171

(0.234)

(0.124)

(0.210)

(0.198)

(0.203)

Efficacy

–2.639***

–0.009

–0.490

–1.064***

–1.458***

(0.477)

(0.312)

(0.540)

(0.397)

(0.411)

Constant

4.313***

0.940*

3.151***

–1.721*

1.761**

(0.904)

(0.550)

(0.924)

(0.908)

(0.759)

Pseudo R-squared

3,230.000

3,230.000

3,230.000

3,230.000

3,230.000

Observations

4.313***

0.940*

3.151***

–1.721*

1.761**

Model 3

Multinomial Logit

(1)

Nonvote

(2)

Labour

(4)

Green

(5)

New Zealand First

(6)

Other

Populism

3.699***

5.503***

4.292***

4.644***

3.034***

(1.163)

(0.490)

(0.780)

(0.645)

(0.633)

After government (AG)

0.909

0.896**

1.180**

–0.636

–0.103

(0.701)

(0.370)

(0.524)

(0.717)

(0.535)

Populism* AG

0.189

–0.929

–1.736

0.805

1.114

(1.261)

(0.707)

(1.078)

(1.192)

(1.024)

Authoritarianism

–1.728***

–3.969***

–8.054***

–0.291

–3.121***

(0.569)

(0.365)

(0.563)

(0.578)

(0.434)

Constant

–1.898***

–1.133***

–0.778**

–3.971***

–1.577***

(0.533)

(0.231)

(0.338)

(0.368)

(0.326)

Pseudo R-squared

0.087

Observations

3,438.000

3,438.000

3,438.000

3,438.000

3,438.000

Note: Bracketed categories are those for reference. Vote for National (3) is the reference category for vote choice.

Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1.

Models 1 and 2 are weighted by demographics, education and party vote. Model 3 is further weighted by party votes, both pre- and post-government formation.

Source: New Zealand Election Study (2017).

Table A4.3: Satisfaction with democracy

Ordinary Least Squares

1

2

3

4

Populism

–0.462***

–0.454***

–0.469***

–0.446***

(0.046)

(0.046)

(0.048)

(0.046)

Authoritarianism

–0.057

–0.053

–0.055

–0.062

(0.037)

(0.037)

(0.040)

(0.039)

Winner

0.052***

0.048***

0.039**

(0.015)

(0.015)

(0.015)

Loser

–0.062***

–0.068***

–0.076***

(0.022)

(0.023)

(0.023)

Boomer

–0.048***

–0.046**

(0.018)

(0.018)

Generation X

–0.062***

–0.060**

(0.024)

(0.024)

Millennial

–0.082***

–0.073***

(0.024)

(0.024)

Generation Y

–0.055

–0.049

(0.036)

(0.038)

Female (male)

0.025*

0.019

(0.013)

(0.013)

Māori

0.047**

0.034

(0.021)

(0.021)

Pasifika

0.110***

0.112***

(0.040)

(0.040)

Asian

0.072*

0.073**

(0.038)

(0.036)

Other ethnic

0.144**

0.128**

(0.058)

(0.060)

University degree

0.018

0.016

(0.018)

(0.018)

Household income

–0.013

–0.010

(0.024)

(0.023)

Low-risk assets

0.056**

0.055**

(0.028)

(0.028)

High-risk assets

0.045

0.044

(0.029)

(0.028)

Major urban

0.025*

0.022

(0.014)

(0.014)

Religious

–0.015

–0.017

(0.015)

(0.015)

Efficacy

0.117***

(0.035)

Constant

0.857***

0.837***

0.825***

0.733***

(0.021)

(0.024)

(0.042)

(0.046)

Observations

3,403.000

3,403.000

3,215.000

3,215.000

R-squared

0.105

0.123

0.148

0.156

Note: Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1.

Source: New Zealand Election Study (2017).

Table A4.4: Democracy is better

Ordinary Least Squares

1

2

3

4

5

Populism

–0.240***

–0.215***

–0.117**

–0.072

–0.336***

(0.043)

(0.044)

(0.049)

(0.045)

(0.126)

Authoritarianism

0.018

0.002

–0.001

–0.015

–0.026

(0.042)

(0.042)

(0.046)

(0.041)

(0.037)

Winner

0.057***

0.030**

0.015

0.014

(0.014)

(0.013)

(0.013)

(0.012)

Loser

0.028

0.015

0.001

–0.000

(0.018)

(0.017)

(0.017)

(0.016)

Boomer

–0.062***

–0.056***

(0.015)

(0.015)

Generation X

–0.122***

–0.115***

(0.017)

(0.017)

Millennial

–0.179***

–0.160***

(0.019)

(0.018)

Generation Y

–0.123***

–0.111***

(0.028)

(0.031)

Age

0.001

(0.001)

Populism x age

0.005***

(0.002)

Female (male)

–0.016

–0.027**

–0.025**

(0.012)

(0.011)

(0.011)

Māori

–0.029

–0.053**

–0.049**

(0.022)

(0.022)

(0.021)

Pasifika

0.011

0.017

0.023

(0.032)

(0.031)

(0.030)

Asian

–0.079***

–0.078***

–0.076***

(0.027)

(0.026)

(0.026)

Other ethnic

–0.030

–0.061

–0.020

(0.067)

(0.069)

(0.060)

University degree

0.072***

0.067***

0.059***

(0.015)

(0.013)

(0.013)

Household income

0.044**

0.048**

0.044**

(0.022)

(0.021)

(0.020)

Low-risk assets

0.060**

0.058**

0.059**

(0.030)

(0.028)

(0.028)

High-risk assets

0.057*

0.057*

0.061**

(0.032)

(0.030)

(0.030)

Major urban

0.033***

0.028***

0.026**

(0.011)

(0.011)

(0.011)

Religious

0.020*

0.015

0.009

(0.012)

(0.011)

(0.012)

Efficacy

0.218***

0.213***

(0.029)

(0.028)

Constant

0.110***

0.143***

0.820***

0.648***

0.550***

(0.016)

(0.020)

(0.038)

(0.044)

(0.067)

Observations

3,367

3,367

3,213

3,213

3,213

R-squared

0.038

0.050

0.145

0.190

0.203

Note: Robust standard errors in parentheses.

*** p < 0.01, ** p < 0.05, * p < 0.1.

Source: New Zealand Election Study (2017).


1 Before the 2017 election, New Zealand First considered (but rejected on grounds of possible racist implications) a placard with the slogan: ‘It’s About You, Not Them’ (Cook & Manch, 2019).

2 The negative sign for authoritarianism in Table A4.2 is relative to the National vote reference category.

3 Chapter 8 explains the rationale and details of this weighting.


Previous Next