New Zealand 2017 Election – The Influence of Voters ’ Social -Economic Status on the Party Choice

A
considerable amount of literature has been published about election studies
around the world. An area that interests
many scholars in those studies is the influence of voters’ social-economic
status on the party of choice they voted for in elections (Coffe, Voweles & Curtin, 2014).
Therefore, this research intends to identify if the social-economic status of
voters influenced their voted choice of party in New Zealand’s 2017 election. 

 

It is significant to
conduct this study because all political parties present themselves as
universal with policies that are beneficial to all citizens (Afnoso, 2016).
However, an increasing gap between the rich and the poor should question those
policies. It should concern the government, the bureaucrats, and the public because
it can affect the social, political, and economic participation of disadvantage
groups (Szewczyk,
2015). And because inequality affects all spheres of
life, it can influence voters’ party choices. The party choice is so
significant because it is the party with the majority of candidates that
eventually forms the government and which decides on all of the country’s
affairs which will affect every individual citizen and future generations too. 

 

Because
of its significance, much research has been conducted in different countries to
identify the influence that voters’ social-economic status has on their party
choice. For example, Winters et al. (2017) conducted a related study in Britain, Pietsch (2017) in Australia
and Suryanarayan (2018) in
India and they discovered that there was strong
correlation between the social economic status of voters and their party
choice. However, a comparable study conducted in the United States by Gelman,
Kenworthy and Su (2010) found that there was no clear pattern of relationship
between voters’ social economic status and their party choice. Gelman et al. (2010) suggest that the
influence must be due to other factors such as religion, education, or gender.
In New Zealand, Coffe et al.
(2014) conducted similar research and found
that inequality does affect the party choice of voters and is a significant
issue to be addressed. The inequality factors include income inequality, gender
inequality, and inequality in the status of other minority groups. 

 

Because
gender inequality exists just like income inequality, the probability of gender
inequality influencing an election outcome is also high. Coffe et al (2014)
explain that it is the support of equality that was widely accepted, giving
rise to gender equality rights, income equality rights and the equality and
respect for minority groups. Even today, these groups still demand more
recognition from the state and the desire of achieving their demands can
influence their party choices. Therefore, this research tries to establish the
relationship between voters’ social-economic status and the party they voted
for (their party choice) in the 2017 election. Because of the nature of the
issue, gender inequality might also influence voters’ choices. Gender is used in this study as an
interfering variable in the relationship between social-economic status and
party-choice.

The first question was designed to establish the relationship (if any) between voters' socialeconomic status and their party choice. Only by establishing the relationship, then the researcher can work out if there is influence of voters social economic status on their party choice, while the second question was to find out if there was gender interference in this relationship.

Hypothesis
There are several theories (i.e. party identification, social approaches, structural etc…) that can be used to analyse the participants' party choices but one that can best explain how the socialeconomic status of the participants influences their party choices is rational choice theory. Scott (2000) explains rational choice theory as when individuals are motivated to achieve their goals, desires, and interest. Therefore in the context of voting, individual voters will choose the party that best represents their interests. So, applying that theory, it can be generally assumed that citizens with a low social-economic status will vote for left-wing parties (i.e. Labour, and the Green) because these parties' policies favour people of low economic status specifically in terms of wealth redistribution (Coffe et al., 2014). However, people with a higher economic status would oppose that idea and would vote for right-wing parties (i.e. the National, and ACT) to maintain their individual interests. In this research, gender was used as an extraneous variable to see if it has some influence on the relationship between social-economic status and partyvotes. If it does, it will sway the voters with low economic status (low income) to vote for right wing parties, and high economic status (high income) voters to vote for left wing parties as shown in the conceptual framework.

Method of Data Collection
The data used were from the New Zealand Election Study (NZES) in 2017. NZES data was made available on the students' canvas by the Statistics and Data Analysis for Policy, a course offered by the University of Auckland. The data was collected from a set of self-completion questionnaires posted to randomly selected New Zealand registered voters and followed up by phone interviews and online surveys (Vowles, 2018;New Zealand Election Study, n.d).

Data Selection and Processing
In this study three variables were used: i.e. household income to measure the participants' socialeconomic status, as an independent variable; party vote which reflects voters' party choices, as a dependent variable; and gender as an extraneous variable. These variables were chosen because the main purpose of this study was to establish the relationship between participants' social-economic status and their party-choice and to identify if social-economic status has some influence on the participants' party choice. Gender is only used as an extraneous variable because it might have some influence in this relationship.
For the purpose of data analysis and processing, gender, as a dichotomous variable, was given (1) for male and (2) for female. Social-economic status was measured using the aggregated annual family (household) income by asking the respondents to indicate their income level on the questionnaire, i.e. For voters' party-choice, the participants were asked to indicate which party they voted for in the 2017 general election i.e. Labour (1), National (2), Green (3), NZ First (4), ACT (5) Maori Party (6), Mana Party (7), The Opportunity Party (8), Another Party (Other Parties) (10). When processing the data, the respondents who didn't know the political parties they voted for were removed from the analysis and the parties with very low numbers of votes were amalgamated into Other Parties. All these details were entered into SPSS and analysed using RStudio.

Results
A total of 3,465 people participated in this study. From this sample, 43.5% (n=1507) were males while the other 56.5% (n=1958) were females. This result shows that there are more female participants in this study.
In addition, the study showed that most participants voted for the National Party (39.6%, n=1 294). The second highest was the Labour Party with 37.6% (n=1 226). The next party after Labour was the Green Party with 8.2% (n=268), then the NZ First Party with 7.5% (n=246), The Opportunities Party followed with 3.4%, (n=110), next was the Maori Party with 2.7% (n=88) and finally the Other parties with 1.2% (n=38). The information revealed that most of the participants chose the National Party. Even though Labour received fewer votes compared to National, the election was extremely competitive between those two parties. All the other parties were significantly less popular with the voters, with each receiving less than 10% of the votes.
The research also revealed that most participants were medium income earners (37.5%, n=1119) and about 36.5% (n=1075) of the participants were high income earners while 26.4% (n=786) were low income earners. This data means that majority of the voters were from an average economic status. Voters with a high economic status were represented by a significantly high number while those with a low economic status were less represented in this election.
The data collected from the study also indicated that there was a low to moderate relationship (r=0.1191945) between the household income and the party. The first relationship shows that party votes for Labour and NZ First increased as the participants' household incomes decreased. A high number of Labour votes were from the low household income group (43.5%, n=316), followed by participants with middle incomes (37.1%, n=394), and the least party votes were from the high household income earners (31.3%, n=322). Just like the Labour Party, the NZ First Party received 3.9% (n=40) votes from high income earners, 9.2% (n=98) from medium earners and their highest percentage of votes from low-income earners (10.3%, n=75).
On the other hand, the National Party's votes increased as voters' household incomes increased. National's 48.0% (n=494) of votes were received from the people with high incomes, 39.0% (n=414) from medium income-earners and 31.6% (n=230) from the low income-earners. The Green Party and other parties did not show a clear relationship between household incomes and party votes i.e. the Green's low-income party vote was 8.1% (n=59), their medium income party vote was 7.2% (n =76) and their high-income party vote was 9.4% (n=97). And the low-income vote for the Other Party was 6.5% (n=47), their medium-income party vote was 7.5% (n=80 and their high-income vote was 7.4% (n=76). Even though the Greens and other parties did not have a clear relationship, the general result (including National, NZ First and Labour) showed that there is a positive correlation between household income and party vote which supports our hypothesis that voters with a high social-economic status will vote for a right-wing party (e.g. National) and voters with a low social-economic status will vote for a left-wing party (e.g. Labour).
In relation to the gender votes, more females voted for the Labour Party (41.2%, n= 757) and the Green Party (8.6%, n=159) compared to males (Labour: 32.6%, n=455; Green: 7.2, n=101). On the other hand, most males voted for the National Party (42.7%, n=585), the NZ First Party (9.5%, n=133) and the Other Parties (9.0%, n=126) when compared to the females i.e. National 38.2% (n=702), NZ First 6.1% (n=112) and Other Parties 5.9% (n=109). This may mean that Labour and the Greens (left wing parties) attracted more female voters than the male voters who showed a voting preference for NZ First, National Party and Other Parties.
In addition, there were more females with low levels of household income (22.9%, n=476) compared to males (22,7%, n=303), while there were more males in the medium (39.1%, n=552) and the high income levels (38.2%, n= 551) when compared to the females i.e. Medium level 36.3% (n=591) and high level 34.5% (n=563). This means that most females are of lower socialeconomic status compared to most males.
When all variables (income, gender, and party votes) were analysed together, the research revealed that almost equal numbers of males (42%, n= 115) and females (43.9%, n=196) at lowincome levels voted for the Labour Party. However, at the medium income level more females (42.3%, n=239) voted for Labour compared to males (30.9%, n=152). As with the medium-income earners, more females (35.9%, n= 194) voted for the Labour Party than males (26.2%, n= 128) at the high-income level. In contrast to Labour, National received an almost equal number of votes from males (32.8%, n=87) and females (32.1%, n=143) at low-income levels but this increased significantly among the males of medium (n=40.9%, n=201) and high-income status (50.8%, n=248). Even though National's female vote continued to increase amongst the medium (37.3%, n=211) and high income (45.6%, n=245) earners, it remained low compared to the male vote.
These analyses show that most people of low-income status, and females (despite their income status), voted for Labour and the Greens while most people of high-income status, and most males, voted for National. Even though NZ First and Other Parties have their votes evenly spread across different income levels, it is clear that they attracted more male voters than female voters. Therefore, gender may have some influence on voters' party choices because a good number of female high social-economic status voters voted for left-wing parties (opposite to our hypothesis) while a handful of males of high social economic status voted for the left wing Green Party compared to the females. Therefore, the deflection of the voters' choice to the right or to the left (as shown in the conceptual framework) is believed to be caused by gender interference.

Discussion
This study reveals that there is a significant (p=<0.01) low to moderate correlation (r=0.1191945) between the social economic status (household income) and voters' party choices (party-votes). This means that one variable does exert its influence over the other variable. In the context of this study, it is believed that it was the social-economic status of the voters that influenced the party-choice, demonstrated through their votes. This idea was supported by Afonso, (2016) indicating clearly that people with different levels of income will influence the voting outcome. It can be argued that it is the voters' party choice that influences the voting outcome however Coffe et al. (2014) indicate that income has a high chance of influencing voters' party choices because of its direct impact on the voters and is considered as an independent variable compared to the party-vote.
In addition, the study confirms the hypothesis that voters have rational behaviour. This is because the majority of voters with low social economic status (43.5%, n= 316) voted for the leftwing Labour party which represents their interest of wealth redistribution. On the other hand, the majority of voters with high social-economic status (48.0%, n=494) voted for the National party which represents their interest in lowering taxation rates as a mechanism to boost private enterprise. Even though other parties (including the Green Party) might not display a clear relationship, possibly due to the interference of other factors, the relationship between the Labour Party and the National Party was noticeably clear: as income increases, National votes increase (i.e. low, 31.6%, n=230; medium 39.0%, n=414; and high 48.0%, n=494) and as income decreases, Labour votes increase (i.e. high 31.3% n=322; medium 37.1%, n=394 and low income 43.5%, n=316) as discussed in rational choice theory. NZ First is a centre party that gained most seats from the low-income earners (10.3%, n=75), medium (9,2%, n=98) and least from high income earners (3.9%, n=40). This may be due to its support and its coalition with left-wing parties, specifically Labour party.
Moreover, there was some gender interference in the relationship between social-economic status and party choice. The study shows that even though there is some relationship between social-economic status and party choice, more women prefer left wing parties despite their economic status. For example, the ratio of Labour male to Labour female votes is: i.e. low income 42.0%: 43.9%, medium 30:9%: 43.3%, high income 26.2% :35.9%. This indicates that Labour not only attracts women voters of low social-economic status, but women of high social-economic status too, whereas most males prefer right wing parties. For example, the ratio of male to female party-votes in National is: low income 31.8%: 32.1%, medium income 40.9%: 37.3% and high income 50.8%: 45.6%. Notably, at low-income levels, there is a higher female to male ratio, but the difference is extremely low. It is not surprising for women to choose the left wing Labour Party because its polices not only favour low income earners but are possibly more favourable to women. It is no coincidence that the Labour Women's Council was formed to assist women in joining the party and to meet the needs of the women across New Zealand's society. That could be a reason for women voting for the Labour Party. This indicates that, even though there is significant correlation between the social-economic status of voters and their party choices, gender also has some influence on this relationship and does interfere with the results by deflecting voters' choices either to left-wing or right-wing parties despite their social-economic status.

Conclusion
Finally, this study reveals that there is a low to moderate relationship between social-economic status and party choice. It also agrees with the hypothesis that voters are rational beings, therefore the participants with a high economic status voted for right-wing parties while those with a low social-economic status voted for left-wing parties. There was gender interference in this relationship deflecting some of the voters' choices. Therefore, it can be concluded that the social-economic status of citizens did influence the party choice of voters in New Zealand's 2017 election with some gender interference.