For decades political scientists agreed that politics in New Zealand was nearly exclusively either concerned with economic issues or based around the left-right class divide. That class and a basic economic cleavage underpin the way New Zealand politics is carried out has become an almost unchallenged assumption for some. This blog post discusses how and why the left-right class cleavage is in decline in NZ parliamentary politics. [Read more below]
The establishment of the NZ party system
The existence and operation of the major parties in New Zealand have traditionally been closely aligned with class interests. This was especially the case after the 1880s, during which a decade-long depression politicised urban manual workers, small farmers, and the landless, leading to the election of the Liberal Party Government in 1890. Prior to the establishment in 1890 of the Liberal Party, which was the first ‘modern’ political party, most ‘parties’ were simply parliamentary factions ‘based on sectional or issue differences rather than class differences’ (Robinson, 1967: p.100).
Nonetheless since the establishment of central government in 1856 these quasi-parties were dominated by the landed oligarchy. Liberal Party politicians, in contrast, were particularly aware of the different economic and political interests existing in the new nation and found advantage in taking up the side of the poor under universal suffrage. For example, prior to becoming Liberal premier, Richard Seddon clearly outlined the ideological division in New Zealand when he said, ‘It is all nonsense to talk of Liberalism and Conservatism in New Zealand. It is the rich and the poor; it is the wealthy and the landowners against the middle and labouring classes’ (quoted in Lipson, 1948: pp.68-69).
With the arrival of the Liberal Party, working men – and later women – were finally involved in parliamentary politics and elected to Parliament. From this moment on, class was properly part of New Zealand politics, with both sides of the class divide represented in Parliament. This was reinforced when the Reform Party formed on the right of the Liberals in 1909 and was dependent upon small North Island farmers and urban businessmen. The Liberals’ support base later shifted towards commercial and manufacturing interests (and the party changed its name to United), and thus the Labour Party emerged as the voice for trade unionists. Then, following the election of Labour in 1935, the National Party was formed out of the combination of the United and Reform parties (See: Robinson, 1967: p.100; Carter, 1956: p.90; and Gustafson, 1986). By the 1930s then, partisan competition in New Zealand was predominately uni-dimensional – organised around economic issues and class divisions, with the result that social structure played a crucial role in determining party support.
As elsewhere in the Western world the New Zealand parties’ core voter bases in the 1930s came to provide their solid long-term backing. These social linkages have been theorised as representing a ‘frozen party system’ – the idea that defining moments in the history of a country produce social cleavages which become politicised, encouraging the emergence of party systems which reinforce and reproduce these cleavages. This was the theory developed in Seymour Martin Lipset and Stein Rokkan’s influential analysis of 1967. They argued that political parties that establish their dominance in crucial periods are often able to retain their positions in the party system even after the social cleavages change. They do this by their skilful adaptation to that change, incorporating new cleavages into the original dimensions of conflict. Subsequently, for fifty years New Zealand politics orientated to the basic socioeconomic cleavage in which Labour and National were in dynamic competition.
The dominant left-right socioeconomic cleavage
For decades political scientists agreed that politics in New Zealand was nearly exclusively either concerned with economic issues or based around the left-right class divide. Leslie Lipson also noted in 1948, ‘Politics in this Dominion are concerned with economic problems to the exclusion of almost all others. Social and moral issues, it is true, have at times entered into the field of party warfare, but none of these has ever dominated political controversy’ (Lipson, 1948: pp.196-197). Writing in 1969, Austin Mitchell had a similar conclusion, arguing, ‘the social cleavage in fact dominates the political scene in unrivalled fashion. Without contrary pull, social self-interest becomes the main determinant of voting behaviour’ (Mitchell, 1969a: p.30).
Half a century later, in the 1996 and 1999 elections, Fiona Barker and Elizabeth McLeay found that voters still overwhelmingly chose political parties based primarily on socio-economic issues (Barker and McLeay, 2000: p.148). According to Barker and McLeay, ‘In the 1990s, New Zealand voters’ primary concerns were with the economy, unemployment, and related social concerns such as health care and education’ (Barker and McLeay, 2000: p.145; see also: Clements, 1982: p.151; Aimer, 1997: p.193; and Nagel, 1998: p.231). Miller has also said that ‘New Zealand lacks the social cleavages found in many other multi-party systems’ (Miller, 1997a: p.39).
Jonathan Boston et al. (1996) also concluded that Labour and National ‘still derive their distinctiveness primarily from socio-economic and related issues’ (Boston et al., 1996b: p.58). Likewise, according to Aimer, ‘one issues dimension – the socioeconomic dimension – over-shadows all others’ (Aimer, 2001: p.277). It is also noteworthy that in his 1994 multi-country study, Arendt Lijphart recognised only the socioeconomic class cleavage operating in New Zealand (Lijphart, 1994). Miller has recorded that ‘Lijphart has argued, only the ethnic cleavage between Pakeha New Zealanders and the indigenous Maori population... prevent New Zealand from being regarded as a totally homogenous society. The party system, Lijphart claimed, was "one dimensional", with the socio-economic factor being the key point of social distinction’ (Miller, 1997a: p.39). According to Aimer, Lijphart regards New Zealand as having ‘pronounced unidimensional tendencies’ shared in strength only by Barbados and the Bahamas (Aimer, 2001: p.277; see also: Lijphart, 1984).
Matthew Gibbons, in his content analysis of election manifestos between 1911 and 1996, found that the left-right dimension (which he says measures economic, redistribution and foreign policy differences) structured party competition throughout the period (Gibbons, 1997: p.16). The results of Thomas Brechtel and Andre Kaiser’s 1997 survey of experts also reinforced the idea that the socioeconomic dimension dominated the party system – just over half of the academic respondents classified New Zealand politics as uni-dimensional, and the survey as a whole gave the left-right dimension a ranking of four out of five on a scale of relevance (Brechtel and Kaiser, 1999: p.7).
That class and a basic economic cleavage underpin the way New Zealand politics is carried out has become an almost unchallenged assumption for some. This is understandable, as for many decades its influence was overwhelmingly obvious, especially in that National electorates tended to be those with a high proportion of employers whereas Labour held electorates that tended to have a low proportion of employers. According to Vowles, Labour electorates were also often those ‘with a higher proportion of people employed in manufacturing than in professional, administrative and other skilled employment’ (Vowles et al. 1995, 16-17; see also: Mulgan, 1997a). Furthermore, key party debates and differences clearly revolved around socioeconomic or materialist issues.
Class dealignment in New Zealand
Traditionally political scientists have therefore viewed party politics in advanced industrial countries like New Zealand as existing in ‘alignment’ – a term referring to the association between the parties that people vote for and the social groups or categories to which these voters belong. A class alignment is therefore one in which the working class votes overwhelmingly for left-wing and the middle class votes overwhelmingly for right-wing parties. A second element of the concept of alignment is the idea of some rigidity or loyalty underpinning voting – that parties have party supporters (voters who actually identify with the party they vote for) rather than just people who vote for the party from time to time.
Throughout advanced industrialised nations political scientists have observed a breakdown in this alignment over the last four decades – a process and result that they label ‘class dealignment’. In New Zealand, this class dealignment is highlighted when the most widely used method of analysing voting along class lines, the Alford index, is applied. This index is calculated by subtracting the proportion of the middle class – defined in this instance by the non-manual household vote (including farmers) – that vote for the main left party (Labour) from the proportion of the working class – defined here as the manual household vote – that also vote for that party. A 100% score would occur only where all the working class and none of the middle class voted for Labour, while a zero score would indicate that the middle class vote for Labour in an equal proportion to the working class. According to Vowles’ calculations, in New Zealand the Alford index has dropped from 30% in 1963 (at a time when dealignment was already underway) to only 13% in 1999, after falling to only 5% in 1990. [Ganley says that, ‘Even if we include the Alliance, as the scale was originally based on all left parties not just the largest, the figure increases by only 0.9, still nothing like the earlier figures’ (Ganley, 1998: p.12). Mitchell also applied the Alford index to statistics from his 1962 Dunedin Central voter survey, arriving at a figure of 42% (Mitchell, 1967: p.6).]
Alford’s Index of Class Voting in New Zealand
Year Percentage
1963 30%
1975 19%
1981 16%
1984 15%
1987 9%
1990 5%
1993 10%
1996 9%
1999 13%
Source: Vowles (2001: p.176).
There is a consensus within the contemporary political science literature that the influence of class on voting has been lessening since the 1960s, and many commentators now question its usefulness for understanding voting behaviour. According to Aimer, ‘The weak and weakening relationship between "class" (i.e. occupation) and voting is by now one of the firmly-established facts of New Zealand electoral politics’ (Aimer, 1992: p.333). Vowles maintains that although membership of social groups ‘has an influence on voting choice in New Zealand’, the ‘relationship between structural influences and political parties in New Zealand is complex, and weaker than many sociologists might expect’ (Vowles, 1994c: p.191). Gold agrees, claiming that recent trends indicate that ‘social differences do influence party choices in New Zealand, but only to a fairly modest degree’ (Gold 1992, p.489). According to Richard Mulgan, too, ‘there has always been a strong tendency for significant numbers of voters to vote against what might be thought of as their normal class interests… and the number of such voters appears to have been steadily increasing since the 1960s’ (Mulgan, 1997a: p.274). All of this is not to say that the social class cleavage no longer plays any part in structuring electoral outcomes, but that it does so to a much lesser degree than in the past.
Academics sometimes ask whether the party system will revert back to its class alignments, again making the socioeconomic cleavage more universal and central. Some writers seem to think so, on the basis that dealignment is automatically followed by realignment and that there is a dynamism in party politics that automatically reinforces and reproduces cleavages of conflict. This is certainly possible, but there is no reason to believe that it is inevitable. Instead it seems more likely that what is occurring in New Zealand is ‘the onward march of party decomposition’ (Gustafson, 1993: p.73).
It remains to be seen whether any substantial class realignment is possible in New Zealand society. This series of blog posts argues that this is unlikely because the political parties are not encouraging an economic/class cleavage by campaigning on those issues. Nor is such a political cleavage reflecting an active and overt social cleavage any longer. Instead, other social and political cleavages are becoming more prominent.
Politicised class conflict replaced by depoliticisation and atomisation
There is no doubt that the class alignments created in the political system decades ago between party and society still remain partially frozen in the party system today, but it is also clear that there is less of a politicised social conflict that kindles rivalry between the parties. Instead, New Zealand society has been growing less cohesive and more individuated in recent years. This is part of a trend throughout advanced industrial democracies. Everywhere, collective organisations such as churches and unions have lost societal influence. The trend of individualism is in the ascendant and is at odds with the essentially collective nature of political parties. According to Bently et al., ‘The rise of individualism, the values of the 1960s, television, economic change and a host of other factors are blamed for fostering an atomised world in which people feel less social connection and less interest in common issues or collective solutions’ (Bently et al., 2000).
There has also been a noticeable decline in class consciousness in New Zealand. That which made individual workers and those in other classes feel that they were part of a particular section of society has largely gone. The fact that classes still exist in New Zealand cannot be disputed, but for various reasons (including the breakdown of the political traditions which have prevailed during the past century), fewer people identify as being members of their class. This is especially the case for the working class, which is made up of people who increasingly experience life and politics as individuals, rather than as members of a class with any collective interest and strength. Because political loyalty was at its fiercest in class-based constituencies, the breakdown of communities means this loyalty is declining.
This depoliticisation of class is seen in the declining level of traditional working class militancy. One indicator of this decline is the sharp fall in strike activity. Whereas in 1986 there were 1,329,054 ‘person days of work lost’ in the New Zealand economy, by 1997 this figure had dropped dramatically to 24,614 days (Statistics NZ, 1998: p.323). Another indicator of the same general phenomenon is the drop in union membership. While in December 1985 there were 259 trade unions and 683,006 union members (43.5 percent of the work force), by December 2002 there were only 174 unions representing 306,687 members (21.7 percent of the work force) (Street, 2001: p.355; Dearnaley, 2003). It is also noticeable that there has been relatively little class mobilisation in the streets in recent years. Despite the incredible reforms of the 1980s and 1990s there was remarkably little participation in protest. Even Jim Bolger noted with surprise the lack of protests that occurred on the streets in response to his government’s reforms (Bolger, 1998: p.55; see also: Trotter, 1997r).
The following blog posts illustrate the decline of the left-right cleavage by examining each party’s class bases, and showing how this change affects the party.
[This blog post will be updated with new data - any feedback and information is appreciated]
