New Zealand politics have always been influenced by the spatial cleavages in society. These are seen in two ways: regional cleavages and the urban-rural cleavage. It seems likely that the decline in the significance of class as a determinant of voting has meant that the geographical cleavage in particular has grown in relative importance in structuring party politics in New Zealand. [Read more below]
The regional cleavage in New Zealand politics involves the political differences between regions in the country, such as the North and South Islands, while the urban-rural cleavage involves the political differences between those living in cities, towns and rural areas. During the nineteenth century, parliamentary politics were strongly affected by regionalism, but this declined due to the abolition of the provinces in 1876, the establishment of strong national political parties after 1890, and increased immigration and internal migration.
The major parties, however, continued to have a geographical bias in the twentieth century, with Labour being strong in Wellington, Christchurch and the South Island West Coast, while National has been particularly successful in the central North Island. Writing in 1979, Levine argued, ‘It is important to recognise that there is a strong (and growing) regional basis to New Zealand politics’ (Levine, 1979: p.91). More recently New Zealand First’s strongest support base has been in the northern North Island (Mulgan, 1997a: p.275), and in 2002 the Outdoor Recreation Party won 5.4% of the party vote in the West Coast-Tasman electorate (despite winning only 1.3% nationally). Overall, however, New Zealand now lacks any deep regional divisions. According to Mulgan, the geographical cleavage therefore plays only a very small part in affecting an individual’s voting habits (ibid).
In contrast to the * regional * aspect of the geographical cleavage, the * rural-urban * nature of politics is much more important – particularly in the past, with Labour predominating in the mainly urban electorates, and National being more successful in the provinces, rural towns and scattered settlements. To a degree, the urban-rural cleavage has been a consequence of the class cleavage because rural areas tend to have a large number of employers, due to the fact that most farmers are self-employed.
Political geographer RJ Johnston asserts that in New Zealand you can better predict how people will vote by knowing where they live than by knowing what class they are: ‘The analysis of the geography of voting suggests that urban-rural and inter-island cleavages have increased during the post-war era…. producing a country that, in electoral terms, is spatially more polarized than ever’ (Johnston, 1992: p.47).
It seems likely that the decline in the significance of class as a determinant of voting has meant that the urban-rural geographical cleavage has grown in relative importance in structuring party politics in New Zealand. Electoral geographers Rex Honey and Ross Barnett have gone as far as declaring that at the 1987 general election, ‘the rural-urban dimension supplanted the socio-economic one as New Zealand’s pre-eminent voting cleavage’ (Honey and Barnett, 1990: p.87). They noted that the ‘rural-urban split became even more pronounced’, as ‘National solidified its margins in the countryside but almost disappeared in the major cities’ (Honey and Barnett, 1990: p.87).
Vowles’ analysis of NZES survey research from the 1996 general election showed that although the urban-rural cleavage continues to be significant, it is more so for some parties than others. He found that while support for National and the Alliance was evenly spread across the urban and rural categories, the other parties had greater variations in support:
>Labour did a little worse in three main centres and in the provincial cities, but significantly worse in the rural areas. New Zealand First did best in the towns and the rural areas, Act best in the rural areas and significantly worse in the provincial cities and towns (Vowles, 1998d: p.40).
In terms of geography Act’s strongest base of support in that first MMP election was in Auckland (Fraser and Zangouropoulos, 1997: p.52). Act made ‘inroads in the affluent, mainly National-held seats of Auckland. All seven seats where Act gained more than 10% of the party vote were in Auckland’ (Luke, 1998c: p.15). While Richard Prebble won the electorate of Wellington Central, in terms of the party vote Act did not do nearly as well.
This geographical bias might stem from the fact that the party was formed in Auckland, mainly by Aucklanders. As well as performing well in cities like Auckland, the strongest support for Act was found in rural areas, although the party was also said to have performed ‘significantly worse in the provincial cities and towns’ (Vowles, 1998d: p.40). According to Act’s own interpretation of the election results, while the party ‘was strongest in urban areas... [it] also gained a credible share of the rural vote’ (Fraser and Zangouropoulos, 1997: p.53).
In the 1999 election campaign it was obvious that the geographical cleavage was a much politicised one, and a pronounced political contest existed between urban and provincial New Zealand. Many political parties, and especially Act, made a strong play for the provincial heartland, as National’s former strong grip on it was weakening as the party came to be more perceived of as an urban party. Many commentators suggested that the growing importance and size of urban New Zealand has been pushing National to become an urban-oriented party much like the Australian Liberals, leaving Act to mop up the provincial vote, much like the rural-based rightwing Australian National Party.
However, at the 2002 election it was National’s urban vote that collapsed significantly, while Act continued to do relatively well in metropolitan Auckland rather than rural areas. According to the 2002 Herald-DigiPoll, ‘Since the 1999 election, National has lost more than a third of its voting share in Auckland, Hamilton, Wellington and Christchurch…. [while] Act confirmed its previous showing as a city party, scoring most strongly in Auckland’ (Collins, 2002). In the 2002 general election, National’s urban vote collapsed significantly, especially in Auckland, where the number of National voters almost halved from 166,000 to 86,000 (ibid). New Zealand First support is also not so evenly dispersed when it comes to the geographic cleavage. In 1996, according to Vowles, New Zealand First drew ‘more on rural electorates, and less from working-class electorates’ (Vowles, 1998d: p.33). In that election, ‘New Zealand First did best in the towns and the rural areas’ (ibid: p.40; see also: Mulgan, 1997a: p.275).
Although it might appear that the geographical cleavage is becoming more significant, this is possibly only in relation to the decline of the class cleavage. It seems likely that these spatial dimensions have in more recent years probably decreased in importance. Certainly the introduction of MMP is likely to have further pushed party politics away from a geographic-orientated nature because the all-important party vote is not geographically-related, and parties are therefore inclined to spend less attention on winning marginal electorates. Instead, social constituencies (such as professionals, trade unions or ethnic groups) are more important to the parties.