Previous Next

Abbott’s Gambit: The 2013 Australian Federal Election

15. Regional Place-Based Identities and Party Strategies at the 2013 Federal Election

Geoff Robinson1

During the prime ministership of Julia Gillard themes of place-based identities were prominent in Australian political discourse. At the 2010 federal election Labor’s two-party-preferred vote reached historic highs in Victoria, South Australia and Tasmania but fell sharply in Queensland and Western Australia. During the last year of the Gillard Government observers competed to produce more gloomy scenarios of a Labor collapse both in states such as Queensland and Western Australia and ill-defined regions such as ‘western Sydney’ (Kenny 2013; Shanahan 2012). On election night, 7 September 2013, these predictions were largely unfulfilled. The final result demonstrated a general, rather than regionally specific, estrangement from Labor. The largest swings against Labor were in the states where the Party had performed best in 2010 (see Antony Green’s chapter in this volume, Chapter 23).

The 2010 election had been one of firsts: a female prime minister, the near-defeat of a first-term government and an unprecedented vote for the Greens. There were also exceptionally high levels of divergence in electoral behaviour between the states. Across the next three years Australian parties sought to learn the political lessons of 2010. Labor sought to shore up its base against erosion to both the left and right, the Coalition refined their (almost) winning appeal and worked to make Tony Abbott less of the (slight) liability he had been in 2010 (Bean and McAllister 2012: 352). The Greens searched for new constituencies, such as farmers, beyond the urban left they had attracted in 2010. The final result of the 2013 federal election demonstrated that the regionally-based estrangement from Labor apparent in 2010 had become a national disenchantment, but that Labor had some success in defending its base. The Greens found themselves pushed back to their heartland in inner Melbourne while Clive Palmer surfed a wave of political disillusion.

Political regionalism in Australia

This chapter focuses on three lines of regional division in Australian politics: (1) the states and territories; (2) mining and agricultural regions; and (3) the alleged division between ‘western Sydney’ and inner Melbourne and Sydney. It complements the analysis in other chapters in this volume (see Chapters 14 and 23 by Jaensch, Miragliotta and Wear, and by Green, respectively) by unpacking in more depth the nature of the regional differences that underlie divergent voting patterns. This chapter argues that there are three distinct forms of ‘regionalism’: social, economic and cultural.

Social regionalism

Social regionalism refers to the fact that the divergent electoral behaviour of different regions is often simply a reflection of their different social composition. Journalistic analyses of regional divergences in Australian politics frequently cite the variant levels of party support in different states as evidence of ‘regionalism’ (Colebatch 2013a). It is not surprising that Australian states have different levels of party support or that the Greens poll better in inner Melbourne than outer western Sydney. ‘Marginal’ electorates may be in the suburbs and regions but this does not indicate, contrary to the arguments of some commentators, that there is a distinct ‘suburban’ or ‘regional’ identity (Henderson 2013; The Australian 2014).

Australia, compared to other federations, is notable for the low level of political divergence between the states and territories. There is no Australian equivalent of Scotland or Alberta, subnational territories in which one side of politics is entirely powerless. Even Western Australia, currently Labor’s graveyard, had a state Labor government from 2001 to 2008.

Democratic politics is mostly about a competition by parties to provide voters with the same public goods, even if voters disagree about which goods to prioritise and their evaluation of the competence of different parties to provide them (King et al. 2008). The uniformity of political behaviour in Australia exemplifies this principle. In Australian democracy voters mostly want the same public goods, such as rising living standards, national security, defence of borders and high quality government services. Voters do diverge in the weight that they place on particular goods and in their estimate of the competence of different parties to provide them. The Australian colonies entered federation with highly variant party systems and divergent levels of enthusiasm for the project of federation itself (Parker 1949). After 1901, Australian voters demonstrated a uniform preference for the public good of racial exclusion and national development. Class politics stabilised the system, workers in Melbourne and Perth largely valued the same public goods and trusted in Labor to supply them. After 1910, the portion of voters socialised in childhood to identify with one party or the other steadily increased and this added a further element of stability to the system (Aitkin 1982).

Economic regionalism

Economic regionalism describes the fact that the geographical concentration of a particular industry may lead voters in a region to develop a distinct sense of shared economic interest. This pattern first clearly emerged with the rise of the Country Party in the 1920s. Farmers were a small portion of the electorate but the new party drew on an ideology of ‘countrymindedness’ to convince town residents and many rural workers that they shared common interests with farmers (Aitkin 1985).

The debate about the taxation of the mining industry at the 2010 election revived a modern form of countrymindedness. Many voters in Queensland and Western Australia resented the claims of other states to share in mineral wealth. The number of electors employed in capital-intensive resource industries, such as mining or forestry, is quite small, but voters tend to overestimate their importance as employers and are also prone to accept claims about the dependence on them of other industries (Blainey 2013; Essential Media Communications 2012; Schirmer 2012). This chapter will show that there was a distinct electoral reaction against Labor in mining regions at the 2010 election and that this was maintained in 2013.

Cultural regionalism

A third form of political regionalism, here called cultural regionalism, could emerge if significant groups of voters come to question the legitimacy of the state as a provider of public goods. Secessionism is one form of this but the debate about asylum seekers has provided another challenge to state legitimacy. Australian governments have competed to prove their credentials as providers of ‘border protection’ in recent years. For some voters ‘border protection’ has a negative value. It is a positional rather than a valence issue. The rhetoric of ‘culture wars’ is hyperbolic but it does pose the question of whether voters in different regions in Australia have developed radically divergent views about the role of government. In the 1940s, voters in Toorak and Mosman shared similar material aspirations with those of Richmond and Redfern, although they disagreed about how to achieve them. Is it now the case that voters in the ‘left-green’ and sometimes highly ethnic inner suburbs such as Fitzroy and Marrickville fundamentally diverge in their aspirations from outer suburban locations such as Knox and Penrith?

Kirsten Phillips (2009: 123) has argued:

The modern state … can tolerate and ‘cultivate’ a population which includes even bodies of a ‘different race’ as long as ordering mechanisms for managing this difference are upheld … Of key importance is the fact that migrants enter the nation-state under the control and surveillance of that biopolitical state, and can be known, categorised and positioned in the social hierarchy effectively.

The debate about asylum seekers in Australia has challenged the biopolitical state and has destabilised political divisions formed around this state.

The emergence of regionalism in Australia

Economic and cultural regionalism emerge out of a process of social interaction. The holders of power within social networks—those with stocks of social capital—are likely to be able to shape local attitudes. Group interaction reinforces radicalism or conservatism within a locality. The political conservatism of Australian rural areas owes much to the fact that local opinion leaders and community activists are nearly all supporters of the Liberal and National Parties. Strong levels of left party support in mining regions have reflected a similar process (Butler and Stokes 1976: 84–6; Taylor and Johnston 1979: 221–69; Ward and Verrall 1982). As class-based party allegiances weaken, group influences may become more important for voters (Johnston and Pattie 1988). The British Conservatives now poll relatively worse in northern England than they did when it was an industrial heartland (Kellner 2013). A distinct ‘northern’ political identity has emerged within England that did not exist during the post-war heyday of class politics, perhaps in response to the impact of Thatcherism and deindustrialisation. A distinct social milieu may even attract particular sorts of voters. There is some evidence from Australia and the United States that gays and lesbians may migrate to regions that they regard as more sympathetic (Tam Cho 2013; Davidowitz-Stephens 2013; Mansillo 2013a: 23–8; Sides 2013).

In Australia it is the ‘suburbs’, in particular ‘western Sydney’ and the ‘inner city’, that have dominated recent discussions of political regionalism. In the 1950s and 1960s some cultural commentators attributed Labor’s woes to the relocation of working class voters from the tight-knit communities of the inner city to the outer suburbs (Scalmer 1997). This interpretation was challenged by the rise of Gough Whitlam and his appeal to outer suburban voters. Even in 1975 and 1977 Labor held many outer suburban electorates, such as Robertson, which it had first won in 1969 (Forward 1976: 18).

After 1996, regional divergences became a central theme in much Australian political discourse. Initially the focus of commentators was on rural disaffection and the spectre of One Nation, but as the Pauline Hanson phenomenon ebbed the focus turned towards the ‘suburbs’, in particular ‘western Sydney’ and most of all the Penrith-based electorate of Lindsay. The proponents of a suburban focus such as Mark Latham sometimes cited the suburbs as evidence of a broader ‘middle classing’ of the Australian population—an example of what this chapter has defined as social regionalism. On other occasions Latham adopted a cultural regionalist approach and cast the suburbs as the base of a distinctive popular conservatism opposed to the social liberalism of the inner city (Latham 2013: 48–51).

Evidence for regionalism

This chapter will now consider the patterns of regional distinctiveness in Australian politics at the 2013 election. This analysis has two components. The first reviews how state-level electoral patterns have changed over time and the extent to which local variations in economic performance may have contributed. The territories are excluded from this analysis due to their small number of electors. The second analysis develops a simple linear regression model that seeks to predict party support from the social characteristics of electorates. The residuals from this model are then examined as possible evidence of regional patterns.

State regional patterns

In the 1970s and early 1980s some radical scholars, such as Humphrey McQueen, predicted that Australia faced an era of regional polarisation between an industrial core of New South Wales, Victoria, Tasmania and South Australia and the periphery of Queensland and Western Australia. The political culture of the core states would remain that of a residual social liberalism but the periphery, McQueen argued, would be dominated by a new and aggressive conservatism (McQueen 1982: 112–6). This prediction was not fulfilled. In the 1980s, Labor committed itself to national economic restructuring under a leader, Bob Hawke, who unlike Whitlam appealed to voters in Queensland and Western Australia. Tasmanian voters reacted against Labor’s environmental turn in 1983 but eventually fell in behind the Labor ascendancy by the early 1990s.

The 2010 election seemed to represent the regional realignment that radical scholars of federalism had predicted decades earlier. States with a large mining sector swung strongly against Labor, whilst the south-eastern states reacted against Tony Abbott’s aggressive conservatism. In the run-up to the 2013 election some commentators predicted that this regional division would increase and that Labor might lose every seat in Western Australia and Queensland (Coorey 2013). Instead the divergence notably closed. The economic anxiety that had impacted on mining regions in 2010 had now spread nationally.

AG15_Robinson_Fig1.tif

Figure 1: Labor primary and two-party-preferred vote by states—variance 1949–2013

Source: Australian Electoral Commission (2011b and 2013b) and University of Western Australia (2013).

At the state level there is no simple relation between economic performance and political behaviour. Unemployment is an imperfect indicator of state-level economic performance: it excludes discouraged workers and also does not take into account the shift of workers from full to part-time jobs. The relative stability of Australian unemployment during the global financial crisis (GFC) obscured a notable softening in the labour market apparent in withdrawal from the labour market and a fall in working hours. This softening was particularly apparent among blue-collar males. By 2013, less than half of Tasmanian males over 15 were in full-time employment (Cowgill 2013). The heavy swing against Labor at the 2013 federal election in this state probably reflected the strong belief among many Tasmanians voters that at a time of economic uncertainty public policy should support resource industries. It is misleading to extrapolate from the Tasmanian case—as does Tim Colebatch (2013a)—a broader thesis that the state-by-state decline in the Labor vote was largely accounted for by divergences in local economic performance. Overall there was no relation between the decline in male full-time employment by states and the decline in Labor’s vote either during the 2010–13 term or across the Rudd–Gillard years as a whole (Australian Bureau of Statistics 2014).

The response of voters to economic trends is dependent on the political environment. At the 1993 federal election unemployed voters were more likely to support Labor, but at the 1996 election they were more likely to support the Coalition (Charnock 1997: 247–9). In 2010, economically anxious voters in Tasmania may have credited federal Labor for its effective response to the GFC but by 2013 they were focused on the welfare of local resource industries, just as Queensland and Western Australian voters had been in 2010.

Electorate level regression analysis

A federal census was conducted in 2006 and 2011. Federal elections were held in 2004, 2007, 2010 and 2013. In 2005, New South Wales and Queensland electoral boundaries were adjusted. In 2008, boundaries were adjusted for Tasmania, Northern Territory and Western Australia. In 2009, boundaries were adjusted in New South Wales and Queensland. Before the 2013 federal election, South Australian and Victorian boundaries were adjusted (Australian Electoral Commission (AEC) 2011a). Electorate-level census data is available for some areas but not combinations of electoral districts and censuses during this period. For some elections the regression analysis in this chapter is based not on the boundaries in place at that election but for the boundaries in place at a later poll for which census data is available. Social data has been matched to electorates using the Australian Bureau of Statistics Tablebuilder.2

Table 1: Regression dependent variables

Dependent variables

Census

Definition

2004 ALP

two-party-preferred

2006

The 2004 Labor two-party-preferred vote applied to the electorate boundaries in place at the 2007 election as estimated by Antony Green (2007). This adjusted two-party vote for the independent electorates of Kennedy and New England is not available so the actual 2004 two-party-preferred vote in these electorates has been used (AEC 2005).

2007 ALP, Coalition and Green primary votes

2006

The 2007 primary vote applied to the electorate boundaries in place at the 2007 election. The Coalition vote includes the total for Liberal and National candidates (AEC 2008).

2007 ALP

two-party-preferred

2006

Votes by boundaries in place for 2007 election (AEC 2008).

2007 ALP, Coalition

and Green primary votes

2006

Votes by boundaries in place for 2007 election (AEC 2008). The Coalition vote includes the total for Liberal and National candidates.

2010

ALP two-party-preferred

2011

The 2010 vote applied to the boundaries in place for the 2013 election (Green 2013).

2010 ALP, Coalition and Green primary votes.

2011

The 2010 vote applied to the boundaries in place for the 2013 election. The Coalition vote includes the total for Liberal and National candidates (AEC 2013a).

2013 ALP Two-party-preferred

2011

Votes by boundaries in place for 2013 election (AEC 2013b).

2013 ALP, Coalition, Palmer United and Green primary votes.

2011

Votes by boundaries in place for 2013 election (AEC 2013b). The Coalition vote includes the total for Liberal and National candidates.

Sources: As indicated in table.

Table 2: Regression independent variables

Variable

Type of variable

Definition

Mining

Nominal

An electorate with more than 5 per cent of workforce employed in mining

Agricultural

Nominal

An electorate with more than 5 per cent of workforce employed in agriculture, forestry and fishing

Manufacturing

Interval

Percentage of workforce employed in manufacturing

Tertiary qualifications

Interval

Percentage of population with tertiary qualifications

No religion

Interval

Percentage of population with no religious affiliation

New religions

Interval

Percentage of population with religious affiliation other than Judaism or Christianity

Source: Australian Bureau of Statistics.

These dependent variables have been selected to cover a range of new and old drivers of electoral choice. Religions other than Judaism or Christianity prove a more relevant contemporary measure of cultural diversity than older measures such as country of birth or language used at home. Contemporary conservative discourse has emphasised the importance of Australia’s Judeo-Christian antecedents and has argued that this identity has come under challenge from Muslim communities in particular regions such as south-western Sydney (Sheehan 2006).

Table 3: Two-party-preferred Labor vote determinants 2004–133

2004

2004

2007

2007

2010

2010

2013

2013

B

Beta

B

Beta

B

Beta

B

Beta

Constant

19.19

27.21

8.94

9.24

Agricultural electorate

-6.26

-0.25

-6.95

-0.28

-9.85

0.32

-8.80

-0.31

Mining electorate

10.32

0.19

11.57

0.21

-0.87

-0.02

-0.66

-0.02

Manufacturing

0.68

0.25

0.91

0.34

1.59

0.46

1.27

0.40

No religion

0.16

0.09

0.05

0.03

0.44

0.25

0.40

0.25

Non Judeo-Christian religion

0.80

0.45

0.71

0.40

0.58

0.35

0.70

0.47

University qualifications

0.08

0.10

0.05

0.06

0.01

0.01

-0.04

-0.49

Public employment

0.72

0.32

0.65

0.29

0.84

0.35

0.86

0.39

R-Sq (adj)

0.45

0.47

0.48

0.53

Source: Australian Bureau of Statistics.

Table 4: Primary Labor vote determinants 2007–13

2007

2007

2010

2010

2013

2013

B

Beta

B

Beta

B

Beta

Constant

22.24

5.60

6.88

Agricultural electorate

-8.18

-0.32

-9.72

-0.33

-9.45

-0.34

Mining electorate

13.36

0.25

0.31

0.01

-0.05

0.00

Manufacturing

1.10

0.40

1.78

0.54

1.29

0.41

No religion

-0.06

-0.04

0.18

0.11

0.10

0.06

Non Judeo-Christian religion

0.70

0.38

0.53

0.34

0.73

0.50

University qualifications

-0.05

-0.06

-0.07

-0.01

-0.12

-0.16

Public employment

0.63

0.27

0.76

0.33

0.83

0.39

R-Sq (adj)

0.53

0.53

0.62

Source: Australian Bureau of Statistics.

The geographical foundations of Labor support are those that would be expected. This pattern demonstrates the persistence of ‘countrymindedness’, despite the overall decline in agricultural employment and the decline of social capital in rural areas. Strong support persists among both city and country voters for the ideals of countrymindedness and the provision of assistance to farmers (Cockfield and Botterill 2012). The electoral significance of this fact has diminished due to the growth of provincial cities and the urbanisation of coastal regions which have given Labor a stronger foothold outside the capital cities than in previous years (see the preceding chapter by Jaensch, Miragliotta and Wear in this volume). Electorates with a high level of manufacturing employment and those with significant populations of a non-Christian religion remain Labor strongholds. The most notable change in the regional determinants of party support is the collapse in 2010 of Labor’s previous significant strength in mining electorates. This reversal was maintained in 2013.

At the 2010 election those states with a higher manufacturing population had rallied to Labor, in contrast to those on the resource-dependent periphery. By 2013, Australian manufacturing was under siege. The high Australian dollar dramatically reduced the competiveness of local industries. Labor had hoped that the mining tax would raise revenues that could assist regions and industries that were under pressure from the high dollar and the general increase in costs associated with the mining boom. The tax came with a heavy political cost but then failed to raise the revenue predicted. Manufacturing remained under pressure (Garnaut 2013: 123, 184). The Coalition challenged Labor in its manufacturing heartlands; it highlighted the alleged impact of the carbon tax on manufacturing, advocated stronger measures against ‘dumping’ and complained vociferously when in July 2013 the Rudd Government amended the fringe benefits tax on cars (Ferguson and Hannan 2013; Kelly 2013). In April 2013, Labor’s primary vote fell sharply at a state by-election in south-eastern Melbourne, an area with a significant manufacturing workforce. This reversal was cited as evidence by some journalists that Labor faced particular challenges in regions with a high manufacturing population (Gordon 2013). During the 2013 election campaign some analysts predicted special difficulties for Labor in electorates, such as Corangamite, situated in regions where the car industry was (or was believed by voters to be) a significant employer (Robinson 2013).

Despite these concerns there was little evidence of a backlash against Labor in manufacturing areas comparable to the backlash it experienced in mining regions. It is an overstatement to suggest, as did one journalist, that the changes to the fringe benefits tax played a large role in the swing against Labor in Victoria and South Australia (Allard 2013).

At the 2010 election, Labor’s performance on multicultural and immigration policy, a traditional area of strength, had been clumsy. Julia Gillard distanced Labor from Kevin Rudd’s advocacy of a ‘big Australia’. This change of tack sought to respond to the Coalition’s effective exploitation of arguments about asylum seekers. This rhetoric, together with candidate selection problems, contributed to sharp declines in Labor’s vote in New South Wales ethnic electorates (Thompson and Robinson 2011). After 2010, the Gillard Government responded to this setback. Labor in government tightened its stand against asylum seekers but worked hard to restore the traditional ties between Labor and ethnic communities. In 2012, a Minister of Multicultural Affairs was restored to cabinet for the first time since the government of Paul Keating. Labor in office demonstrated itself marginally more sympathetic to Palestinian aspirations than the Coalition, an orientation some predicted would appeal to Muslim voters in Sydney (Hetherington 2011; Jakubowicz 2013). Rudd’s return to the prime ministership probably assisted Labor’s appeal to culturally diverse voters. Many electors of Chinese ethnicity had been particularly enthused by Rudd in 2007. It is possible that a married male prime minister may have been more attractive to some socially conservative ethnic voters than an unmarried female one.

In 2010, the Green surge owed something to disillusion among Labor’s traditional constituency of public sector human service workers, especially in education. Labor’s championing of standardised testing was unpopular among teachers (Fitzsimmons 2010). Many of the policies of the Gillard Government sought to appeal to this constituency. These included the Gonski funding reforms in education, together with the National Disability Insurance Scheme and support for higher pay for community sector workers (Schneiders 2011). The fact that Labor’s vote held up in areas of high public employment suggests some success for this strategy.

Table 5: Greens’ vote determinants 2007–13

2007

2007

2010

2010

2013

2013

B

Beta

B

Beta

B

Beta

Constant

0.03

-2.48

-2.42

Agricultural electorate

-0.99

-0.11

-2.17

-0.17

-1.24

-0.09

Mining electorate

-1.78

0.01

-1.20

-0.08

-0.28

-0.02

Manufacturing

-0.16

-0.17

-0.04

-0.03

-0.14

-0.09

No religion

0.23

0.38

0.38

0.51

0.27

0.34

Non Judeo-Christian religion

0.04

0.06

0.04

0.05

-0.05

-0.07

University qualifications

0.10

0.37

0.14

0.39

0.21

0.55

Public employment

0.07

0.09

0.03

0.03

-0.10

-0.10

R-Sq (adj)

0.51

0.63

0.59

Source: Australian Bureau of Statistics.

Green support was not simply a ‘chip off the left block’. The sectoral economic cleavages that counted against Labor were largely irrelevant. In 2013, new Greens leader Christine Milne had sought to appeal to farmers. The Greens emphasised their opposition to coal-seam gas mining and also criticised the uncompetitive nature of the Australian supermarket sector which they argued had a negative impact on farm produce prices. In response conservatives condemned the Greens as enemies of agriculture. Both these positive and negative appeals had little effect (Milne 2013; Devine 2012). ‘Countrymindedness’ is a distinctively anti-Labor ideology. As Green support fell from 2007 to 2013 the party found its most loyal supporters in electorates with a high population of university graduates. This may bolster Lindsay Tanner’s argument that Green support is a reflection not only of support for particular policy positions but also a manifestation of a broader distaste among the educated for the stunts and gimmickry of contemporary politics (Tanner 2011). Some conservatives identified a regional dichotomy between areas with high proportions of university graduates and the ‘real’ Australia of the suburbs. For the broader left, as defined by the Labor two-party-preferred vote, this dichotomy did not exist (Davidson 2013).

The pattern for the Coalition was largely the obverse of that for Labor.

Table 6: Coalition vote determinants 2007–13

2007

2007

2010

2010

2013

2013

B

Beta

B

Beta

B

Beta

Constant

64.05

78.71

75.20

Agricultural electorate

5.63

0.22

7.2

0.24

7.77

0.28

Mining electorate

-13.2

-0.25

0.57

0.02

1.29

0.04

Manufacturing

-0.83

-0.31

-1.37

-0.41

-1.05

-0.34

No religion

-0.03

-0.02

-0.32

-0.19

-0.39

-0.25

Non Judeo-Christian religion

-0.71

-0.39

-0.51

-0.32

-0.67

-0.46

University qualifications

-0.07

-0.01

0.04

0.05

0.15

0.21

Public employment

-0.62

-0.27

-0.87

0.38

-0.82

-0.39

R-Sq (adj)

0.38

0.34

0.42

Source: Australian Bureau of Statistics.

The Palmer United Party (PUP) was not a right-wing version of the Greens. Green support was a distinctive component of the left, but Palmer support was much more amorphous. Palmer United preferences split only very narrowly in favour of the Coalition although its how-to-vote material advised preferences to the Coalition (Brent 2013; Green 2014). PUP was a ‘catch-all’ party, with the faint qualification that it was less popular in agricultural regions and electorates with a high proportion of university educated voters. The association between PUP support and educational levels has been noted by some observers (Mansillo 2013b). An electorate level analysis, undertaken later in this chapter, demonstrates that there was a strong regional component to its support.

Table 7: Palmer United Party vote determinants 2013

2013

2013

B

Beta

Constant

19.63

Agricultural electorate

-1.63

-0.16

Mining electorate

-1.26

-0.10

Manufacturing

-0.39

0.33

No religion

-0.01

-0.01

Non Judeo-Christian

-0.05

-0.09

University qualifications

-0.19

-0.70

Public employment

-0.14

0.17

R-Sq (adj)

0.33

Source: Australian Bureau of Statistics.

The 2013 election was a much more ‘normal’ election than 2010; what remained from 2010 was the realignment against Labor in mining regions. The impact of this realignment was significant. The rise of ‘countrymindedness’ in the 1920s contributed to the conservative inter-war political ascendancy as the Country Party won formerly Labor rural electorates (Rydon 1979: 62). In 2010, the alienation of mining regions cost Labor an overall majority. In 2013, if the 2007 pattern of support for Labor in mining regions had been restored, Labor would have held the central Queensland coal electorates of Flynn, Dawson and Capricornia and probably several electorates in the Perth region.

The other aspect of regionalism highlighted by many observers pitted the ‘suburbs’—in particular ‘western Sydney’—against the ‘inner city’. Implicitly this argument defined the ‘suburbs’ as formerly Labor-voting outer suburban areas. This would predict that the Labor (and overall left) vote would be notably higher in the ‘inner city’ but notably lower in the ‘suburbs’. An examination of the electorate residuals offers some evidence for the first proposition but none for the second.

Table 8: Largest negative Labor residuals 2004–13

2004 ALP two-party

Residuals

2007 ALP two-party

Residuals

2007 ALP primary

Residuals

2010 ALP two-party

Residuals

2010 ALP primary

Residuals

2013 ALP two-party

Residuals

2013 ALP primary

Residuals

Mitchell

16.7

Curtin

15.1

New England

21.9

Tangney

15.1

Lyne

18.6

Mitchell

15.8

Indi

16.0

Aston

14.8

Murray

14.8

Kennedy

17.5

Mitchell

14.4

New England

14.9

Tangney

14.1

Mitchell

12.3

Murray

14.2

Tangney

14.7

Franklin

14.1

Fowler

13.9

Forrest

13.8

Murray

11.1

Tangney

11.4

Tangney

13.5

Mitchell

13.6

Murray

13.9

Murray

13.9

Fowler

13.5

Berowra

10.1

Denison

11.1

Curtin

13.4

Aston

13.0

Curtin

11.8

Curtin

11.8

Tangney

13.3

Forrest

10.0

Wright

11.0

Source: Australian Bureau of Statistics.

Table 9: Largest positive Labor residuals 2004–13

2004 ALP two-party

Residuals

2007 ALP two-party

Residuals

2007 ALP primary

Residuals

2010 ALP two-party

Residuals

2010 ALP primary

Residuals

2013 ALP two-party

Residuals

2013 ALP primary

Residuals

Grayndler

20.4

Grayndler

21.2

Throsby

18.1

Melbourne

25.9

Lalor

19.8

Melbourne

26.1

Sydney

17.8

Throsby

19.2

Throsby

20.7

Fraser

15.4

Lyons

21.0

Hunter

19.7

Grayndler

19.2

Shortland

15.7

Shortland

16.6

Sydney

16.4

Shortland

14.8

Hunter

20.6

Lyons

19.2

Wills

17.5

Hunter

15.2

Lyons

15.9

Fraser

15.7

Grayndler

14.4

Willis

17.1

Shortland

15.7

Batman

17.1

Grayndler

13.6

Fraser

15.8

Batman

15.6

Lalor

12.4

Batman

17.1

Braddon

14.5

Sydney

16.3

Lyons

12.7

Source: Australian Bureau of Statistics.

Labor has always struggled in the affluent suburbs and in farming regions (Forward 1976: 23; Jupp 1982: 30–1). The Labor-held seats where the Party has notably underperformed at recent elections are those where it faced particular local problems of candidate selection, such as Franklin and Fowler. The rural independents took primary votes from Labor.

The electorates in which Labor performed better than the model predicted are largely in two categories. One is inner urban electorates such as Grayndler, which according to the ABC’s Vote Compass was the most ‘left-wing’ electorate in Australia (ABC News 2013). The second category is Labor’s traditional strongholds of the Illawarra and Hunter Valley in New South Wales. Labor’s problems in mining regions were less marked in coal producing areas.

Green support has a strong place-based component around ‘alternative’ communities such as inner Melbourne, part of inner Sydney (in particular the electorate of Grayndler), inner Hobart and the far north coast of New South Wales (Mansillo 2013a: 57–9). In 2007, the Greens benefitted in Bass by virtue of their opposition to the construction of a pulp mill in the area (McCall 2010). There is some evidence of a Melbourne–Sydney dichotomy in Greens support, particularly in 2013. The Greens over-performed the model by 7.1 percentage points in Grayndler, compared to 10.4 in Willis and 13.0 in Batman. In the electorate of Sydney the Greens actually underperformed by 2.5 points. The massive residual for Melbourne reveals that Adam Bandt’s victory in the seat of Melbourne reflected more than the demographics of the electorate (Badham 2013; Miragliotta 2013). Bandt’s campaign appealed to a particular Melbourne identity—socially progressive, educated and culturally aware.

Table 10: Largest positive Green residuals 2007–13

2007 Green primary

2010 Green primary

2013 Green primary

Bass

7.4

Melbourne

15.9

Melbourne

24.9

Denison

7.1

Willis

7.4

Batman

13.0

Melbourne

7.0

Lyons

7.3

Willis

10.4

Cunningham

6.9

Batman

7.2

Richmond

7.8

Batman

6.8

Grayndler

7.0

Grayndler

7.1

Source: Australian Bureau of Statistics.

The Greens tend to underperform in ethnic electorates where Labor can rely on traditional allegiances and community ties to bolster its vote (Mansillo 2013a: 28, 60). There is little evidence of a distinctively place-based anti-Green sentiment.

Table 11: Largest negative Green residuals 2007–13

2007

2010

2013

Bennelong

4.6

Bennelong

7.3

Denison

5.5

Kingston

4.3

Chisholm

5.4

Chisholm

5.2

Herbert

3.9

Lyne

4.1

Bennelong

4.9

Lingiari

3.9

Sturt

4.1

Banks

4.8

Makin

3.4

Parramatta

4.0

Adelaide

4.3

Source: Australian Bureau of Statistics.

The pattern of residuals for the Coalition is largely the obverse of Labor. Like Labor it lost votes to the rural independents in 2007 and 2010, but in 2013 Bob Katter surrendered much of his appeal to Coalition voters. If there is an electorate that incarnates a distinctively Sydney conservatism it is Mitchell not Lindsay.

Table 12: Largest negative Coalition residuals 2007–13

2007

Residuals

2010

Residuals

2013

Residuals

New England

28.7

New England

31.5

Melbourne

28.7

Grayndler

21.4

Kennedy

29.0

Grayndler

18.5

Throsby

18.9

Melbourne

26.9

Willis

18.3

Sydney

16.8

Hunter

19.3

Batman

18.1

Newcastle

16.7

Batman

17.9

Hunter

17.1

Source: Australian Bureau of Statistics.

Table 13: Largest positive Coalition residuals 2007–13

2007

Residuals

2010

Residuals

2013

Residuals

Murray

15.1

Murray

16.8

Mitchell

16.1

Curtin

14.5

Mitchell

16.1

Tangney

13.2

Mitchell

13.7

Tangney

13.9

Forrest

12.4

Aston

13.1

Fowler

13.2

Murray

11.3

Riverina

12.7

Forrest

13.1

Hughes

11.1

Source: Australian Bureau of Statistics.

There is no evidence of a distinctive regional pattern of conservatism in western Sydney contrary to some interpretations (Cater 2013). This absence is compatible with a social rather than cultural regional model. If western Sydney is the mirror of the nation it would be expected to follow the national trend. Voters in western Sydney who switched their vote to the Coalition cited the same concerns as voters elsewhere: Labor disunity and the economy (Carswell 2013). In the Melbourne suburbs Labor in 2010 failed to win Aston, Casey and Dunkley, all of which should have been Labor gains under this model.

Table 14: Western Sydney Labor two-party-preferred residuals 2004–13

2004

2007

2010

2013

Banks

-0.7

1.0

0.0

-0.2

Blaxland

-5.1

-8.3

-3.4

-3.2

Chifley

8.2

9.0

4.9

7.2

Fowler

-8.9

-10.9

-13.9

-3.2

Greenway

-10.6

-10.3

-5.5

-0.7

Lindsay

0.5

2.9

0.2

0.2

McMahon (Prospect in 2004)

4.4

3.4

3.5

4.7

Parramatta

-4.6

-2.5

-5.0

-7.2

Reid

-6.8

-6.3

-2.3

-3.9

Werriwa

0.7

1.5

-4.3

-5.7

Source: Australian Bureau of Statistics.

Conclusion

The result of the 2010 election appeared to anticipate a major shift in Australian politics: the emergence of a cultural and political polarisation between the states akin to that apparent in the United States. The 2013 election result demonstrated that Australian electoral politics had returned to a more normal pattern. The 2010 result now appeared as the first instalment of a general shift against Labor. During the Rudd–Gillard years Labor lost support across the country, with a particular estrangement in mining regions. There were some signs in the 2013 election of economic regionalism, but little evidence of a distinctively conservative cultural regionalism.

The implications of this regional pattern for the future are unclear. In 2000, Clive Bean argued that although Labor’s loss of core support at the 1996 election was disheartening to it as a major party, it had left an easier task for Labor than would have been the case if there had been a general swing against the party. Labor, he argued, could focus on the task of regaining traditional supporters (Bean 2000: 87–8). An alternative argument would be that it is easier for the Labor Party to recover from a general repudiation such as that of 2013. Voters across all regions in 2013 judged Labor harshly for its failure to deliver basic public goods such as economic security and political stability. Kevin Rudd’s reforms to the Labor leadership may enable new Opposition Leader Bill Shorten to draw a line under the disunity of the Rudd–Gillard years. Economic uncertainty now counts against an incumbent Coalition government.

References

ABC News. 2013. ‘Australia’s most left-leaning and right-leaning seats revealed’. 4 September, viewed 18 December 2013: <www.abc.net.au/news/2013-09-04/vote-compass-left-right-electorates/4929064>.

Aitkin, Don. 1982. Stability and change in Australian politics. 2nd edn. Canberra: The Australian National University Press.

Aitkin, Don. 1985. ‘“Countrymindedness”: The Spread of an Idea’. Australian Cultural History 4: 34–41.

Allard, Tom. 2013. ‘Southern discomfort’. Sydney Morning Herald, 8 September, viewed 28 January 2014: <www.smh.com.au/federal-politics/federal-election-2013/southern-discomfort-the-states-where-labor-lost-the-election-20130908-2tdi1.html>.

Australian Bureau of Statistics. 2014. Labour Force Australia, 13 February, viewed 14 February 2014: <www.abs.gov.au/Ausstats/abs@.nsf/exnote/6202.0>.

Australian Electoral Commission (AEC). 2005. The Official 2004 Federal Election Results. Australian Electoral Commission, viewed 1 November 2013: <results.aec.gov.au/12246/results/default.htm>.

AEC. 2008. The Official 2007 Federal Election Results. Australian Electoral Commission, viewed 1 November 2013: <results.aec.gov.au/13745/Website/Default.htm>.

AEC. 2011a. Redistributions. Australian Electoral Commission, viewed 7 January 2014: <www.aec.gov.au/Electorates/Redistributions/index.htm>.

AEC. 2011b. House of Representatives—Two Party Preferred Results 1949—Present. Australian Electoral Commission, viewed 15 January 2014: <www.aec.gov.au/Elections/australian_electoral_history/House_of_Representative_1949_Present.htm>.

AEC. 2013a. 2013 federal election downloads. Australian Electoral Commission, viewed 5 November 2014: <www.aec.gov.au/Elections/Federal_Elections/2013/downloads.htm>.

AEC. 2013b. The Official 2013 Federal Election Results. Australian Electoral Commission, viewed 4 January 2014: <results.aec.gov.au/17496/Website/Default.htm>.

Badham, Van. 2013. ‘Adam Bandt’s victory in Melbourne is no fluke—and that’s worth celebrating’. Guardian Australia, 8 September, viewed 29 January 2014: <www.theguardian.com/commentisfree/2013/sep/08/adam-bandt-melbourne-victory>.

Bean, Clive. 2000. ‘Who Now Votes Labor?’ In John Warhurst and Andrew Parkin (eds), The Machine: Labor Confronts the Future. Sydney: Allen & Unwin.

Bean, Clive, and McAllister, Ian. 2012. ‘Electoral Behaviour in the 2010 Federal Election’. In Marian Simms and John Wanna (eds), Julia 2010: The caretaker election, Canberra: ANU E-Press.

Blainey, Geoff. 2013. ‘Mining a rich seam’. The Age, 19 October.

Brent, Peter. 2013. ‘The 2PP wait is over’. Mumble, 29 November, viewed 18 December 2013: <blogs.theaustralian.news.com.au/mumble/index.php/theaustralian/comments/the_wait_is_over>.

Butler, David, and Stokes, Donald. 1976. Political Change in Britain. New York: St. Martin’s Press.

Carswell, Andrew. 2013. ‘Labor’s light dims in the west as Sydney seats fall to Coalition’. Daily Telegraph, 8 September, viewed 29 January 2014: <www.dailytelegraph.com.au/news/special-features/labor8217s-light-dims-in-the-west-as-sydney-seats-fall-to-coalition/story-fnho52jp-1226714368648>.

Cater, Nick. 2013. ‘Labor resists western values’. The Australian, 8 November, viewed 29 January 2014: <www.theaustralian.com.au/opinion/columnists/labor-resists-western-values/comments-fnhulhjj-1226700141869>.

Charnock, David. 1997. ‘Spatial variations, contextual and social structural influences on voting for the ALP at the 1996 federal election: conclusions from multilevel analyses’. Australian Journal of Political Science 32(2): 237–54.

Cockfield, Geoff and Botterill, Linda Courtenay. 2012. ‘Signs of Countrymindedness: A Survey of Attitudes to Rural Industries and People’. Australian Journal of Political Science 47(4): 609–22.

Colebatch, Tim. 2013a. ‘Geography is destiny: states of red and blue’. Sydney Morning Herald, 7 December, viewed 29 January 2014: <www.smh.com.au/federal-politics/political-news/geography-is-destiny-states-of-red-and-blue-20131206-2ywri.html>.

Coorey, Phillip. 2013. ‘Labor faces annihilation in marginal seats’. Australian Financial Review, 28 March, viewed 29 January 2014: <www.afr.com/p/national/labor_faces_annihilation_in_marginal_MRqll2ldnEA2yvStbw4IMM>.

Cowgill, Matt. 2013. ‘Low employment in Tasmania: it’s not just the seniors’. We are all dead, 20 March, viewed 29 November 2013: <mattcowgill.wordpress.com/2013/03/20/low-employment-in-tasmania-its-not-just-the-seniors>.

Davidowitz-Stephens, Seth. 2013. ‘How Many American Men Are Gay?’ New York Times, 7 December, viewed 29 January 2014: <www.nytimes.com/2013/12/08/opinion/sunday/how-many-american-men-are-gay.html?_r=0>.

Davidson, Sinclair. 2013. ‘Testing the lucky culture’. Callataxy Files, 21 May, viewed 29 January 2014: <catallaxyfiles.com/2013/05/21/testing-the-lucky-culture>.

Devine, Miranda. 2012. ‘Greens’ country girl Christine Milne no friend of the bush’. Herald Sun, 15 February, viewed 29 January 2014: <www.heraldsun.com.au/news/opinion/greens-country-girl-christine-milne-is-no-friend-of-the-bush/story-e6frfhqf-1226326602622>.

Essential Media Communications. 2012. ‘Essential Report’, 20 August.

Ferguson, John and Hannan, Ewin. 2013. ‘Ford blames Kevin Rudd’s FBT tax changes for plant shutdowns’. The Australian, 23 August, viewed 29 January 2014: <www.theaustralian.com.au/national-affairs/election-2013/ford-blames-kevin-rudds-fbt-tax-changes-for-plant-shutdowns/story-fn9qr68y-1226702474062>.

Fitzsimmons, Hamish. 2010. ‘Teachers boycott literacy, numeracy tests’. ABC Lateline, 12 April, viewed 14 December 2013: <www.abc.net.au/lateline/content/2010/s2870861.htm>.

Forward, Ann. 1976. A Geographical Analysis of the Labor Vote in Victoria, 19661972. Melbourne: Department of Geography, Monash University.

Garnaut, Ross. 2013. Dog Days: Australia After the Boom. Melbourne: Black Inc.

Gordon, Josh. 2013. ‘Federal woes trashing state Labor’s brand’. Sydney Morning Herald, 2 May, viewed 29 January 2014: <www.smh.com.au/comment/federal-woes-trashing-state-labors-brand-20130501-2it0m.html>.

Green, Antony. 2007. Antony Green’s Election Guide. ABC—Australia Votes 2007, viewed 15 November 2014: <www.abc.net.au/elections/federal/2007/guide/>.

Green, Antony. 2013. ‘Electoral Pendulum’. ABC News, viewed 17 September 2013:<http://www.abc.net.au/news/federal-election-2013/guide/pendulum>.

Henderson, Gerard. 2013. ‘Abbott’s anguish: inner-city types in media dislike him’. Canberra Times, 3 December, viewed 29 January 2014: <www.canberratimes.com.au/comment/abbotts-anguish-innercity-types-in-media-dislike-him-20131202-2ylyd.html>.

Hetherington, David. 2011. ‘Australian Labor re-embraces multiculturalism’. Policy Network, 7 March, viewed 10 February 2014: <www.policy-network.net/pno_detail.aspx?ID=3970&title=Australian+Labor+re-embraces+multiculturalism>.

Jakubowicz, Andrew. 2013. ‘How Labor won the “ethnic vote” in Western Sydney’. The Drum, 18 September, viewed 11 January 2014: <www.abc.net.au/news/2013-09-13/jakubowicz-politics-of-ethnicity/4954660>.

Johnston, Ron J and Pattie, Charles J. 1988. ‘People, Attitudes, Milieux and Votes: An Exploration of Voting at the 1983 British General Election’. Transactions of the Institute of British Geographers 13(3): 303–23.

Jupp, James. 1982. Party Politics: Australia 19661981. Melbourne: Cambridge University Press.

Kellner, Peter. 2013. ‘Reclaiming the north’. Prospect, 17 October, viewed 22 October 2012: <www.prospectmagazine.co.uk/magazine/features/reclaiming-the-north-tories-peter-kellner>.

Kelly, Matthew. 2013. ‘Carbon tax destroying confidence: Mirabella’. Newcastle Herald, 9 April, viewed 29 January 2014: <www.theherald.com.au/story/1420852/carbon-tax-destroying-confidence-mirabella>.

Kenny, Mark. 2013. ‘Labor facing wipeout: poll’. Sydney Morning Herald, 2 March, viewed 29 January 2014: <www.smh.com.au/national/labor-facing-wipeout-poll-20130301-2fc6u.html>.

King, Gary, Rosen, Ori, Tanner, Martin and Wagner, Alexander F. 2008. ‘Ordinary economic voting behavior in the extraordinary election of Adolf Hitler’. The Journal of Economic History 68(04): 951–96.

Latham, Mark. 2013. Not Dead Yet: What Future for Labor? Melbourne: Black Inc.

Mansillo, Luke. 2013a. ‘Spatial Dimensions of the 2010 Australian Greens Vote: The Gay and Lesbian Effect on Voting’. Honours thesis, The Australian National University. Available at: <genderinstitute.anu.edu.au/sites/default/files/webform/LJM Luke Mansillo Thesis—Spatial Dimensions of the 2010 Australian Greens Vote THe Gay and Lesbian Effect on Voting_0.pdf>.

Mansillo, Luke. 2013b. ‘Palmer United Party: who voted for them?’ Guardian Australia, 28 October, viewed 29 January 2014: <www.theguardian.com/news/datablog/2013/sep/09/who-voted-for-palmer-united>.

McCall, Tony. 2010. ‘Tasmania’. Australian Cultural History 28(1): 95–102.

McQueen, Humphrey. 1982. Gone Tomorrow: Australia in the 80s. Sydney: Angus & Robertson.

Milne, Christine. 2013. ‘Christine launches the Greens’ election campaign’. GreensMPs, 24 August, viewed 10 February 2014: <christine-milne.greensmps.org.au/content/video/christine-launches-greens-election-campaign>.

Miragliotta, Narelle. 2013. ‘Reassessing Melbourne three years’. The Conversation, 12 August, viewed 28 October 2013: <http://theconversation.com/reassessing-melbourne-three-years-on-15515>.

Parker, Robert S. 1949. ‘Australian Federation: the Influence of Economic Interests and Political Pressures’. In Jennifer Jill Eastwood and Francis Barrymore Smith (eds), 1964, Historical Studies: Selected Articles, First Series, Melbourne: Melbourne University Press.

Phillips, Kristen. 2009. ‘Immigration Detention, Containment Fantasies and the Gendering of Political Status in Australia’. PhD thesis, Curtin University of Technology. Available at: <espace.library.curtin.edu.au/R?func=dbin-jump-full&local_base=gen01-era02&object_id=129031>.

Robinson, Geoff. 2013. ‘Ultra-marginal Corangamite is a mirror of the nation’. The Conversation, 6 September, viewed 5 October 2013: <theconversation.com/ultra-marginal-corangamite-is-a-mirror-of-the-nation-16883>.

Rydon, Joan. 1979. ‘The Conservative Electoral Ascendancy Between the Wars’. In Cameron Hazlehurst (ed.), Australian Conservatism: Essays in Twentieth Century Political History, Canberra: The Australian National University.

Scalmer, Sean. 1997. ‘The Affluent Worker or the Divided Party? Explaining the Transformation of the ALP in the 1950s’. Australian Journal of Political Science 32(3): 401–18.

Schirmer, Jackie. 2012. ‘Still here: why Tasmanian forest industry job figures are misleading’. The Conversation, 20 November, viewed 28 October 2013: <theconversation.com/still-here-why-tasmanian-forest-industry-job-figures-are-misleading-10827>.

Schneiders, Ben. 2011. ‘20% wage boost for low-paid workers’. The Age, 10 November, viewed 29 January 2014: <www.theage.com.au/national/20-wage-boost-for-lowpaid-workers-20111110-1n8fq.html>.

Shanahan, Dennis. 2012. ‘Even Kevin Rudd at risk in Qld Labor wipeout’. The Australian, 2 July, viewed 29 January 2014: <www.theaustralian.com.au/national-affairs/even-kevin-rudd-at-risk-in-qld-labor-wipeout/story-fn59niix-1226413912633#>.

Sheehan, Paul. 2006. ‘Ideals become casualties of war’. Sydney Morning Herald, 13 August, viewed 29 January 2014: <www.smh.com.au/news/paul-sheehan/ideals-become-casualties-of-war/2006/08/13/1155407666922.html08/13/1155407666922.html>.

Sides, John. 2013. ‘Most Americans are not like Antonin Scalia’. The Monkey Cage, 18 October, viewed 22 October 2013: <www.washingtonpost.com/blogs/monkey-cage/wp/2013/10/18/most-americans-are-not-like-antonin-scalia>.

Tam Cho, Wendy K. 2013. ‘Voter migration is a significant factor in the geographic sorting of the American electorate’. LSE American Politics & Policy, 19 November, viewed 5 December 2013: <blogs.lse.ac.uk/usappblog/?p=2457&preview=true>.

Tanner, Lindsay. 2011. Sideshow: Dumbing Down Democracy. Melbourne: Scribe.

Taylor, Peter J and Johnston, Ron J. 1979. Geography of Elections. London: Croom Helm.

The Australian. 2014. ‘Here comes the end of a bad political marriage’. The Australian, 10 January, viewed 29 January 2014: <www.theaustralian.com.au/opinion/editorials/here-comes-the-end-of-a-bad-political-marriage/story-e6frg71x-1226798518852>.

Thompson, Elaine and Robinson, Geoff. 2011. ‘New South Wales’. In Marian Simms and John Wanna (eds), Julia 2010: the caretaker election, Canberra: ANU E-Press.

University of Western Australia. 2013. ‘Australian Politics and Elections Database’. Australian Politics and Elections Database, viewed 13 October 2013: <elections.uwa.edu.au>.

Ward, Ian and Verrall, Derek. 1982. ‘Liberal Ascendancy in the Western District’. In Brian Costar and Colin A Hughes (eds), Labor to Office: the Victorian State Election 1982, Melbourne: Drummond.

1 Thanks are due to participants in the 2013 Election Workshop for comments, to Antony Green for assisting in the location of electoral statistics and to Lisa Hay for editing.

2 Available from <http://www.abs.gov.au/websitedbs/censushome.nsf/home/tablebuilder>.

3 In this table and all subsequent tables, values in bold are significant at 0.01, underlined at 0.05.


Previous Next