Explaining the Australian marriage equality vote: An aggregate-level analysis

AuthorAndrea Carson,Timothy B Gravelle
DOI10.1177/0263395718815786
Date01 May 2019
Published date01 May 2019
Subject MatterArticles
/tmp/tmp-1761eDmIzGO0E9/input
815786POL0010.1177/0263395718815786PoliticsGravelle and Carson
research-article2018
Article
Politics
2019, Vol. 39(2) 186 –201
Explaining the Australian
© The Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
marriage equality vote: An
https://doi.org/10.1177/0263395718815786
DOI: 10.1177/0263395718815786
journals.sagepub.com/home/pol
aggregate-level analysis
Timothy B Gravelle
The University of Melbourne, Australia
Andrea Carson
La Trobe University, Australia
Abstract
The Australian public voted in November 2017 in favour of changing the law to allow for same-
sex marriage – only the second such national popular vote after Ireland in 2015. Though 61.6% of
the Australian public voting in the Marriage Law Postal Survey voted Yes in support of marriage
equality, this support was not uniformly distributed across the country, with support at the
electoral division level varying between 26.1% and 83.7%. What, then, explains such variation
in support for same-sex marriage among the Australian public? In this article, we advance an
aggregate, electoral division-level explanation of the Yes vote that links support for the legalisation
of same-sex marriage to a set of local-level political and socio-demographic factors.
Keywords
Australia, same-sex marriage, voting behaviour
Received: 14th May 2018; Revised version received: 20th September 2018; Accepted: 16th October 2018
Introduction
On 15 November 2017, the Australian Bureau of Statistics (ABS) announced the results
of the Marriage Law Postal Survey, ending a contentious, months-long campaign and
confirming majority support (61.6%) for legalising same-sex marriage among the
Australian public. With just over three in five Australians responding Yes to the survey,
the right-of-centre Liberal-National Coalition government of then-Prime Minister
Malcolm Turnbull committed to legislating marriage equality by the end of the year, ful-
filling this promise on 7 December 2017. This means Australia has joined the 25 countries
Corresponding author:
Timothy B Gravelle, School of Social and Political Sciences, The University of Melbourne, John Medley
Building, Parkville, VIC 3010, Australia.
Email: tim.gravelle@unimelb.edu.au

Gravelle and Carson
187
(at the time of writing) recognising marriage equality in law, more than 16 years since
same-sex marriage was first legalised in the Netherlands in 2001.
The government-run voluntary postal vote was unusual in the Australian case, since
turning up to vote in elections and referenda is legally mandatory. The campaign ahead of
the vote through mid- to late-2017 was also politically contentious, with the governing
right-of-centre Liberal Party taking no official position and Liberal cabinet ministers sid-
ing with both the Yes and No campaigns, and the Labor Party and Green Party both
endorsing the Yes campaign. The campaign also featured duelling television ads and accu-
sations of misinformation from advocacy groups Australian Marriage Equality (support-
ing Yes) and the Coalition for Marriage, mainly backed by Australian religious
organisations (supporting No). It entailed court challenges to the lawfulness of the postal
vote intended to halt it, and a controversial donation of one million Australian dollars by
the Anglican Archdiocese of Sydney to the No campaign. The postal vote thus captured
Australians’ attitudes on a highly salient, noncomplex, symbolic social policy issue (or
‘easy’ issue) (Carmines and Stimson, 1980; Haider-Markel and Meier, 1996).
In the end, every Australian state and territory recorded a majority Yes vote, as did 133
of the 150 electoral divisions of the House of Representatives. It was soon apparent, how-
ever, that while support for marriage equality was highest in major urban areas (the per-
centages voting Yes were 70% or higher in each of Sydney, Melbourne, Brisbane, Perth
and Adelaide), there were nevertheless parts of metropolitan areas where the majority
voted No, including the electoral divisions of Blaxland in New South Wales, west of
Sydney (73.9%) and Calwell in Victoria, northwest of Melbourne (56.8%). Support for
marriage equality was therefore far from uniformly distributed across the country (see
Figure 1).
This chapter in Australian politics has importance beyond Australia because it repre-
sents only the second case (after Ireland, in 2015) of legalising same-sex marriage follow-
ing a national popular vote on the issue. Still, detailed studies of the Irish same-sex
marriage referendum are relatively few (but see Elkink et al., 2017; Murphy, 2016;
O’Mahony, 2017). Studies of political behaviour and public policy at the subnational
level are numerous yet focus almost exclusively on the American case (Camp, 2008;
Campbell and Monson, 2008; Fleischmann and Moyer, 2009; Haider-Markel and Meier,
1996; Lax and Phillips, 2009; Lofton and Haider-Markel, 2008; Soule, 2004; Wald et al.,
1996). As such, this specific form of mass political participation is relatively understud-
ied, particularly in cross-national perspective. What, then, explains the Australian pub-
lic’s aggregate support for marriage equality? The aim of this article is to uncover the
contextual determinants of the aggregate-level Yes vote using the publicly reported results
from the Australian Bureau of Statistics (ABS) joined to other data capturing local-level
socio-demographic and political characteristics.
We undertake an aggregate, Federal Electoral Division (FED)-level analysis of the
marriage equality vote for two principal reasons. The first is necessity: unlike the case of
Australian federal elections, no large-scale surveys of the Australian public measuring
voting intentions (during the campaign) or self-reported vote (after the campaign) were
conducted; this means that the FED-level data from the ABS are the best data available
for studying Australians’ voting behaviour in this case. Second, given that legislation was
required to change Australian marriage law in response to the postal survey result, it is
important to understand electoral division-level dynamics because these districts are the
contexts inhabited by legislators. Such aggregate-level (or ecological) explanations of
political behaviour also have a long history, dating back to the seminal work of V.O. Key


188
Politics 39(2)
Figure 1. The geographic distribution of the support for marriage equality (% yes).
(1949) who sought to explain racial attitudes and voting behaviour in the southern United
States with reference to local contextual factors. Extensions of this tradition of analysis
include contextual analyses of voting behaviour in other country contexts, including
Australia (Eagles, 2002; Jones, 1981) and the local-level correlates of anti-gay hate crime
(Green et al., 2001). Thus, aggregate-level analyses of political behaviour are a valuable
tool in explaining a range of aggregate political and social phenomena (Eagles, 1995).
With the analysis presented here, we aim to contribute to ongoing internationalisation of
research on public attitudes towards same-sex rights (cf. Brewer, 2014).
We structure the rest of the article as follows. First, we provide some necessary back-
ground on the politics of marriage equality in Australia, and survey the existing literature
on support for same-sex marriage to motivate our research expectations. We then describe
our contextual, aggregate-level data sources and our modelling approach before discuss-
ing our results.
Background and research expectations
From the perspective of survey methodology, the Marriage Law Postal Survey would be
considered a great success. The survey featured a clear question: ‘Should the law be
changed to allow same-sex couples to marry?’ Further, the national postal survey partici-
pation rate of 79.5% of enrolled voters was high by global standards for a voluntary bal-
lot. Electoral division-level participation ranged between 50.0% and 85.2%. (The
significance of varying participation rates is discussed in our findings.) In terms of

Gravelle and Carson
189
ascertaining majority public opinion, the postal survey was considered redundant by
many commentators because numerous publicly released polls had indicated that major-
ity support for same-sex marriage had existed for a decade (Coorey, 2015). Still, the
voluntary survey was politically expedient for the Liberal-National Coalition govern-
ment, allowing it to keep its promise from the 2016 election campaign to hold a plebiscite
of sorts on the issue, even though it was subsequently twice unable to command support
in the Senate to pass the necessary enabling legislation to do so. The postal survey did not
require legislation to be passed, and thus allowed the government to deal with the impasse
on same-sex marriage policy caused by conflict between factions within the Coalition
over the issue (Carson et al., 2018). McAllister (2011) argues that Australia’s main politi-
cal parties typically avoid adjudicating on ‘moral’ issues, since they tend to promote
intra-party divisions. Consistent with this approach, then-Australian Prime Minister
Malcolm Turnbull argued the popular ballot was necessary to ‘give all Australians a say’
(Koziol and Remeikis, 2017). A similar motivation underpinned the Irish decision to put
same-sex marriage to a popular vote. By doing so, Ireland’s political parties avoided
intra-party division, and avoided electoral retribution for their stances on the issue
(O’Mahony, 2017).
The...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT