The opinion–policy link in Australia

Abstract

Whether or not policy is responsive to public opinion is central to questions of representation. Democracy by many accounts is premised on there being a strong correspondence between public opinion and policy. This link has not, however, been examined in detail in Australia. This article examines the policy–opinion link in a more robust way than that has previously been achieved in Australia, through the use of legislative data from the Australian Policy Agendas Project (APAP) and public opinion data from Roy Morgan. The article asks: is policy congruent with public opinion in Australia? In addressing this question, we examine to what extent policy accords with the preferences of the public. This in turn provides us with some answers about how representative Australian democracy is, as well as contributing to an international debate about the responsiveness of policy to public opinion.

政策是否要对舆论负责，这是代表制的核心问题。民主的前提往往就在于舆论和政策之间的息息相通。但在澳大利亚，人们并没有仔细地研究过二者间的关联。本文根据澳大利亚政策议题计划提供的立法资料以及罗伊摩根提供的舆情资料，对政策—舆论关联做了比以往充分多的研究。本文试问：政策与舆论在澳大利亚是否一致？我们考察了政策在多大程度上符合民意。这也为澳大利亚民主在多大程度上具有代表性的问题提供了答案，对于国际上争论政策对民意的回应问题也具有参考意义。。

Keywords:

• Australian politics
• policy agendas
• policy–opinion link
• policy research
• public opinion
• representation

We would like to thank Murray Goot, Peter John, Will Jennings, Ian McAllister, two anonymous reviewers and participants at the Australian Political Science Association conference in Perth in September 2013 for their helpful comments on earlier versions of this paper. The ARC Grant DP110102622 ‘Policy Agendas in the Australian Government’ funded the research reported in this paper.

Notes

1 This question was preceded by one about what is the MII facing the world, which is likely to have conditioned responses to this question. The sample size is >500, and the surveys were carried out by telephone, usually over two days. The open-ended responses were coded into a large number of categories that we then used to match to the APAP codes.

2 There are a few other problems with the Roy Morgan data that should be pointed out. Roy Morgan did not conduct the survey every year (hence the years missing from the analysis below). Roy Morgan also often conducted the survey more than once a year (e.g., in 2010, it was run four times), so the figures here have been aggregated by year.

3 Jones and Baumgartner (2004: 7) in looking at congruence had no lagged effect but, as they acknowledge and argue, they are looking at hearings that are more sensitive to public opinion (at the ‘front end’ of the process), whereas we would expect some delay in the ‘back end’ in terms of policy-making.

4 The equation is: PT = P1 + [(T − t1) (P2 − P1)]/(t2 − t1).

5 This is because for the period 1992–2002, respondents chose the three MII so responses did not sum to 100; and in some years where we did not have a matching policy agenda code we deleted the Roy Morgan response, entailing the responses did not sum to 100.

6 Defence was actually the fourth MII using averages, but because this is largely owing to one time point it was excluded from this analysis.

7 The 2012 Roy Morgan data are not yet available.

8 See Note 6.