Morality policy processes in advanced industrial democracies

ABSTRACT

In recent years, several countries have experienced widespread, intense debates about morality issues such as the death penalty, abortion, ART/stem cell research, same-sex marriage, and euthanasia. Assuming the distinctiveness of morality policies from other policy fields, this article analyses three alternative institutional models for explaining variations in the amount of conflict over these morality issues across 24 Western democracies. Is either the US-developed “policy type” model or the European-developed “two worlds” of morality politics, based on religious and secular party systems, applicable more broadly? Are there regional patterns (Europe and non-Europe) to any institutional findings? How does each model contribute to our understanding of morality policy comparatively across Western democracies? We find broader cross-national support for the policy type model, with the two worlds model largely restricted to Europe. The US has more morality policy conflict because of its unusual combination of a political party targeting religiously oriented voters within an institutional framework of multiple venues. The US and other non-European countries have similar patterns of institutional deliberation through decentralization, and the US also has similarities in multiple venues with religious party systems in Europe. Secular European party systems with centralized institutions have the fewest venues for morality policy debate.

KEYWORDS:

• Morality policy
• advanced industrial democracies
• policy processes
• religious party systems
• secular party systems
• US exceptionalism
• policy type

Acknowledgements

Earlier versions of this article were presented at the American Political Science Association, Chicago, and the Irish Political Studies Association, Dublin. Thanks to participants in those sessions, to William Harrison of West Virginia University (now at Fairmont State University) for data assistance, and to Bob Duval of WVU for methodological advice.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Donley T. Studlar is Emeritus Professor of Political Science at West Virginia University. The author of five books and some 150 articles, his current research interests are comparative morality policy and policy on non-communicable disease risks such as tobacco, alcohol, and obesity.

Gordon J. Burns is a graduate student in Political Science at West Virginia University. He has published previously in Policy Sciences.

Alessandro Cagossi holds a Ph.D. from West Virginia University and has previously published articles on morality policy, regulation, and federalism in the European Union.

Notes

1 Knill, Adam, and Hurka (2015) find that “manifest” morality issues (abortion, euthanasia, homosexuality, pornography, and prostitution) have different regulatory processes than do “latent” morality issues (gambling, illegal drugs, handguns), suggesting that morality policies do have distinct institutional developments.

2 Aside from same-sex domestic relationships, these issues all qualify as “core” morality issues of life and death. See Knill (2013).

3 Any general explanation of morality policy processes should include what is arguably the first morality policy discovered to deviate from responsible party government (Christoph 1962). However, two of the recent models based on general comparative studies (Engeli, Green-Pedersen, and Larsen 2012; Knill, Adam, and Hurka 2015) do not include the death penalty. It differs from the other four issues in that the Roman Catholic Church no longer takes a conservative position, but this occurred only during Vatican II in the early 1960s, and the effect appears to have been more pronounced on countries later transitioning to democracy (McGann and Sandholtz 2012).

4 “Party-centred” does not necessarily mean “party-whipped”. Outshoorn (1996) and Halfmann (2011) demonstrate that having a completely unified party vote on abortion is an extremely rare event in Europe. While most parties do avoid emphasizing morality issues in party manifestos (Pennings 2010) and disciplined votes in the legislature, the religious/secular party system divide argues that leftist parties use these issues to keep religiously based parties on the defensive while in secular systems, there is no such “party strategy”. Hence one would expect stronger party voting in legislatures in religious party systems.

5 Because we have no data on interest group mobilization or public opinion, we exclude this part of the policy types theory from consideration.

6 We refer to these as “venues” for ease of comparability but some of them are more procedural rather than having a completely different institutional locus. Both legislative initiatives and divided party voting occur in legislatures, and constitutional procedures can result from multiple sources, including the legislature, judiciary, and referendum. Parties can be treated as actors within an institutional framework (two worlds), but also as venues in which divisions occur (policy type). See Engeli, Green-Pedersen, and Larsen (2013), Smith (1975), Smith and Tatalovich (2003), and Tatalovich and Daynes (1988, 2011).

7 Some European countries have had courts active on morality issues, but this is largely a recent development. See Tate and Vallinder (1995).

8 In addition to previously cited studies, see Lindsey (2010), Blank and Merrick (2001), Council of Europe (1999), Hood and Hoyle (2008), International Planned Parenthood Federation European Network (2009), Isasi and Knoppers (2006), McGann and Sandholtz (2012), Mitsumori and Rogers (2004), and Singh et al. (2009).

9 Introducing time as a variable might result in some changes, but institutions, including party systems, generally have been stable over the almost 70 years considered here. For some analyses across European democracies of three of the policies examined here by decades, 1960–2010, using different models of policy, see Knill, Adam, and Hurka (2015).

10 Having the formal possibility of a referendum does not, of course, mean it will be used, that depends on the political development of the issue.

11 Although Hooghe, Marks, and Schakel (2010) have several policy indicators for different levels of government in Europe, morality policy is not among them. Most discussions of morality policy in individual countries focus on the central level although variations by decentralized authorities are sometimes acknowledged (Engeli, Green-Pedersen, and Larsen 2012). We consider Denmark to be a decentralised country because of the status of Greenland and the Faroe Islands, where exceptions to central laws apply.

12 The Christian Democratic International became the Centrist Democrat International in 2001, thus confusing the formal distinction between Christian Democratic and other conservative parties. Italy has at least two confessionally oriented parties (with former Christian Democratic members) similar to Spain. In the latter country, the Popular Party is highly religiously influenced through church organisations and positions taken on morality issues, as the 2014 attempt to restrict abortion laws demonstrates (see Chaqués Bonafont and Palau Roqué 2012).

13 The Mann–Whitney U typically provides comparable power to the standard two-sample t-test when the assumptions of the t-test are met, and is somewhat more robust when the assumptions are violated (e.g. normality). For an extended examination of the relative power of the two approaches, see Duval and Groeneveld (1987).

14 Because our analysis is constrained by sample size, we adopt α = .10 criterion for statistical significance. We are balancing the cost of two types of errors (acceptance, Type I, versus rejection of findings, Type II) with the plausibility of the hypotheses being tested. If our hypotheses make sense, the sample size is small, and the cumulative set of results provides consistency across the hypothesis tests, having a higher Type I rate is acceptable. See Duval and Groeneveld (1987).

15 The measure for the judiciary category barely misses significance at .104.

16 As Table 1 shows, six of the seven federal countries (Australia, Belgium, Canada, Germany, Switzerland, and US, with Austria excepted) have decentralised debate on some morality policies, as does Austria on another issue, prostitution (Wagenaar, Altink, and Amesberger 2013). However, two countries with institutionally devolved government (Spain and United Kingdom) also have morality policy debate. Even before formal devolution, morality issues such as capital punishment and abortion were subject to non-central authorities in the United Kingdom (Christoph 1962; Cohan 1986; Cowley 1998).

17 In order to test whether the US is statistically significantly different from other countries on morality policy, the Mann–Whitney U for the difference in two sample medians is used since there is no direct non-parametric equivalent to the single sample t-test. This necessitates a sample size of 1 for the US “group”, but this does not violate the assumptions of the test.

18 Secular party systems with decentralisation (UK, Canada, Australia) use this venue at a high level except for Denmark.

19 Israel is an anomaly because of its combination of few venues and a religious party system. The party system consists of 20–25% electoral and legislative support for several religiously connected parties rather than one major one, as in European countries.

20 This may also help explain why US scholars, especially those studying state politics, have been most active in studying morality policy.