Free Votes, MPs, and Constituents: The Case of Same-Sex Marriage in Canada

Abstract In the summer of 2005, with the passage of the Civil Marriage Act, Canada became the third country to extend full rights to same-sex marriages. This article explores passage of the CMA, focusing on parliamentary voting behavior on the free vote used in the House of Commons. Using multivariate empirical analysis we find unusually strong evidence of constituency characteristics influencing the voting behavior of MPs, a rare outcome given the existing scholarly literature on free votes. In a concluding section, we discuss what these findings imply about the increasingly important debate in Canada about parliamentary accountability. Keywords: free votesconstituency influencesame-sex marriage Acknowledgements This research was supported by a generous Faculty Research Grant from the Canadian Government. Previous versions of this article have been presented at the Working Group on Canadian Studies at the University of Missouri and at the 2008 meeting of the Midwest Association for Canadian Studies. Notes 1. Franks (Citation1997, 34) has gone so far as to argue that "greater use of free votes would require massive changes in the process of representation and decision-making in the parliamentary system. It would change the system of responsibility and accountability: if members of Parliament rather than the government make the decisions, then members instead of the government should be held accountable … [t]he choices facing the electorate would be more blurred than at present." 2. Preliminary analysis, not reported fully here, led us to conclude that in most cases abstainers were motivated by idiosyncratic concerns. Treating abstentions as missing data, excluding them from our analysis, and running our models with binary logistic regression estimators yields very similar results. 3. As with most studies of parliamentary systems, we have no direct measure of MPs' ideology or personal policy preferences. We are unaware of any systematic ideology measure for Canadian MPs and decided against creating our own proxy measure for this study based on public statements regarding the same-sex marriage vote; not all MPs took public stances on the issue in advance of the vote and it is unclear whether such statements should be interpreted as sincere policy preferences. 4. In what would certainly be considered odd by observers of American politics, in which virtually all members of Congress publicly proclaim membership in some religious denomination, only 27 Canadian MPs opted to list a religious preference in the Canadian Parliamentary Guide. While we assume many more MPs are religious, this measure likely taps those who are most open about and motivated by their faith. 5. In an article on Bill C-38 (in which it described same-sex marriage as "a grave offense against God"), Catholic Insight identified 89 members of Parliament as Catholic. The publication admits that this number is "incomplete," since it could not determine the status of 44 MPs from Quebec whom it assumed but could not verify to be Catholic. While not a perfect measure, we believe the Catholic Insight list likely identifies the most publicly Catholic members of the House. 6. Details regarding construction of these variables are available from the authors. 7. The need for this interaction term is demonstrated by comparing the deviances of the models with and without the interaction: the difference is 5.123 with one degree of freedom (p<.05). 8. We also tested interactive effects using the other constituency variables, yet none of these yielded significant results. 9. Indeed, there are reasons to believe that our findings may actually understate constituency effects. First, as noted above, partisanship is itself based in part on constituency preferences. In free vote analyses, where party is included as an independent variable, constituency variables tap only that portion of the effect that pulls against partisan leanings. Second, free votes involve a selection effect (Carrubba et al. 2006; Carrubba, Gabel, and Hug 2008; Kam 2009): party leaders have the option of releasing their backbenchers or not and it is reasonable to suspect they do so only on those matters where countervailing constituency pressures do not threaten to shatter the party completely. 10. In Table 2, this variable become insignificant in models 2 and 3. This is due to multicollinearity problems with the percentage married persons in the riding; dropping this variable restores the significance of the percentage Francophone variable to the p < .01 level. 11. We also tested a model including both interactions from models two and three. The results were that the interaction between Conservative and vote share became insignificant while the interaction between Conservative and first-term status became marginally significant (p = .063). To better show the contrasts between the two effects, we present only the interaction between Conservative membership and first-term status in model 3. 12. Indeed, of the five Conservatives who either abstained or voted yea, only one was a first term MP, and this member (Jim Prentice of Calgary North Centre) was a novice in name only, since he had been involved in the party since 1976, including serving as chief financial officer and treasurer of the Progressive Conservatives. Additionally, another of the three Conservative defectors who voted yea, James Moore, is known for his libertarian views on social issues which have put him in conflict with the more socially conservative wing of the Conservative Party. On his website (http://www.jamesmoore.org/ssm.htm), he has even argued that the government should "not be in the marriage business at all." 13. In March 2010, a whipped Liberal motion to include family planning services in an aid bill to developing countries failed when several of Liberal MPs voted against the party leadership and others failed to attend the vote (Galloway 2010). In September 2010, a Conservative private member's bill to repeal the federal long-gun registry nearly passed because several rural NDP MPs voted to scrap the registry in a free vote that divided their ranks. Several rural Liberal MPs, who voted with the Conservatives on the second reading, also threatened to defy the party whip on the third reading. The Conservatives targeted these NDP and Liberal MPs by campaigning against the registry in their constituencies, warning these MPs that their constituents would hold them accountable for their votes in the following election (Chase and Galloway 2010). 14. The Conservative Party's platform on democratic reform calls, among other things, for "restor[ing] democratic accountability in the House of Commons by allowing free votes. A Conservative Government will make all votes free, except for the budget and main estimates. On issues of moral conscience, such as abortion, the definition of marriage and euthanasia, the Conservative Party acknowledges the diversity of deeply held personal convictions among individual Party members and the right of Members of Parliament to adopt positions in consultation with their constituents and to vote freely."

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.