Explaining voting behaviour on free votes: Solely a matter of preference?

Abstract

While studies examining free votes find MPs' preferences influence their voting behaviour, most studies also show MPs tend to divide along party lines even after the whips have been withdrawn. Recent work offers a possible alternative explanation for this finding: this sustained party cohesion represents the impact of MPs' party identification similar to party identification effects in the electorate. This argument is tested using a series of free votes on same-sex relations. Even after controlling for preferences using several direct measures, party continues to shape voting behaviour. Although indirect, this provides evidence in favour of the party-asidentification argument. NotesWe would like to thank Pippa Norris for generously sharing the British Representation Study data with us. All responsibility for their use, and especially any errors, belongs to us.

Simply a matter of context? Partisan contexts and party loyalties on free votes

“…Previous research examining the voting behaviour of MPs on un-whipped divisions concludes that 'party' continues to play an important role in deciding the outcomes of these divisions. Although most studies argue the reason parties continue to remain cohesive even when MPs are not compelled by their whips is because MPs of the same party share similar preferences (Hibbing and Marsh, 1987;Marsh and Read, 1988;Mughan and Scully, 1997;Plumb, 2013Plumb, , 2015, other research suggests that parties also remain cohesive due to their MPs' identification with and sense of the loyalty to the party (Raymond and Overby, 2016;Raymond and Worth, 2016;Russell, 2014). Until now, there had been insufficient evidence to draw wider generalisations regarding the effects of party identification across a broad range of divisions, as most of the evidence in favour of party identification effects had been derived from a handful of divisions.…”

“…Finally, MPs may vote with the rest of their party's MPs-even if this means voting against constituents' demands or their own preferencesout of a lingering sense of loyalty to the party. This party loyalty is rooted in psychological identifications with their parties-similar to the party identifications of voters (Butler and Stokes, 1969;Campbell et al, 1960)-that are independent of preference-and disciplinebased effects (Raymond and Overby, 2016;Raymond and Worth, 2016;Russell, 2014).…”

“…This party loyalty is rooted in psychological identifications with their parties-similar to the party identifications of voters (Campbell et al 1960;Butler and Stokes 1969)-that are independent of preference-and discipline-based effects (Russell 2014;Raymond and Overby 2016;Raymond and Worth, 2016).…”

“…Even when preferences are not clearly aligned along party lines, parties often continue to display high levels of cohesion (Andeweg and Thomassen, 2011;Norton, 2003;Plumb and Marsh, 2013). Some research has argued-and found evidence in support of such arguments-that MPs vote along party lines on free votes due to their identification with the party, which creates a sense of loyalty to the party that is independent of personal and constituency preferences (Raymond and Overby, 2016;Raymond and Worth, 2016;Russell, 2014;Van Vonno et al, 2014). As a consequence of the in-group/out-group psychology inherent in party identification, MPs will want to act in the interests of their parties even if they are not compelled by the whips to do so, and even if doing so means voting against their own preferences (Russell, 2014).…”

“…Because MPs are free to choose how to vote on their own, previous research suggests MPs' personal preferences have significant effects on their voting behaviour and that the high levels of cohesion observed on free votes are due to shared preferences among MPs of the same party (Hibbing and Marsh 1987;Marsh and Read 1988;Mughan and Scully 1997;Plumb 2013;ibid 2015). However, a growing body of research suggests these high levels of cohesion are also due to MPs' party identifications, which are independent of shared preferences and the whip (Russell 2014;Raymond and Overby 2016;Raymond and Worth 2016;Raymond 2016).…”