Sometimes you cannot have it all: Party switching and affiliation motivations as substitutes

Abstract

Existing research on when legislators switch parties reports inconsistent results about motivations for switching (e.g., office, ideology, and votes). I treat the motivations for party switching as substitutes and argue that many of the inconsistencies that persist can be explained by modelling the interactive effects between these motivations. For example, scholars differ in terms of whether they find that electoral considerations are an important determinant of party switching. The conflicting findings on the independent effects of electoral considerations are explained here by demonstrating that these effects are conditional on the level of office benefits a legislators enjoys, as well as the ideological distance between the legislator and party. More generally, the empirical analysis provides strong support for the substitution effect hypothesis. Thus, modelling interactive effects increases our understanding of party switching.

Perfect timing: Leveraging the timing of defection to assess the effect of electoral rules on party switching

“…When choosing whether to stay put or to seek affiliation with another party, legislators are confronted with a multichotomous choice, not a dichotomous one (Desposato, 2006; Desposato and Scheiner, 2008; McElroy, 2009; McElroy and Benoit, 2009; Radean, 2019). The reality of there being more than one choice alternative is reflected in the data structure as, for each deputy, there are as many legislator-party dyads as there are parties.…”

“…While the motivations that push legislators to leave their home party are diverse, they can be classified into three broad categories: office benefits, policy considerations, and reelection incentives (Mu ¨ller and Strøm, 1999;Radean, 2019). Legislators are naturally attracted to parties that can provide high levels of private and club goods, as well as to those that are close to them ideologically.…”

“…Thus, each cabinet spell is a renewed opportunity for legislators who seek to maximize office benefits. Therefore, the unit of analysis is legislator-party/cabinet (Desposato and Scheiner, 2008; Radean, 2019). Practically, there is a new entry for all legislators whenever there is a new government coalition—this updates the cabinet element of the unit of analysis, but not legislators’ party affiliation.…”

“…As a result, I can easily identify during which cabinet spell (for which I also have the exact start and end dates) a defection took place. Importantly, defectors reenter the sample with their new party affiliation (reflected by the updated Home Party indicator) in the same cabinet spell because, theoretically, they may switch again during the tenure of that cabinet (see also Desposato and Scheiner, 2008; Heller and Mershon, 2005; Klein, 2016; Radean, 2019).…”

Policy preferences and party switching: Evidence from the 2012 Japanese election

“…Indeed, the most common explanation for party switching in the literature is that politicians choose to switch parties as part of their reelection calculus. That said, part of the difficulty of analyzing this calculus empirically is the fact that politicians may consider several factors when weighing their electoral chances (Kato and Yamamoto 2009; Radean, 2019).…”