Institutional and Sociological Perspectives on Parliamentary Behaviour and Party Cohesion

and Background: This is the penultimate empirical chapter of my dissertation (indeed, it is still structured as a chapter rather than a standalone paper). The dissertation itself investigates what factors lead Members of Parliament to toe the party line or to dissent. In the aggregate, this amounts to asking what factors foster or undermine party cohesion. Previous chapters demonstrated that a simple economic model of parliamentary behaviour advocated by Krehbiel (1993, 1999) – in which MPs vote simply vote their preferences (quite aside from what their party leaders might desire) – is not empirically supportable. This chapter considers two alternative explanations: an institutional explanation that focuses on the party’s control of career advancement, and sociological explanation that sees loyalty and cohesion as a function of group norms. The institutional perspective assumes that MPs are rational actors intent on securing re-election – but considers the possibility that they also desire career advancement, a good that, in parliamentary systems, is controlled by party leaders (in contrast to a seniority system as is employed in the U.S. congress). Hence, MPs frequently confront a difficult tradeoff: dissent from the party line in order to preserve their electoral prospects or toe the party line in the hope of securing advancement. Party cohesion in parliamentary systems is very much a product of how MPs ultimately come to terms with this trade-off. The sociological perspective (e.g. Searing 1994) takes a much different tack, and indeed balks at casting party cohesion in such starkly rational choice terms. Instead, the MP’s loyalty (and hence aggregate party cohesion) is seen to be due to the internalization of norms of loyalty and group solidarity. Students of parliamentary politics often cite the confidence convention as the key contributing factor to the high levels of party cohesion that are typically observed among parliamentary politics (Huber 1996; Diermeier and Feddersen 1998). I consider the above explanation in light of this institutional constraint, but point out that parliamentary politics is only rarely played out under conditions in which the survival of the government is at stake: most votes, even if they result in a government defeat, are simply not matters of confidence. The empirical analysis supports this argument, and demonstrates that the confidence convention is not generally a sufficient condition for party cohesion. Using aggregate from Great Britain, Canada, Australia, and New Zealand, I examine the evidence in favour of the two perspectives. This analysis looks at the frequency, depth, and extent of backbench dissent in Great Britain, Canada, Australia, and New Zealand over the post-war period. The results indicate that even after one controls for the size of the government’s majority (hence the degree to which confidence is a real threat), dissent is more strongly connected to the size of the ministry than to turnover in the parliamentary party. This strongly indicates that it is the availability of career opportunities rather than socialization (which is difficult to do properly when personnel turnover is high) that contributes to the cohesion of parliamentary parties. These results support the institutional perspective on parliamentary behaviour at the expense of the sociological perspective, and in doing so cast doubt on the utility of employing social norms and roles as the chief independent variables in models of parliamentary behaviour.