Volume XXX, Number 3
August 2005

Editor’s Introduction


Journalists know that the reshuffling of cabinets in parliamentary systems is politically important, but political scientists have rarely studied this practice. That omission is even more surprising since there is a large body of literature on cabinet stability, presumably out of the belief that stability is important because it contributes both to competence in governing and clear lines of accountability to the electorate. But in fact, cabinet stability as measured by the duration of governments is not at all the same thing as stability in the membership of cabinets. In the United Kingdom , where cabinets have been among the most stable in postwar Europe , turnover in the membership of cabinets is high and ministers are therefore relatively inexperienced because of frequent reshuffles. Christopher Kam and Indriđi Indriđason investigate why prime ministers reshuffle their cabinets. With data on changes in the composition of cabinets in five democracies over 40 years, they analyze six models of the possible factors influencing reshuffles. They conclude that prime ministers reshuffle their cabinets in order to maintain their position in response to declining electoral popularity and challenges from within their own party. They show that reshuffling occurs most frequently in party and political systems that limit the power of the prime minister, and at times when the parliamentary or electoral popularity of a prime minister is declining. The article points to the need to give greater attention to the role and consequences of cabinet reshuffling in parliamentary politics.

The role of party in the U.S. Congress continues to be a puzzle. A steady stream of research has followed Keith Krehbiel’s work in the last decade that raised the question of whether parties play any independent role at all. Two articles in this issue of the Quarterly add pieces to solving this puzzle. Jeffrey A. Jenkins, Michael H. Crespin, and Jamie L. Carson analyze the voting behavior of members of Congress in the last six months before an election, comparing three sets of members: those who seek reelection and presumably have incentives to vote with their party as well as in responsiveness to their constituents, those who seek higher office and therefore have only party concerns, and those who are retiring and are presumably free of party and constituent pressure. They test predictions based on the Cox and McCubbins cartel theory that holds that the purpose of party is to give the majority organizational control. They find that, consistent with that theory, members of the majority party who are retiring from Congress move away from their parties on matters of procedure—but not on substantive matters—as they approach the end of their term when only their own preferences continue to matter. The authors interpret that move as evidence of the prior influence of the majority party in procedural and organizational matters.

Another way to identify the impact of party is to compare the voting behavior of party switchers with members who stay with their party. Antoine Yoshinaka pursues this line of analysis by exploring the committee assignments of party switchers to see whether the new party that they join disproportionately favors them. His measure is the incidence of violations of the seniority norm. Yoshinaka’s analysis shows that members who switch do receive favorable treatment, easing their party transition and demonstrating that party leaders influence committee assignments.

The influence of gender on legislative behavior is another long-standing subject of research. Michele L. Swers uses cosponsorhip activity on issues of traditional importance to women to assess the impact of gender in the U.S. House. Her research builds on previous work that used cosponsorship as an indictator of members’ entrepreneurial activity. Swers shows that women were disproportionately among cosponsors of bills on education, children and family issues, and health issues, but not on welfare bills.

Among the most satisfying research results are those that show where conventional wisdom is wrong. The final article in this issue does just that by reexamining the known connection between divisive primaries and close election outcomes. Jeffrey Lazarus analyzes that relationship in U.S. House elections between 1976 and 1998. He disentangles the confounding effect of incumbent vulnerability using a game theoretic model. The author demonstrates that perception that a seat is winnable increases the number of candidates entering a primary and increases the prospects that the primary will be divisive. But it is just such elections that are likely to be close because they are races in which the incumbent seems vulnerable. Thus, Lazarus concludes that divisive primaries, rather than being the cause of close general elections, are the unintended consequences of the perception of candidates that an election is likely to be close. The direction of causality, the author suggests, is the opposite of the received wisdom on this subject.

                                                                                                                                                                —Gerhard Loewenberg

  Return to August 2005 Titles

  Return to LSQ home page