Volume XXXIII, Number 2
May 2008




        This issue opens, as did the last one, with further evidence that the subject of agenda formation continues to receive prominent attention from legislative scholars. Gary W. Cox and Mathew D. McCubbins have collaborated with William Heller, an expert on European legislative politics, to test the procedural cartel theory of agenda control that they developed in their book on Setting the Agenda: Responsible Party Government in the U.S. House of Representatives (2005). The three authors investigate agenda power in the Italian Chamber of Deputies to determine whether, in that seemingly chaotic parliament, the governing majorities use agenda power to prevent being overruled on the floor of the chamber. They examine the “roll rates” experienced by governing and opposition parties between 1988 and 2000. The authors explain the procedures available to the government in Italy to either obtain a consensus on the agenda or to circumvent it by using emergency decrees to substitute for legislation. They find that the governing parties are almost never rolled, while opposition parties are rolled over 20% of the time—the roll rate increasing with the distance of a party’s ideal point from the ideal point of the floor median. The government’s agenda control rests on its ability to block bills from the agenda and results principally from its use of emergency decree power and of the budget process, both of which follow a different procedure from ordinary legislation. The unusually great delaying power available to the opposition reduces the roll rates of the opposition parties compared to their counterparts in other European parliaments. The authors find that government control is impressive, confirming the procedural cartel theory that Cox and McCubbins developed in the setting of the U.S. House of Representatives.

        Frances Lee investigates the influence of the agenda in a very different setting, that of the U.S. Senate. She finds that the conspicuous increase in partisanship in the Senate over the past 25 years is not entirely due to exogenous factors such as changes in Senate membership and changes in the electorate. Lee constructed a dataset classifying Senate roll-call votes between 1981 and 2004 according to their ideological content and their effect on polarization between the parties. She finds that social issues that are less divisive have been increasingly replaced on the Senate’s agenda by more polarizing economic issues. Using a multivariate analysis, she shows that this change in the agenda accounts for nearly two fifths of the increase in party polarization in the Senate. While leaders of the Senate do not enjoy the procedural control over the agenda available to leaders of the U.S. House, collective strategies among individual members in the Senate have a systematic effect on the agenda that helps to explain the rise of partisanship in that chamber.

        The next two articles deal with two aspects of representation, descriptive representation and electoral accountability. Thomas L. Brunell, Christopher J. Anderson, and Rachel K. Cremona bring new aggregate- and individual-level data to bear on the question of whether African Americans’ attitudes toward Congress and toward their representative are affected by the racial homogeneity of their district and by whether their member of Congress is African American. The authors used survey data from the National Black Election Study as well as demographic information on congressional districts. They find an interaction between the effect of the presence of a large number of black voters in a district and its representation by a black member. The first imparts a sense of empowerment that enhances African Americans’ views of their member of Congress—whether or not that member is black—and enhances their view of Congress as a whole. The second, that of having a black representative in Congress, affects African American voters positively toward that member but it affects their attitude toward Congress as a whole only if they perceive that blacks are underrepresented there. In short, having a black representative does not independently affect attitudes toward Congress as a whole. Those attitudes are affected most strongly by perceptions of how extensively blacks are represented in Congress overall. That conclusion has implications for the continuing controversy over whether the racial composition of congressional districts should have an influence on redistricting.

        The second article on representation deals with the influence of constituents’ approval of their Senator on his or her reelection. A new set of data on job approval ratings make possible this previously neglected subject of research. Benjamin Highton examines data for the period 1986 to 2004 and finds that job approval has an independent effect on senators’ reelection, much as presidential approval affects the outcome of presidential elections. Constituents’ approval of their senator influences reelection by influencing the senator’s propensity to campaign again and the probability of attracting effective challengers.

        It has long been known that by strategic voting on the floor of a legislature, opponents of a bill that appears to have majority support can kill it if they can successfully attach an amendment to it that causes some of the bill’s original supporters to turn against it. But relatively few “killer amendments” have ever been identified in the U.S. House of Representatives and the exact conditions under which such amendments succeed have not been established. This is what Charles J. Finocchiaro and Jeffery Jenkins set out to do. Searching through all roll-call votes in the U.S. House between 1953 and 2004, they identify just five cases. The authors show that for a killer amendment to succeed, some of the supporters of the original bill must find it impossible to vote strategically against the amendment that would kill it because the amendment itself is so attractive to their constituents that they could not explain their vote against it. The authors then develop a procedure for identifying killer amendments involving a spatial analysis of the relevant roll-call votes to identify the necessary multidimensionality of the relevant issues. That they find astonishingly few cases in half a century of roll-call votes does not diminish the importance of the phenomenon because it is likely that the threat of killer amendments affects the legislative process at stages prior to floor votes. They also suggest that killer amendments may occur more readily in legislatures other than in the U.S. House.

        One of the distinctive powers of the U.S. Congress is its power to investigate the executive branch and to hold hearings. By contrast, the oversight function is limited in parliamentary systems because the party composition of the executive branch tends to mirror that of parliament. That raises the question of whether the investigatory activity of Congress varies between united and divided government. Douglas Kriner and Liam Schwartz have extended the research of David Mayhew on this subject. Their analysis is based on 35 high profile investigations conducted by the U.S. House of Representatives between 1947 to 2002. They find that a shift from unified to divided government increases the number of investigations by more than 25%, regardless of the ideological distance between President and Congress. Divided government stimulates investigatory activity especially when the opposition party is cohesive, lending support to the theory of ­conditional party government. Initial examination of the pattern of Senate investigation produces similar if less conclusive results. Kriner and Schwartz have therefore found new support for the proposition that partisanship matters in executive-legislative relations in the United States, helping to explain a contrast between presidential and parliamentary systems.

        The final article in this issue exemplifies the value of comparative research across U.S. state legislatures. Jesse Richman investigates the relationship between the preferences of the median member of the legislature and the members of its committees, an important aspect of legislative organization. Are committee “outliers” dominated by narrow interests or inliers serving the informational needs of the average member? Keith Krehbiel’s informational model predicts that outliers will exist only when the floor is uncertain about its committees. Richman examines the role of uncertainty in shaping the opportunity for outliers, using data from roll-call votes for 1999 and 2000 in U.S. state legislatures. By looking at this subject across U.S. state legislatures, he obtains far greater variance in information and uncertainty than scholars who have looked only at the U.S. Congress. He confirms that committee outliers are more common when legislators are ill-informed compared to their committees. His findings have implications for explaining the declining deference to committees in Congress as ideological uncertainty has diminished. Richman’s article reinforces the nuanced predictions of Krehbiel’s informational model of legislative organization.

                                                                                                     —Gerhard Loewenberg,                                          
Comparative Legislative Research Center


John C. Wahlke, lead author of the path-breaking comparative study of four U.S. state legislatures, The Legislative System (1962), died in April at the age of 90. Wahlke’s work influenced the conceptualization of legislatures, informed legislative theory, and spurred comparative research. When he heard of the establishment of this Quarterly, he instantly became its first subscriber. His favorite description of a scholar he admired was to call that individual "broad gauged." That describes him best. He inspired two generations of legislative scholars and was a champion of this journal from the start.


 Return to May 2008 Titles

 Return to LSQ home page