Search

Can inconsistency in legislators' voting records be explained by the uncertainty they have about their constituency? Why does a two-dimensional model of roll-call voting explain the votes of some legislators better than others? In this paper, we present an uncertainty-based theory of legislative voting and inconsistency. We test this theory by examining whether U.S. senators' uncertainty about their constituency leads to higher levels of inconsistent voting in Congress. Specifically we compare southern senators who faced a large influx of black voters in the 1960s and 1970s with southern senators from more stable constituencies. We also examine other factors that can cause legislator uncertainty, such as legislators' backgrounds. Our results suggest that what some call "maverick" behavior by legislators can be better characterized as best responses by incumbents to a changing environment. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

This paper answers the following question: Under what conditions are states more likely to repeal disfranchisement provisions pertaining to the voting rights of persons convicted of a felony? We also address two broader questions: (1) Do parties seek electoral gain by changing the electoral system and voting laws? And (2) do parties override the ideological preferences of constituents in order to advance their electoral agenda? The answer to both questions is ‘yes’. We find that disfranchisement provisions are more likely to be repealed under a unified Democratic state government, but that states where electoral support for Democrats is generally weak are also more likely to repeal such provisions. We find no evidence that citizen ideology effects changes in voting rights. The implications of our analysis for students of parties and voting rights are clear: Rules are changed by those who stand to benefit, and parties matter at the state level. [ABSTRACT FROM AUTHOR]

In a recent APSR article, Hajnal (2009) explores the question of who loses in American democracy, and finds that African-American voters are significantly more likely than white voters to vote for a losing candidate in recent U.S. elections. This finding is extremely troublesome, as it suggests that African Americans' voices are systematically being shut out of the political process. We revisit this question and show that Hajnal's analyses are too broad in their scope, and that they do not account for the sampling procedures unique to each national exit poll. When we disaggregate his results on the basis of each election year, weight the data according to each exit poll's probability sampling weight, cluster standard errors to take account of the first-stage precinct-wide sampling frame, and examine a longer time period, we arrive at different conclusions. We find a surprising number of elections where minority voters are more likely than whites to vote for a winning candidate. However, we also find many elections where minority voters lose, and we find that Hajnal overestimates the extent to which Latinos and Asian Americans win in some U.S. elections. We conclude that the picture of "who loses" in American democracy may not be as bleak as Hajnal suggests for African Americans, but may be somewhat less positive than he suggests for Latinos and Asian Americans. [ABSTRACT FROM AUTHOR]

Can ideological inconsistency in legislators' voting records be explained by the uncertainty they have about their constituents? Why does a two-dimensional model of roll-call voting explain the votes of some legislators better than others? We present an uncertainty-based theory of legislative search and ideologically inconsistent voting. We test this theory by examining whether U.S. senators' levels of uncertainty about their constituents leads to higher levels of inconsistent voting in Congress. Specifically we compare senators who faced large shifts in the racial makeup of their electorate due to enfranchisement with senators from more stable constituencies. We also examine other factors related to legislator uncertainty, such as legislators' backgrounds. We find that uncertainty due to black enfranchisement and uncertainty due to legislators' backgrounds both lead to higher ideological inconsistency among U.S. senators. Our results suggest that what some call "maverick" behavior by elected officials can be better characterized as best responses to a changing environment. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Does district competitiveness cause members of Congress to change their rollcall voting behavior? Ansolabehere et al. (2001) find that congressional races where the candidates fought tough primaries tend are less polarized than if both candidates won uncompetitive primaries. However, Burden (2004) shows the exact opposite: competitive primaries drive Republican and Democratic candidates further apart on the ideological spectrum. Thus, the effect of primary competitiveness on ideological extremism is still misunderstood. With general election competitiveness, there is a larger consensus showing that competitiveness has a moderating effect on candidate ideology.Whereas prior research treats competitiveness in these two electoral settings as separate entities, we model "district competitiveness" as the product of primary and general election competitiveness. We hypothesize that competitive primaries absent competitive general elections will push candidates away from the middle. Similarly, competitive general elections absent competitive primaries should pull candidates towards the middle. Most interesting theoretically are those cases in which a) neither primary nor general election is competitive, or b) both primary and general election are competitive. We intend to show that the dynamics underlying district competitiveness are more complicated than previously shown and might explain the discrepancies in the literature. We further expect open and closed primaries to pull candidates in different directions.We analyze election data from 1990-2004 and examine every congressional race during that time period in a selected number of states. Our paper should be of particular interest to those who argue that uncompetitive elections are a primary cause of polarization in Congress. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Are legislators who present inconsistent voting records truly mavericks or are they simply responding to uncertainty they have over the preferences of their own constituencies? We investigate the impact of these uncertainty-causing and uncertainty-reducing factors on the (in)consistency of senators’ voting records for the period 1915-2004. We theorize and show that uncertainty about constituents’ preferences explains inconsistent voting behavior by legislators nationally. We also show that for southern senators, the interplay between party and the influx of black voters presents a particularly interesting case of legislator response to a changing environment. ..PAT.-Conference Proceeding [ABSTRACT FROM AUTHOR]

Why do some legislators present ideologically inconsistent voting records to constituents while other legislators are voting in a more consistent manner? We present an uncertainty-based theory of legislative search and ideologically inconsistent voting. We argue that legislators who face a substantial change in constituency composition—through either redistricting or substantial population change—are likely to react by voting in a less consistent and more ambiguous manner than legislators from districts with little constituency change. The ideologically inconsistent roll-call records are rational attempts by the legislator to search for the position of critical supporters on various issues whose preferences are unknown to the legislator.Specifically, we estimate models of ideological inconsistency on all U.S. House members from 1970 to 2006. We measure a legislator's voting record inconsistency by examining the frequency that a legislator's overall ideological position in a given Congress (e.g., NOMINATE scores) does not correctly predict his/her roll calls in Congress. We expect legislators from rapidly changing districts to be more inconsistent than members from more stable constituencies. Furthermore, we expect that legislators whose district boundaries were significantly altered will react to this uncertainty by casting votes that are less ideologically consistent than members whose district boundaries were left unaltered. The effect of redistricting should diminish over the 10-year cycle, as legislators update their beliefs and become more certain about the preferences of their constituents. The effect of redistricting should also be contingent on the processes that lead to the creation of the new districts (e.g., partisan, bi-partisan, court-imposed). Specifically, we predict opposition legislators "targeted" by redistricting to react by voting inconsistently in the new district (conditional on surviving their reelection attempt).This paper extends the theory elaborated by Grose and Yoshinaka (2006). The theoretical framework and empirical results should be of specific interest to students of Congress, legislative roll-call voting, congressional elections, and redistricting, and of a more general interest to students of representation and principal-agent models with uncertainty. Our paper provides a theoretical framework that can explain legislative behavior that is sometimes referred to as "erratic" or "maverick" behavior. Instead, our theory explains this behavior as a legislator's best response to a changing environment. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]