Search

This paper is the first in a larger project to estimate Congress-Administrator-President (CAP) scores, or ideal point estimates for federal cabinet secretaries, presidents, and members of congress on a common scale. In this paper, we present ideal point estimates for labor secretaries, presidents, and members of the Senate from 1991-2002. Two sets of estimates have been generated: (a) single-congress W-NOMINATE estimates with bootstrapped standard errors (Lewis and Poole 2004), and (b) Markov Chain Monte Carlo (MCMC) methods in a Bayesian framework (Clinton, Jackman, and Rivers 2004; Martin and Quinn 2003). Administrative positions have been constructed through a content analysis of congressional testimony in which an administrator expresses a position on a bill submitted to a roll-call vote in the current Senate session. Results indicate that most labor secretaries have ideal points distinct from their appointing presidents, questioning the assumptions of much of the theoretical literature on the executive branch. Also, using the CAP scores for labor secretaries and senators, we examine the effect of secretary preferences on the allocation of distributive policy projects. We find that the distances between the labor secretary and senators of the same party (Republican) in the 107 th Congress are directly related to the amount of labor department funding allocated to the senator?s state: the further the secretary?s ideal point is from senators of her own party, the less those senators receive in labor outlays. The implications are that the preferences of unelected officials have a direct impact on policy outcomes in Congress. [ABSTRACT FROM AUTHOR]

Does the allocation of ?pork? projects to constituents affect legislative position-taking? Are legislators who deliver substantial amounts of federal largesse more likely to diverge from their campaign opponents and from their constituency medians? Three competing, though overlapping, literatures are examined in this paper: valence theories of position-taking, trust theories of position-taking, and marginality theories of position-taking. These three literatures lead to competing predictions in terms of the extent that legislators deviate from their constituencies and diverge from their campaign opponents given the extent of valence advantages available to incumbents. Examples of valence advantages are numerous (e.g., charisma, constituency service), though we measure an incumbent?s valence advantage as the extent of federal project outlays distributed to a legislator?s constituency. The marginality hypothesis suggests a linear and negative relationship between project allocations and legislator deviation from the district. The trust hypothesis also suggests a linear relationship, though in a positive direction. The valence hypothesis suggests a nonlinear relationship, where increased levels of project allocations at first lead to more legislator convergence toward the median, but eventually lead to more legislator divergence from the median. These competing expectations are tested by examining candidate convergence data from the 1996 U.S. House elections and data on senator divergence from their states? median voters during the 104th-107th Congresses. One key contribution of this paper is the creation of ideal point estimates of legislators and constituency medians on a common scale using MCMC ideal point estimation techniques. The findings are that valence theories are demonstrated when examining incumbent divergence from the constituency median, though these theories are not demonstrated when examining candidate divergence from one another. In addition, we also find that the extent of project outlays (though not deviation from the constituency median) is related to the margin of victory for legislators. The deviation from the constituency median, and not the extent of project outlays, is directly related to the likelihood of winning. [ABSTRACT FROM AUTHOR]

This paper systematically assesses the common claim that the United States is becoming increasingly isolated from the rest of the world in multilateral institutions. It employs a new data set that includes post-Cold War roll-call votes in multilateral institution that are identified as important by the U.S. State Department. A multilevel item-response model is developed that allows for differentiation between trends in the preferences of states and trends in the agenda. The model is estimated using Markov Chain Monte Carlo methods. The results reveal that states have indeed gradually shifted their preferences away from the U.S. from 1991 to 2001. But, these changes are independent from shifts in the agenda. During the first half of the 1990s the agenda became more positive for the U.S. thanks to its multilateral initiatives, whereas unilateralist tendencies led to an opposite shift since 1998. Changes in preferences, however, develop at a constant rate. This suggests that the unilateralist turn in U.S. foreign policy is not solely responsible for the increased gap between the U.S. and the rest of the world. The paper discusses the policy and theoretical implications of these findings as well as interesting deviations from the common trend by individual countries. Check author’s web site for an updated version of the paper. [ABSTRACT FROM AUTHOR]

This paper examines the perspective from which we should select moral principles to guide states as they decide whether or not to engage in humanitarian intervention. I distinguish between first person and legislative perspectives. In the former, agents select principles as the best for them to interpret and apply; in the latter, they select principle for themselves as well as others. I argue that, given the non-ideal world in which we live, a legislative perspective is superior. I suggest how the legislative point of view sheds light on three different "non-ideal facts" about the world (the bias and fallibility of human agents, the lack of correlation between justice and power, and the unpredictable contingency of political outcomes) as they relate to the problem of humanitarian intervention. The legislative point of view can account for the main arguments about how a principle of humanitarian intervention should be articulated. The legislative point of view will generally direct us toward a more cautious principle of intervention. [ABSTRACT FROM AUTHOR]

Current democratic theorizing is based either on the classical model of democracy, which assumes that self-rule and political equality are achievable, or on the procedural democracy of the modern nation state, which accepts that the presence of political and civil liberties is a fundamental right. The weakness of the former model lies in its assumption that self-rule and political equality are possible in modern nation states, whereas the latter fails to account for the variation in the level of influence or power that citizens possess due to differences in income and knowledge. Given these deficiencies, I formulate an open-ended theory of democracy. Such a system goes beyond the mere presence of political rights and civil liberties and allows for a continuous achievement of a more equal distribution of power. I hypothesize that the main driving force for the dynamic nature of democracy is economic development. In addition, two other factors, the political process and external factors, could facilitate the process of democratic development. Democracy would continue to evolve indefinitely but because of the presence of some inherent tension between liberty and equality would fail to achieve the ideal democracy. However, I hypothesize that as the level of economic development grows over time, the levels of income, knowledge, and power will be normally distributed. I call such a state of politics normal democracy. [ABSTRACT FROM AUTHOR]

Do party preferences or constituency preferences affect legislator positions in support of the president? Spatial theories have long implied that legislators will locate themselves at the median of their district. However, most empirical evidence related to roll-call support of the president’s agenda in Congress indicates the opposite, showing that party and ideology often cause a legislator to deviate from the district median (and that constituency preferences thus have a negligible or muted effect on presidential support). We answer our research question and explain this paradox by showing that legislators’ platforms are different from legislators’ votes on roll-calls. Spatial election theories are based upon the platforms that legislators take when running for office. The platforms communicated to constituents may be related to roll-call voting records, but clearly are not one and the same. Our findings show that constituency preferences are the only significant predictor of legislative support for the president when analyzing legislators’ platforms, and that party preferences have no effect. However, when analyzing roll-call votes, our analysis supports the large body of research on executive-legislative relations, suggesting that party and ideology are the key predictors of presidential support. We analyze legislators’ platforms and the messages they send to their constituents with newsletters and other mass mailings from the 107th Congress. The data were collected from the Legislative Resource Center and coded by the authors. [ABSTRACT FROM AUTHOR]