Senate: Votes on Cloture and Confirmation of John Brennan as CIA Director
Ahead of the start of the sequester tomorrow, below we use Optimal Classification (OC) in R to plot the Senate’s 81-16 vote to invoke cloture on the confirmation of John Brennan as CIA Director and its 63-34 vote on confirmation itself. Senator Rand Paul’s (R-KY) talking filibuster of the nomination the day before garnered considerable popular attention.
Sens. Rand Paul (R-KY) and Ted Cruz (R-TX) (who assisted Paul during the filibuster) voted Yea on cloture but Nay on the confirmation, while Sen. Mike Lee (R-UT) (also a participant in the filibuster) voted Nay on both motions. The sixteen Republican Senators who opposed cloture and the thirty-one Republican Senators who opposed confirmation are mostly among the most conservative members of the caucus, but this issue is one that does not map very clearly onto the liberal-conservative divide. For instance, three of the Senate’s most liberal members—Sens. Patrick Leahy (D-VT), Jeff Merkley (D-OR), and Bernie Sanders (I-VT)—also voted Nay on confirmation. These particular votes were no doubt complicated by partisan considerations; but at least among the Republican caucus, ideology is a fairly good indicator of position on President Obama’s drone program.
Click images to enlarge
As a technical note, we are able to plot members of the 113th Senate so early in the new Congress by combining the roll call voting records of returning members in the 113th Senate with the 32 roll call votes held so far in the 113th Senate. Until more votes are compiled for the freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
Senate: Votes to Invoke Cloture on the Republican and Democratic Sequester Replacement Plans
Ahead of the start of the sequester tomorrow, below we use Optimal Classification (OC) in R to plot the Senate’s 38-62 vote to invoke cloture on a Republican plan to give the president more discretion in applying the spending cuts and its 51-49 vote to invoke cloture on a Democratic plan to make more limited cuts and increase tax revenues. Both votes failed to meet the 60-vote threshold to bring the proposals themselves to a vote.
Eight Senate Republicans voted against cloture on the Republican proposal, a group that was ideologically mixed between moderate Senators like Sen. Susan Collins (R-ME) and very conservative Senators Mike Lee (R-UT) and Rand Paul (R-KY). They eight Republican defectors are also mixed along the second dimension. On the Democratic proposal, four Senate Democrats voted against cloture: three Southern Democrats who face reelection next year (Sens. Kay Hagan (D-NC), Mary Landrieu (R-LA), and Mark Pryor (R-AR)) and Senate Majority Leader Sen. Harry Reid (D-NV), who voted Nay in order to retain his ability to bring the bill back up in the future.
Click images to enlarge
As a technical note, we are able to plot members of the 113th Senate so early in the new Congress by combining the roll call voting records of returning members in the 113th Senate with the 27 roll call votes held so far in the 113th Senate. Until more votes are compiled for the freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
House: Vote to Reauthorize the Violence Against Women Act (VAWA)
Below we use Optimal Classification (OC) in R to plot the House’s 286-138 vote to reauthorize the Violence Against Women Act (VAWA). The vote was in part noteworthy because it marks the third time in the 113th House that the majority Republicans were “rolled” — that is, a bill won passage despite a majority of the majority party voting against it.
House Republicans were also rolled on the Hurricane Sandy relief package, a vote on which Republicans split 49-179 against the measure. The cutting line from the Sandy vote is quite similar to the cutting line from the VAWA vote, with both dividing the most conservative contingent of House Republicans from the rest of the caucus. This intra-party schism also occurs along the second dimension, which we have conjectured may represent an “insider-outsider” divide. Members with low second-dimension scores have been more likely to buck party leadership on a range of votes.
Click image to enlarge
As a technical note, we are able to plot members of the 113th House so early in the new Congress by combining the roll call voting records of returning members in the 113th House with the 55 roll call votes held so far in the 113th House. Until more votes are compiled for the freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
Senate: Cloture and Confirmation Votes of Chuck Hagel for Secretary of Defense
Below we use Optimal Classification (OC) in R to plot the Senate’s 71-27 cloture vote and its 58-41 confirmation vote on Former Senator Chuck Hagel (R-NE) for Secretary of Defense. Eighteen Senate Republicans joined all voting Senate Democrats and Independents to invoke cloture on the nomination, and four Republicans — Senators Thad Cochran (R-MS), Richard Shelby (R-AL), Mike Johanns (R-NE), and Rand Paul (R-KY) — crossed over to support confirmation itself.
The cloture vote more clearly divides the most conservative contingent of Senate Republicans from the remainder of the caucus (with only six classification errors), with the confirmation vote occurring essentially on party lines. Below we include separate plots for both votes, but we first show a combined plot that shows Senators’ voting patterns on both votes. Senators can either vote “YY” (Yea on both votes), “YN” (Yea on cloture, Nay on confirmation), “NY” (Nay on cloture, Yea on confirmation), or “NN” (Nay on both votes). As can be seen, the cutting lines from both votes fan out and do a good job of isolating the moderate slice of the Senate Republican caucus: the 15 “YN” voters who supported ending the filibuster on the Hagel nomination but voted Nay on confirmation. One Senator — Sen. Rand Paul (R-KY) — voted the reverse (“NY”), an odd pairing that may be a product of political maneuvering.
Click images to enlarge
As a technical note, we are able to plot members of the 113th Senate so early in the new Congress by combining the roll call voting records of returning members in the 113th Senate with the 21 roll call votes held so far in the 113th Senate. Until more votes are compiled for the freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
Senate: Vote to Invoke Cloture on the Confirmation Vote of Chuck Hagel as Secretary of Defense
Below we use Optimal Classification (OC) in R to plot the Senate’s 58-40 vote to invoke cloture on the conformation vote of Former Senator Chuck Hagel (R-NE) for Secretary of Defense. The vote fell two votes short of the 60-vote supermajority threshold required to end debate and bring a measure to a vote in the Senate. Note that Senate Majority Leader Harry Reid (D-NV) switched his vote from Yea to Nay at the last minute as a procedural matter so that he can bring the measure back up for a vote after the Senate reconvenes from recess on February 25.
All other 54 Senate Democrats voted Yea and were joined by 4 Senate Republicans: Senators Thad Cochran (R-MS), Susan Collins (R-ME), Mike Johanns (R-NE), and Lisa Murkowski (R-AK). Senator Orrin Hatch (R-UT) voted Present. Where are the Democrats most likely to pick up the extra vote to invoke cloture? In the second panel of the plot below, we display the locations of the 4 Republicans who voted Yea, Senator Hatch, and 4 Republicans who voted Nay but are close to the cutting line: Senators Mark Kirk (R-IL), Dean Heller (R-NV), John McCain (R-AZ), and John Hoeven (R-ND). The Four Yea Republicans are among the 10 most moderate Senate Republicans (according to OC), and so it seems likely that any new Yea votes would come from the moderate strata of the Republican caucus. Senators Kirk, Hoeven, and Heller are also among this group of 10. Senator Hatch, the Present vote, is also close to the cutting line, and so it would be reasonable from an ideological standpoint if he switched his vote to Yea after the recess.
Click image to enlarge
As a technical note, we are able to plot members of the 113th Senate so early in the new Congress by combining the roll call voting records of returning members in the 113th Senate with the 21 roll call votes held so far in the 113th Senate. Until more votes are compiled for the freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
Lincoln and the Ratification of the Thirteenth Amendment
The plot of Best Picture-nominee Lincoln centers around the dramatic passage of the Thirteenth Amendment to abolish slavery in the United States. Following the Senate’s 38-6 passage of the Thirteenth Amendment in April 1864, Daniel Day-Lewis as President Lincoln must maintain Republican support and peel off enough Democrats in order to reach the supermajority threshold for passage of the amendment in the House. On January 31, 1865, the Thirteenth Amendment squeaks by in the House by a 119-56 margin (or 68% support, just above the 2/3 requirement).
The film has been hailed by political scientists for its portrayal of Presidential and Congressional (especially the House) politics and procedure. Below, we use DW-NOMINATE scores to plot the Senate and House votes on ratification of the Thirteenth Amendment. As can be seen below, there is considerable ideological structure to both votes. There are no spatial misclassifications in the Senate vote, and only 10 in the House vote (with all clustered around the cutting line). It appears that most of the defectors (Democrats who voted in favor of ratification or Unionist Republicans who voted against it) were those prone to defect from their party on other votes.
Click images to enlarge
Senator Robert Dole (R-KS) and Polarization in the US Senate
Mr. Ritchie’s observation is backed up by the use of DW-NOMINATE Common Space scores to compare Senators’ ideological positions over time. The DW-NOMINATE Common Space procedure estimates a constant ideal point for legislators over their entire congressional career: a -1 represents the most liberal position, +1 the most conservative position, and 0 the center between the two. Using this metric, Senator Robert Dole’s DW-NOMINATE Common Space score is 0.338.
As shown in the plots below, when Senator Dole first entered the Senate in 1969 (the 91st Senate), he was more conservative than 61% of his Republican colleagues. When he left the Senate 27 years later to focus on the 1996 presidential campaign, he was — as Senate Majority Leader — slightly to the left of the median Republican Senator in the 104th Senate, more conservative than only 44% of Republican Senators. The shift is most dramatic when we compare Senator Dole’s ideological position to the Republican caucus in the most recent, 112th Senate. If Senator Dole served in that Senate, he would be among the most moderate Republicans, less conservative than 79% of Republican Senators.
Indeed, this would place Senator Dole among the group of 8 Republican Senators who voted for ratification of the UN Treaty on Disabled Rights. Senator Dole returned to the Senate floor last December to advocate for the treaty.
Click images to enlarge
House: Vote to Extend the Debt Ceiling for Three Months (No Budget, No Pay Act)
Below we use Optimal Classification (OC) in R to plot the House’s 285-144 vote on the No Budget, No Pay Act. The measure extends the debt ceiling for three months (until 18 May 2013) but requires that the House and Senate (which hasn’t passed a budget since 2009) agree on a budget by April 15. Otherwise, members’ pay will be withheld.
House Republicans voted 199-33 in favor of the measure, while House Democrats opposed it by an 86-111 margin. As can be seen in the plot, the cutting line divides the Democratic Caucus between the more liberal wing (which mostly opposed the bill) and its centrist wing (which mostly supported it). Many of the 33 Republican Nay votes came from the more conservative, “anti-establishment” (with low second dimension scores) faction of the caucus. Consequently, this vote features a “two ends against the middle” pattern, in which the ranks of both ideological poles (i.e., Reps. Michelle Bachmann (R-MN) and Paul Broun (R-GA) / Reps. Alan Grayson (D-FL) and Nancy Pelosi (D-CA)) vote against the middle.
Click image to enlarge
As a technical note, we are able to plot members of the 113th House so early in the new Congress by combining the roll call voting records of returning members in the 112th House with the 28 roll call votes held so far in the 113th House. Until more votes are compiled for the 81 freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
House: Vote on $50 Billion Hurricane Sandy Relief Package
Below we use Optimal Classification (OC) in R to plot the House’s 241-180 vote on a $50 billion Hurricane Sandy relief package. The vote was unique in that the majority party was “rolled”, as House Republicans voted 49-179 in opposition to the bill which nonetheless passed with the support of all but one Democrat.
As can be seen in the plot, the 49 Republicans who voted Yea are not only more ideologically moderate (with a mean first dimension OC score of 0.36, compared to 0.51 for those Republicans who voted Nay), but also have higher second dimension scores. On several votes in the 112th Congress, we conjectured that the second dimension has come to represent am establishment vs. anti-establishment divide. In this case, “outsider” House Republicans (with low second dimension scores) were less willing to sacrifice ideological purity on government spending to rescue the party from controversy surrounding the issue.
Click image to enlarge
As a technical note, we are able to plot members of the 113th House so early in the new Congress by combining the roll call voting records of returning members in the 112th House with the 23 roll call votes held so far in the 113th House. Until more votes are compiled for the 81 freshman members, the estimation of their ideal points will remain somewhat imprecise, but by bridging data from the 112th this problem will not be too severe.
An Update on the Presidential Square Wave
Below we plot the first dimension DW-NOMINATE Common Space scores of the presidents in the post-war period, which we refer to as the “presidential square wave” due to its shape. DW-NOMINATE is a statistical procedure that estimates the ideological positions of members of Congress based on their roll call voting records. In the Common Space procedure, members of Congress who have served in both the House and the Senate are used as “bridges” to estimate scores that are comparable between the two chambers over time. First dimension scores represent the standard liberal-conservative spectrum in American politics, with increasingly positive scores indicating greater conservatism, increasingly negative scores greater liberalism, and the “0″ position denoting the dead ideological center. Presidents are included in DW-NOMINATE scaling by using their “votes” on a subset of roll calls on which the president announces a position on legislation before Congress (measured through CQ Presidential Support Scores).
We find that President Obama is the most ideologically moderate Democratic president in the post-war period, with a first dimension DW-NOMINATE Common Space score of -0.329. President Lyndon Johnson, the second-most moderate Democratic president in this period, has a score of -0.345. President Obama’s ideological position is estimated from his “votes” (statements of support or opposition) on 282 congressional roll call votes. This amount is somewhat low; for example, President George W. Bush “voted” 453 times during his last term in office. However, it is adequate to recover his latent ideological score.
Among members of the 112th Congress, President Obama is ideologically closest to Rep. Jim Moran (D-VA), whose score is -0.328. President Obama is also proximate to Senators Bob Casey (D-PA, -0.332), Kent Conrad (D-ND, -0.326), the recently deceased Daniel Inouye (D-HI, -0.331), and Jeanne Shaheen (D-NH, -0.317). Among Democratic Congressional leaders, President Obama is to the left of Senate Majority Leader Sen. Harry Reid (D-NV, -0.304), and to the right of House Minority Leader Rep. Steny Hoyer (D-MD, -0.395), Senate Majority Whip Sen. Dick Durbin (D-IL, -0.401), and House Majority Leader Rep. Nancy Pelosi (D-CA, -0.530). Interestingly, President Obama has a nearly identical score to former Senate Democratic Leader Sen. Tim Daschle (D-SD, -0.328), whom President Obama had initially nominated for Secretary of Health and Human Services after controversy over unpaid taxes.
Our results may seem surprising to those who consider President Obama among the most — even the most — liberal president of the post-war era. We would respond with a couple of observations. First, President Obama has tacked to the right on many national security and foreign policy issues. For example, President Obama — who once opposed FISA (The Foreign Intelligence Surveillance Act ) — recently supported and signed legislation a five-year extension of the program. Drone strikes — certainly not a source of liberal enthusiasm, even if vocal opposition from the left has been absent under President Obama — have been central to foreign policy during the Obama administration. Furthermore, while foreign policy issues come up relatively infrequently in congressional roll call voting, they constitute a greater proportion of the votes on which the president (i.e., the commander in chief) announces a position.
Second, personal or symbolic factors are not captured in roll call data. DW-NOMINATE scores are estimated only from roll call voting records, and so perceptions of President Obama — for instance — that he is a polarizing president or is aloof towards Republicans are not considered (though, of course, because congressional Republicans have become increasingly conservative, a wide policy distance from them does not necessitate strident liberalism).
We note that there are other ways to quantify latent ideological positions; for example, research by Adam Bonica indicates that President Obama has a donor base that would place him further to the ideological left. However, based on President Obama’s announced positions on actual legislation, we find that he is closer to the ideo