Ideological Dispersion of the Parties in Congress

A recent New York Times Magazine article examined ideological differences between the progressive and centrist wings of the Democratic Party, using the potential Democratic Senatorial primary race in Maryland between Reps. Donna Edwards and Chris Van Hollen as an example of “the great Democratic crack-up.”

Below we use DW-NOMINATE scores to illustrate the ideological positions of Democrats and Republicans in the current House and Senate (Note: these scores are from a new stand-alone Common Space DW-NOMINATE program that can be run weekly as new roll calls are cast; data available here.)

In the present (114th) Congress, it is clear that Republicans occupy a wider swath of ideological territory than Democrats. It is easy to identify distinct ideological clusters of Republicans, but not for Democrats (a point we make here). The distance between, for instance, Republicans Senators Lisa Murkowski and Ted Cruz dwarfs that between Democratic Senator Elizabeth Warren and former Senator Hillary Clinton. In fact, the distance between Democratic Representatives Donna Edwards and Chris Van Hollen is more comparable with the distances between Senators Lisa Murkowski and John McCain or between Senator McCain and Representative Paul Ryan.

Click image to enlarge


The images below show trends in the ideological dispersion of the parties in Congress since 1879 by plotting the variance in DW-NOMINATE scores among House and Senate Democrats and Republicans in each Congress. Higher values indicate greater ideological dispersion within the parties in both chambers.

From a peak in the mid-twentieth century (when the Democratic Party was split between its Southern and non-Southern wings), the ideological variance among House and Senate Democrats has steadily declined. Congressional Democrats are now more ideologically unified than at any point since the early 1900s. The ideological variance of Congressional Republicans has been flatter over the last century, but there is some suggestion of a jump in dispersion among Senate Republicans in the 112th Congress (following the 2010 midterm elections).

Click images to enlarge



Ideology and the Republican Split on the Medicare ‘Doc Fix’ Bill

Below we use Common Space DW-NOMINATE scores to examine the ideological positions of the 33 House Republicans who opposed the Medicare ‘Doc Fix’ bill. These scores represent House members’ positions along the liberal-conservative spectrum, with higher scores indicating greater conservatism.

The 33 House Republicans who voted Nay are among the most conservative members of the House Republican Caucus. Their mean score is 0.640, compared to a mean score of 0.497 for House Republicans as a whole. The group of Republicans who have been an enduring headache for Speaker John Boehner do appear to be very conservative (both in contemporary and historical terms), suggesting that the split between “establishment” and “outsider” congressional Republicans is not just a difference in personal styles, but also rooted in ideological differences.

This image is from a new (nearly) stand-alone Common Space DW-NOMINATE program that can be run weekly as new roll calls are cast.

Click image to enlarge


New Estimates of Polarization in the 114th Congress

Below we use Common Space DW-NOMINATE scores to compare polarization in the current (114th) and previous Congresses. Common Space scores permit comparability across time and between the House and Senate.

The distribution of the ideological (first dimension) scores of the House and Senate Democrats and Republicans in both Congresses are shown below, with the mean scores of members in both parties and chambers marked in the plot.

The results indicate that polarization (i.e., the ideological distance between the two parties) has increased slightly in the House, but has decreased slightly in the Senate. The entry of freshman Republicans such as Senators Mike Rounds (R-SD) [with a score of 0.189], Ben Sasse (R-NE) [0.277], and Dan Sullivan (R-AK) [0.285] has moved the Republican Senate mean slightly to the left in the 114th Congress. In the House, both the Democratic mean and the Republican mean have moved away from the center, but only slightly. The long-term polarization trends in both chambers are shown in the second plot.

These images are from a new (nearly) stand-alone Common Space DW-NOMINATE program that can be run weekly as new roll calls are cast.

Click image to enlarge



Congressional Policy Shifts, 1879-2014

Below we use first dimension DW-NOMINATE scores, which represent legislators’ positions along the familiar ideological (liberal-conservative) spectrum, with lower (negative) scores indicating greater liberalism, and higher (positive) scores denoting conservatism, to plot the chamber means for legislators and winning outcomes on roll calls for Congresses 46 to 113 (2014). These are updates to Figure 4.1 (page 60) in Poole and Rosenthal’s Congress: A Political-Economic History of Roll Call Voting and Figure 4.1 (page 80) of Ideology and Congress.

We see that the overall chamber means (plotted with red squares) remain mostly stable over time, a reflection of a competitive two-party system. Of the two chambers, the House mean has shifted more to the right in the period following the 104th Congress (the 1994 “Republican Revolution”). However, the position of the mean winning coordinate in each chamber has proved much more volatile, particularly in recent Congresses. This reflects the frequency of party-line votes between rival partisan coalitions that have moved steadily apart in recent decades.

Consequently, the mean winning coordinate–-which is an approximation of the ideological location of policies enacted in the chamber–-has diverged from the overall mean chamber score in both chambers: to the left under Democratic control, to the right under Republican control. Indeed, the mean winning coordinate in the 111th Senate (a session in which the number of Senate Democrats fluctuated between 57 and a supermajority of 60) was the furthest to the left since the 75th Senate during the Great Depression and the New Deal. Further, the mean winning coordinate in the 113th House was the most conservative since the late nineteenth and early twentieth centuries, following the realigning election of 1896, after which Republicans controlled the House for 32 of the next 38 years. Conversely, the mean winning coordinate in the 113th Senate was the most liberal since the late nineteenth century.

Click images to enlarge



House: Vote on Clean DHS Funding Bill

Below we use updated DW-NOMINATE scores to plot the House’s 257-167 vote to pass a clean funding bill for the Department of Homeland Security (DHS).

The cutting line (separating predicted Yea votes from predicted Nay votes) divides the 75 Yea Republicans from the 167 Nay Republicans along both the first dimension (representing liberal-conservative position) and the second dimension. The meaning of the second dimension has largely shifted from representing regional differences within the parties (e.g., between northern and southern Democrats) to intra-party divisions that are more subtle and less clear. One of these divisions appears to be an “insider vs. outsider” cleavage that pops up on votes such as raising the debt ceiling, domestic surveillance, and government funding bills.

This image is from a new stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click image to enlarge


Note: The plot shows only 256 Yea votes because Speaker John Boehner (R-OH) voted Yea but has not cast enough votes (25) to be included in the scaling.

Senate: Vote to Override Keystone XL Pipeline Veto

Below we use updated DW-NOMINATE scores to plot the Senate’s 62-37 vote to override President Obama’s veto of a bill approving construction of the Keystone XL pipeline. The Senate fell five votes short of a successful veto override.

Eight Senate Democrats joined all 54 Senate Republicans in supporting the override. These eight Democrats are among the most moderate members of their party’s caucus, and DW-NOMINATE accounts for this in projecting the cutting line as running through the right edge of the Senate Democrats.

It is also worth noting the degree of ideological dispersion among Republicans in the 114th Senate, which runs the gamut between moderates like Senators Mark Kirk (R-IL) and Lisa Murkowski (R-AK) and strong conservatives like Senators Ted Cruz (R-TX) and Rand Paul (R-KY), with a first dimension (ideological) distance of approximately 0.8 between these two clusters.

This image is from a new stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click image to enlarge


Note: The plot does not include Senate Minority Leader Harry Reid (D-NV), who voted Nay but has not yet cast enough votes to be included in the scaling.

House: Vote on Three-Week DHS Funding Bill

Below we use updated DW-NOMINATE scores to plot the House’s 203-224 vote to reject a three-week funding bill for the Department of Homeland Security (DHS).

Ideology appears to do a good job of dividing the 191 Yea Republicans from the 52 Nay Republicans, with (predictably) those Republicans opposing the measure being more conservative than those supporting it. There is somewhat of a “two-ends-against-the-middle” pattern on this vote, with the 12 Democrats who voted Yea being more moderate than the remainder of their caucus.

This image is from a new stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click image to enlarge


Note: The plot shows only 202 Yea votes because Speaker John Boehner (R-OH) voted Yea but has not cast enough votes to be included in the scaling.

An Early Look at Polarization in the 114th Congress

Now that the 114th House has conducted 66 roll call votes and the 114th Senate has voted 53 times, we now have enough data to take a very preliminary look at ideological polarization in the new Congress. Following standard practice, we use first dimension DW-NOMINATE scores as measures of legislators’ liberal-conservative positions.

The first two plots below show the mean score of Democrats (Northern and Southern) and Republicans in both chambers over time. Though we are hesitant to put too much stock in these results at this early date, it looks like polarization (the ideological distance between Democrats and Republicans) is on pace to take a slight uptick in the 114th House but may level off in the 114th Senate. In the Senate, this is due to a slight shift in the Republican mean back to the center. Some of this may be due to the exit of Senator Tom Coburn (R-OK) [whose score is 0.792] and the entry of Senators Cory Gardner (R-CO) [0.216] and Mike Rounds (R-SD) [0.398]. There may also be a procedural aspect as Majority Leader Mitch McConnell (R-KY) has allowed several votes on amendments.

Interestingly, the defeat of Democratic moderates like Senators Mark Begich (D-AK), Kay Hagan (D-NC), Mary Landrieu (D-LA), and Mark Pryor (D-AR) has had very little effect thus far in moving the Democratic mean leftward. To some degree, their exit may be counterbalanced by the retirements of liberal Democrats like Senators Tom Harkin (D-IA) and Carl Levin (D-MI).

The second set of images shows the percentage of Democratic and Republican legislators in both chambers with DW-NOMINATE scores less than -0.5 or greater than 0.5, making them more ideologically extreme. There has been little change in these values between the 113th and 114th Congresses, with the exception of a marked increase in the proportion of House Republicans with scores greater than 0.5.

These image are from a new stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click images to enlarge





The House and Senate Votes to Approve the Keystone XL Pipeline

Below we use updated DW-NOMINATE scores to plot the House’s 266-153 vote and the Senate’s 62-36 to approve the Keystone XL oil pipeline.

28 House Democrats and 9 Senate Democrats joined all voting Republicans in supporting the measure. As seen in the plots below, NOMINATE does a good job of modeling voting patterns on the basis of ideology, with moderate Democrats in both chambers predicted to defect from the rest of their party.

These images are from a new Stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click images to enlarge



The Speaker Vote: 6 January 2015

Updated 30 January 2015

Below is the vote to elect John Boehner (R-OH) as Speaker of the House. The actual roll call was 216 for Boehner, 164 for Nancy Pelosi (D-CA), 24 Republicans voted for other Republicans, 4 Democrats voted for other Democrats, Babin (R-TX) voted Present, and 25 members did not vote. Grimm (R-NY) resigned before the vote was taken.

The 24 Republicans who did not vote for Boehner were treated as voting for Pelosi and the 4 Democrats who did not vote for Pelosi were treated as voting for Boehner. This was done solely for display purposes. Most of the non-voters were Democrats — 20 of the 25. The opposition to Boehner was from the right-wing of the Republican caucus and the 4 Democrats opposed to Pelosi were relatively moderate.

This image is from a new stand-alone DW-NOMINATE that can be run daily as new roll calls are cast. We will have more to say about this software at a later date.

Click image to enlarge