Election 2016: Republican (and Democratic) Polarization in 6 Graphs

Eric M. Uslaner, University of Maryland-College Park

Patrick Egan argues in a post on the Monkey Cage that House Republicans have not become internally divided in recent years.  He argues: "Those seeking a reason for why the GOP appears to have so much trouble governing lately will need to look beyond the explanation that its members are more internally polarized than usual."  His evidence is that the standard deviation of D-W NOMINATE scores (a measure of left-right voting on roll calls in Congress) has not grown over time.  The emergence of a new conservative bloc, the Freedom Caucus, is not a sign of the splintering of the GOP.

Egan rules out the distribution of ideology (as measured by roll call votes) as the explanation for problems of governance among House Republicans.  But there are two other factors that he does not consider.  The first cannot be readily tested with data from roll calls: There is a growing section of the party that simply rejects compromise and threatens to challenge the party leadership—as it did recently in provoking Speaker John Boehner's resignation.  The second is more readily testable: What matters is less party unity than the overall ideology of the party.  The problem, as I see it, is that moderate Republicans have almost disappeared.  There are few members who might build bridges to the other side to make governing less fractious.   And there is little incentive for legislators from either party to do so.  As Republicans in the electorate have become more conservative and Democrats more liberal, the electoral base for moderation has shrunk.  The traditional strategy of seeking the middle, the centrist median voter, no longer is the path to success.  The middle has become political "roadkill," moderates such as Senator Mark Warner (D, VA) and Rep. Jim Cooper (D, TN) say.

On the voteview.com website, Keith Poole presents data over time for the share of moderates in each party in each chamber over time.  Moderates are members with D-W NOMINATE scores between -.25 (moderate liberals) and +.25 (moderate conservatives).  

In 6 graphs below I detail how the center has vanished in both the House and the Senate, for both Republicans and Democrats. The share of Democratic moderates in the House has fallen from 58 percent in 1948 to 13 percent in 2012.  It shrank from 75 percent in the Senate to 13 percent in the Senate over the same time frame.  For Republicans, the share of moderates fell from 56 percent in the Senate in 1948 to 4 percent and from 52 percent in the House to just less than one percent (.4 percent).  

Both the Republican and Democratic share of moderates have fallen over time—and largely in step with each other.  But the patterns may not be as strong as we might expect—and the answer lies in the much steeper declines in moderation among Republicans, especially in the House (see Figures 5 and 6).  There are fewer moderate Democrats because there are fewer Democrats from the South.  There are fewer moderate Republicans because there are fewer Republicans from outside the South.  From the start of these data (1948) until the late 1960s, there were no Republicans from the South and the GOP had substantial representation in both the House and (especially) the Senate from the Northeast.  In the 83rd Congress (1953-1954) Republicans, mostly moderates, held 17 of the 22 Senate seats from the Northeast.  In the 114th Congress (2015-2016), they only held three seats and are in danger of losing at least one of them to a liberal Democrat in next fall's election.  In 13 Southern states, there was a lone Republican (John Sherman Cooper of Kentucky) amidst 25 mostly conservative Democrats in the 83rd Senate.  By 2015, there were just four Democrats (one each from Florida and Kentucky, two from Virginia).  The Northeast was a bastion for liberal Democrats, the South for conservative Republicans.

There are a handful of moderate Democrats in both chambers but hardly any Republicans.   In the 1970s the moderate advantage went to the Republicans, especially in the House.  The share of Republicans below the regression line is now consistently greater than the share above the line—indicating that they have fewer moderates than the Democrats.  Using Sarah Binder's measure of gridlock—the share of laws passed relative to the size of the legislative agenda, I find that fewer moderates lead to legislative action.  The effect is moderately strong for both parties in both houses.  Yet it is stronger for House Republicans (r = -.58) over time than for House Democrats (r = -.46) and for Senate Republicans (r = -.54) than for Senate Democrats (r = -.48).  These may not be large effects (and they are bivariate), but they are consistent and tell the same story that form much contemporary punditry, even among "moderate" conservatives such as David Brooks.

The Republican problem in governing is not its dispersion of ideology.  It is a center that no longer holds.  

This center is not confined to the Congress.  The share of Democratic party identifiers calling themselves liberals has increased from 26 percent in 1972 (the first year in which mass ideology was measured) to 45 percent in 2014 from data in the American National Election Studies and the Pew Center.  The share of Republican conservatives has grown from 42 to 68 percent. The decline in moderates in the House closely tracks the rising ideological identity in the mass electorate: The greater the share of liberals among Democrats the fewer Democratic moderates there are in the House (r = -.8) and the Senate (r = -.73).  And the more Republican identifiers call themselves conservatives, the fewer Republican moderates there are in the House (r = -.8) and the Senate (-.85).  These correlations can't establish causality—whether the public provoked the legislators to become more ideological or vice versa.  They do indicate that the legislative parties are following their electoral bases.  And there seem to be few penalties for doing so.

Eric M. Uslaner is Professor of Government and Politics at the University of Maryland—College Park, Honorary Professor of Political Science at Aarhus University (Denmark), and Senior Research Fellow at the Center for American Law and Political Science at the Southwest University of Political Science and Law, Chongqing, China.