As part of this project, I was asked to head a sub-group that focused on how rising levels of polarization and gridlock have impeded negotiation, compromise and good governance. By design, the polarization sub-group reflected a variety of viewpoints with respect to the nature and effects of polarized politics in the United States. But in the spirit of a task force on negotiation and compromise, I believe that the report reflects a reasonable consensus about what political scientists know (and, more importantly, don’t know) about partisan polarization. So the report serves as a useful point of embarkation for our Monkey Cage series on polarization.
Here are the main points of agreement, briefly stated. Readers are encouraged to check out the actual report for the details.
- Based on both qualitative and quantitative evidence, the roots of our current polarization go back almost 40 years to the mid-1970s. Indices of polarization based on roll call voting in Congress have been nearly monotonical in both chambers of Congress since around 1978. This evidence is primarily important for the explanations of polarization that it rules out. It casts doubt on explanations focused on more contemporary events such as the Clinton impeachment, the 2000 presidential election, the election of Barack Obama or the emergence of the tea party. That both chambers have been affected suggests a limited role for explanations based on the institutional differences between the House and the Senate. The timing is much more consistent with explanations based on large historical trends such as the post-Civil Rights realignment of Southern politics and increased levels of economic and social inequality.
- The evidence points to a major partisan asymmetry in polarization. Despite the widespread belief that both parties have moved to the extremes, the movement of the Republican Party to the right accounts for most of the divergence between the two parties. Since the 1970s, each new cohort of Republican legislators has taken conservative positions on legislation than the cohorts before them. That is not true of Democratic legislators. Any movement to the left by the Democrats can be accounted for by a decline in white representatives from the South and an increase in African-American and Latino representation.
- The increase in party polarization has also reduced the dimensionality of political conflict. Many issues that were once distinct from the party conflict dimension have been absorbed into it. Congressional voting can be increasingly accounted for by a single dimension that distinguishes the parties. This situation directly contrasts with that of the mid-twentieth century, when the parties divided internally on a variety of issues, especially those related to race and region. While public opinion remains more multi-dimensional than congressional voting, there is a similar trend among the public.
- While significant disagreement persists as to how much voters have polarized by taking increasingly extreme views, there is a consensus that voters are much better sorted within the party system. Conservative voters are much more likely to identify as Republican and liberals as Democrats than two generations ago. Moreover, voters’ partisanship increasingly predicts their positions on issues. Voters are primarily changing their issue positions to match the partisanship rather than switching parties. Since voters seem to be responding to the positions of their party leaders, the causal arrow seems to run from elite polarization to partisan sorting. Whether partisan sorting has an additional feedback effect on elite polarization is less clear.
- Features of our electoral system such as political gerrymandering and partisan primaries are not likely to be important causes of polarization. That the House and Senate have polarized in tandem suggests that partisan districting cannot be a primary cause and researchers have failed to find much of an incremental contribution. Similarly, scholars have not identified any substantial impact of the primary system on polarization. The relationship between our system of private campaign finance and polarization is more complex. While there is little evidence that the origins of greater polarization lie in campaign finance, the growing participation of ideologically oriented donors appears to have exacerbated the problem.
- Polarization in Congress derives from both sincere ideological differences about policy means and ends and strategic behavior to exploit those differences to win elections. The combination of high ideological stakes and intense competition for party control of the national government has all but eliminated the incentives for significant bipartisan cooperation on important national problems. Consequently, polarization has reduced congressional capacity to govern. Congress has been less productive in legislation, more prone to delays in appropriating funds, and increasingly slow in handling executive and judicial appointments. While hard to quantify, there is considerable evidence for a decline in the quality of legislative deliberation and legislation. Of significant concern is the extent to which this reduction in legislative capacity has contributed to a shift in the constitutional balance as it enhanced opportunities for executive and judicial encroachments on legislative prerogatives.
In the next few weeks, we’ll be asking various political and social scientists to weigh in on different questions related to political polarization, both in government and among American voters. Those posts will tackle the puzzle of polarization from various angles. Some will ask whether ideological polarization is really the problem in today’s Congress, while others will examine the causes and extent of polarization among the mass public. In this series, we’ll have multiple posts about the effects of the changing media landscape–and multiple posts on polarization in other developed democracies.
Author’s note: The membership of my working group on U.S. polarization included many notable scholars, journalists, and public servants including Andrea Campbell, Thomas Edsall, Morris Fiorina, Geoffrey Layman, James Leach, Frances Lee, Thomas Mann, Michael Minta, Eric Schickler, and Sophia Wallace. The report was co-authored with Michael Barber.