Thomas Carsey is the Pearsall Distinguished Professor of Political Science and the Director of the Odum Institute for Research in Social Science at the University of North Carolina at Chapel Hill. Geoffrey Layman is Professor and Director of Graduate Studies in Political Science at the University of Notre Dame.
The Democratic and Republican parties are as far apart on issues of public policy as they have been in at least a half-century. The clear growth in polarization between the parties’ elected officials, activists and coalitions in the mass electorate is well recognized and has been widely touted by political observers, leaders and scholars (see the reviews by Juliana Horowitz and us as well as Marc Hetherington of the vast academic literature on polarization). However, we believe that much of the public discussion of party polarization misses two important points. First, far from being something new, party polarization has been the natural state of American politics throughout our history. Second, contemporary party polarization may well be different, characterized by what we call “conflict extension.”
Since they first formed at the end of the eighteenth century, the major American parties have nearly always been polarized over some set of policy issues. The most famous and tragic example is, of course, slavery. But battles over issues like the relative power of the national and state governments (embodied in the early debates about a national bank), the Gold Standard, the New Deal, and civil rights resulted in deep divisions between the nation’s political parties. Party polarization is not new. In fact, a primary function of political parties is to clearly organize political conflict, providing some clear policy separation between the parties virtually inevitable.
However, the conventional wisdom in political science says that for most of our history such party polarization applied to only one general policy area at a time. New divisions only emerged when they successfully pushed the old division that had dominated political conflict to the side. This process, called conflict displacement, meant that parties were not locked into polarized positions on every major issue facing the country. The parties might be polarized on one major issue but still able to cooperate on various others.
Our research suggests that the process of conflict displacement has itself been replaced by what we call conflict extension. Thus, while party polarization is not new, growing polarization across multiple issue dimensions may well be new. The graph just below illustrates what we mean with data from the American National Election Studies (ANES) series of national opinion surveys. The graph shows the degree of polarization between citizens who identify themselves as Democrats and Republicans on four different issues that were asked about consistently from 1972 through 2012.
Two questions—whether it is government’s responsibility to ensure that everyone has a job and a good standard of living and whether government should provide health insurance for all citizens—harken to the political divisions over social welfare that emerged with the New Deal. A question about the government’s responsibility to help improve the social and economic position of African Americans taps divisions on civil rights that emerged in the 1960s. A question about the legality of abortion represents the cultural and moral issues that have separated the parties since the 1980s. All issues have been scaled to range from zero for the most liberal position to one for the most conservative position, and the figure shows the difference in party means.
The most important feature is the clear trend toward greater polarization on all four issues. Had conflict displacement occurred, as it did in prior epochs, the increase in polarization on newer issues like abortion would have coincided with less polarization on older issues such as race and social welfare. Instead, polarization on these older issues has actually increased.
Moreover, the same conflict extension is evident among members of Congress and the activists represented at the parties’ national conventions.The graph below is adapted from one of our recent academic publications and shows levels of party polarization on social welfare, racial, and cultural issues (standardized to means of 50 and standard deviations of 25) for the parties in Congress (based on congressional roll-call votes), party activists (based on surveys of national party convention delegates), and the party identifiers in the electorate (based on ANES data).
The same is true for racial issues and cultural issues, as these graphs demonstrate:
The data are clear: across all three major domestic issue areas—social welfare, race, and culture—there has been a steady increase in the gap between Democratic and Republican citizens, elected officials and activists. In short, we have witnessed conflict extension.
These data raise an obvious question: who leads the process of conflict extension and who follows? No doubt trends among citizens, elected officials and party activists all influence each other. However, our research suggests that party activists have been the key catalyst.
Activists are more likely than elected officials and ordinary citizens to champion ideologically extreme positions on new issues and bring them into the party system, and changes in our party system have made it easier for activist groups to enter party politics. Since the 1960s, the Democratic and Republican parties have steadily lowered barriers to participation in party politics. Nomination processes have moved out of the hands of party bosses and into the hands of primary and caucus participants. Activist groups and candidates themselves are increasingly able to finance their own campaigns without the help of parties. In several instances, the parties have actively recruited newly politicized groups into the party rank-and-file.
Thus, part of the reason why party polarization in earlier periods may have been limited to one issue area at a time is that party leaders made it harder for activists and insurgents to enter the party system. This has changed and polarized groups of activists have helped to polarize the parties on multiple policy agendas.
Conflict extension has evolved to such a degree that insurgent groups who in the past might press a single-issue agenda are now likely to advance more strident positions across multiple issues. For example, recent research by David Campbell and Robert Putnam shows that while Tea Party activists tend to emphasize economic and small-government conservatism, they are equally conservative on cultural issues. Similarly, activists from groups like Occupy Wall Street, religious seculars, and members of the LGBT community who are typically associated with liberalism on a single-issue agenda tend to hold strongly liberal views across all issue agendas.
Where does this leave us? Current discussions of party polarization must recognize that polarization in general is not new. The two major parties almost always have disagreed deeply about some policy issues. However, what may be different is the emergence of party polarization across all major dimensions of domestic political debate. That political polarization seems, to most people, to be worse now than in prior eras may be due partly to this conflict extension. Where parties in earlier periods may have found many areas of agreement even as they fought bitterly over some issues, parties today disagree on virtually everything.
This is the latest post in our ongoing series on political polarization. The previous posts include: