What Causes Congressional Polarization: Changing Politics, or Changing Politicians?

A new study coauthored by Assistant Professor Daniel Moskowitz of the University of Chicago Harris School of Public Policy sheds new light on the increasing polarization in Congress, challenging existing assumptions about partisan divides and finding that the replacement of legislators with more extreme successors is an important driver of polarization. The study, which analyzes data from surveys of congressional candidates over time, provides a clearer distinction between polarization driven by legislators growing more extreme, polarization arising from the changing legislative agenda set by party leaders, or changes by the replacement of retiring legislators with new legislators who are more extreme. 

Historically, polarization in Congress has been measured primarily through roll-call votes, which assess how members of different parties vote on various issues. However, a key limitation of this approach is that changes in polarization may be more reflective of an evolving legislative agenda—one shaped by party leaders—rather than shifts in the ideological positions of legislators themselves. Specifically, when party leaders prioritize divisive issues for roll-call votes, it can create the appearance of increased polarization, even if individual legislators' positions have remained relatively constant.

"The prevailing narrative on polarization has largely relied on analyses of roll-call votes," Moskowitz said. "However, the increasing roll-call divide between the parties could be an artifact of a legislative agenda that is increasingly restricted to items that neatly divide the parties. Party leaders can select issues for votes that are more likely to divide the parties, and this could distort our understanding of how polarized Congress truly is. Our approach allows us to hold constant the issue agenda so that we can assess whether legislators have in fact grown more ideologically polarized.”

A key contribution of the paper is its use of survey data from congressional candidates, who are asked to state their positions on the same set of issues during each election cycle. By analyzing how these positions have shifted over time, the study allows researchers to assess whether members of Congress are adopting more extreme positions independently of changes in the legislative agenda. This method provides a more nuanced picture of polarization, distinguishing between two primary mechanisms: adaptation (where existing members of Congress adopt more extreme positions over time) and replacement (where new members, who hold more extreme positions, replace those who hold more moderate stances).

The authors find that almost all of the increase in polarization is explained by candidate replacement rather than shifts in policy positions by sitting officeholders. Moskowitz and his co-authors also reveal that polarization is not equal across all policy fronts: Most is taking place in the areas of social and environmental policy. The paper also finds that a majority (but not all) of the increase in extremism comes from the Republican Party.

The study's insights are crucial for understanding the dynamics of legislative behavior and the ongoing political polarization in the United States. By isolating the effects of ideological shifts from agenda-setting strategies, this research lays the groundwork for future studies that can better inform the public and policymakers about the true causes of political division in Congress.