To Moderate, or Not to Moderate: Strategic Media Sharing by Cong. Campaigns

• In Event: Ideology in Times of Polarization

Abstract

Congressional campaigns strategically communicate as they work to achieve their political goals; they signal the candidates’ beliefs and ideology when they make public statements across platforms. In this paper, we take advantage of a new method to estimate the ideology of congressional candidates running in 2020 using their behavior on Twitter (Eady, 2020). This measure is based on URLs of media domains shared on Twitter by candidates. Because the measure is based on behavior, it can be computed over any time period where the candidate shares news on Twitter. By utilizing the ideology of every media link shared by a candidate’s campaign, this method offers temporally granular data which allows us to observe changes in communication of ideology by candidates over time. This means we can estimate an ideology measure for candidates both before a congressional primary, and between the primary and the general election.

We treat the ideology of the media shared as a strategic signal by the candidate, and thus as a measure of the ideology the candidate is seeking to project to voters and other audiences. We are then able to test a long-held conjecture about congressional politics -- that campaigns moderate their message following a successful primary as they look forward to the general election. We calculate these measures not only for incumbents, but also their challengers. We test whether the entrance of more ideologically extreme primary challengers shifts the incumbent further towards the left (for Democrats) or right (for Republicans), and whether they return to the center after for the general election. And we consider whether successful primary challengers then moderate their message in the general election.

We use campaign contribution data from the FEC to proxy for challenger quality to test whether the entrance of primary challengers who are able to successfully fundraise further galvanize incumbents to change their publicly communicated ideology to appeal to primary voters. And we assess whether incumbent behavior change is further incentivized when these contributions come from potential voters in the candidate’s district. As viable primary challenges have become more feasible with the advent of crowd-sourcing funding, understanding how incumbents respond to them has taken on renewed importance. With these time-varying revealed-ideology scores of members of Congress as well as their challengers, we offer a powerful test.

Authors

• Maggie Macdonald, New York University

• Joshua A. Tucker, New York University

• Richard Bonneau, NYU

• Jonathan Nagler, New York University