LSE - Small Logo
LSE - Small Logo

Blog Admin

January 25th, 2016

Primary election candidates change their campaign strategies in response to both their current and (potential) future opponents

1 comment

Estimated reading time: 5 minutes

Blog Admin

January 25th, 2016

Primary election candidates change their campaign strategies in response to both their current and (potential) future opponents

1 comment

Estimated reading time: 5 minutes

Share this:

Kevin Banda 80x108Thomas Carsey 80x108Do candidates respond to their opponents during primary elections? If so, which opponents do they engage with: those running against them in the same primary or those who they might face in the general election? In new research, Kevin K. Banda and Thomas M. Carsey show that US Senate and gubernatorial candidates alter their television advertising strategies in response to the ads aired by both their primary election opponents and their potential general election opponents. These strategies are driven in part by the competitiveness of their primary elections; candidates who won their primary elections are more likely to react to their eventual general election opponent during the primary stage of the campaign.

As the Iowa caucuses near, the Democratic and Republican presidential nomination campaigns continue to become more intense. Though Election Day is nearly a year away, the candidates will face many state level primary elections between now and then. Members of the news media and political pundits usually bemoan the lack of discourse in American politics, but is this actually a problem? In the Democratic race, Hillary Clinton and Bernie Sanders have recently had a series of spirited exchanges about endorsements. On the Republican side, Donald Trump and Ted Cruz have begun airing attack advertisements against one another in Iowa. In addition, Democratic candidates occasionally discuss the Republican candidates (usually Trump) whereas Republican candidates sometimes talk about the Democratic candidates (usually Clinton). Thus, it appears that at least some discourse takes place during primary elections, but how much discourse, and why does it occur?

In new research, we observe the degree to which candidates involved in primary elections engage in discourse with the two sets of opponents with whom they are faced: their primary and potential general election opponents. We argue that the two-stage nature of American elections – e.g. candidates must win their party’s nomination before competing in the general election for a given position – leads candidates to adopt a mixed strategy that weighs both sets of opponents in order to come up with a strategy that maximizes the likelihood that they will win both the primary and the general elections. Acknowledging this two-stage nature is important because the people who participate in primary elections tend to have different preferences than do those who participate only in general elections. Candidates need to craft messages that can reach these two groups knowing that what they say during a primary may limit what they can say during the general election – candidates rarely have a true “Etch a Sketch” moment.

We examine the advertising behavior of candidates involved in 56 US Senate and gubernatorial primary elections. More specifically, we focus on the confluence of issue ownership and advertising. Issue ownership is the notion that some issues are so strongly associated with specific parties that they become “owned.” In US politics, the Republican Party is typically viewed as owning issues like taxes, defense, and crime while Democratic-owned issues include health care, education, and welfare. Our research centers on the attention given by candidates to these families of related issues.

The level of competitiveness in a primary election should affect the degree of responsiveness to a candidate’s opponents exhibited by that candidate in their advertising strategies. That is, when faced with competitive opponents, candidates respond to one another by devoting greater attention to the families of issues stressed by their opponents. For example, when Candidate A’s opponent in the primary airs more ads discussing Republican-owned issues, we expect Candidate A to respond by airing more ads that focus on Republican-owned issues. At the same time, Candidate A should also be responsive to their potential general election opponent(s) running in the opposing party’s primary. Because of the two-stage nature of US elections, we expect that candidates who face competitive primary election opponents will be more responsive to those opponents than to their potential future opponents running for the other party’s nomination.

We test these propositions in two contexts. First, we look at contests in which only one party’s primary was contested. Figure 1 shows the average number of advertisements mentioned either Democratic or Republican-owned issues that candidates air in response to an increase of 100 ads of the same type aired by their opponents. For example, the panel on the left of Figure 1 shows that when a winner of a contested primary airs 100 additional ads mentioning Democratic-owned issues, the other candidates involved in that primary increase the number of ads they sponsor mentioning Democratic-owned issues by about 50. At the same time, when the eventual loser(s) of a contested primary increase the number of advertisements mentioning Democratic-owned issues by 100, the eventual winner responds with an increase of nearly 70 such ads. Similarly, as shown in the right panel of Figure 1, when candidates in contested primaries increase the number of Republican-owned issue ads they air by 100, the candidates running uncontested in the other party’s primary increases the number of Republican-owned issue ads they sponsor by 17 for eventual primary winners and 27 for eventual primary losers.

Figure 1 – Candidate Responsiveness in Contests with One Contested Primary

Banda Fig 1

Second, we also observe how candidates respond to their current and potential future opponents’ advertising strategies in contests in which both parties’ primaries were contested. These relationships are plotted in Figure 2 and are largely consistent with the results described above: candidates alter their advertising behavior in response to the advertising strategies employed by all of their current and potential future opponents, but generally respond more strongly to the former. For example, the light gray bar in the panel on the left of Figure 2 shows the number of additional advertisements mentioning Democratic-owned issues that Democratic primary winners air when each of their three sets of current or potential opponents sponsor 100 new ads of the same type: about 60 in response to their primary opponents and approximately 17 in response to either the eventual winner or the loser(s) of the Republican primary.

Figure 2 – Candidate Responsiveness in Contests with Two Contested Primaries

Banda Fig 2

On the whole, our results show that candidates adopted mixed campaign strategies that accounted for both stages of their elections, particularly when their primaries were strongly contested. Our analysis shows that candidates routinely shape their issue agendas during primary elections in response to the agendas of the candidate(s) running against them in the primary. Such candidates are clearly focused on the first stage of the two-stage election process. However, candidates who eventually win their primaries also showed strong responsiveness to their eventual general election opponent even during the primary election phase of the campaign. In other words, candidates involved in two-stage elections are influenced during the first stage by the behavior of the opponent they are likely to face in the second stage.

Our research implies that we can expect the patterns of candidate response that we have seen so far in the presidential primaries to continue. It is likely that Democrats will continue to respond to one another and to at least the front running Republican candidates like Trump and Marco Rubio, and that Republicans will similarly continue to alter their messaging strategies in response to one another and to Democrats – particularly Clinton.

This article is based on the paper “Two-Stage Elections, Strategic Candidates, and Campaign Agendas,” published in Electoral Studies. 

Featured image credit: Gage Skidmore (Flickr, CC-BY-SA-2.0), Hillary for America (Flickr, CC-BY-NC-SA-2.0)

Please read our comments policy before commenting.          

Note:  This article gives the views of the author, and not the position of USAPP – American Politics and Policy, nor the London School of Economics.

Shortened URL for this post:


About the authors

Kevin Banda 80x108Kevin K. Banda University of Nevada, Reno
Kevin K. Banda is an Assistant Professor of Political Science at the University of Nevada, Reno. His research interests include campaigns, elections, public opinion, and political communication. His research has appeared in journals like Public Opinion Quarterly, Electoral Studies, and American Politics Research. 


Thomas Carsey 80x108Thomas M. CarseyUniversity of North Carolina at Chapel Hill
Thomas M. Carsey is the Thomas J. Pearsall Distinguished Professor in the Department of Political Science and the Director of the Odum Institute for Research in Social Science at the University of North Carolina at Chapel Hill. His research interests center on electoral behavior, campaigns, political parties, public opinion, state politics, and legislative politics in the U.S, along with quantitative methods. His research has appeared in journals like the American Political Science Review, American Journal of Political Science, and the Journal of Politics.

About the author

Blog Admin

Posted In: Elections and party politics across the US | Kevin K. Banda | Thomas M. Carsey


LSE Review of Books Visit our sister blog: British Politics and Policy at LSE

RSS Latest LSE Events podcasts

This work by LSE USAPP blog is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported.