What’s in a name? Gauging the effect of labels on third party vote shares

ABSTRACT

In this paper we examine an understudied question: Does the inclusion of a party label boost the vote shares of third party candidates? For years, third parties in Tennessee have contended that the high barrier to appear on the ballot with a label is a decided disadvantage for their candidates. Our empirical analysis, however, tells a different story. Various federal court decisions have allowed third party candidates in Tennessee to appear on the ballot with their respective party label in the 2000, 2012, and 2014 election cycles. This fact creates a ready-made natural experiment whereby the electoral fortunes of third party candidates with a label can be compared to those running without a label. Using data from 1992 through 2016 we test whether the inclusion of a label for third party candidates enhances their vote share. The results of our analysis make it clear that in a dominant two-party system it makes no difference, in terms of electoral viability, whether or not a third party candidate appears on the ballot with a party label.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Tennessee Code Annotated § 2-5-101 (b). At present, there are 11 presidential electors allocated to the State of Tennessee.

2 Tennessee Code Annotated § 2-1-104 (a) (30).

3 Tennessee Code Annotated § 2-13-107 (a). The previous deadline for third parties to submit a ballot access petition was 119 days prior to the date of the August primary election.

4 Tennessee’s election code does not require third parties to nominate candidates by primary elections. Following changes in 2012, third parties are allowed to nominate candidates using any method authorized under the rules of the party or by primary election (Tennessee Code Annotated § 2-13-203 (a)). If a third party does not utilize a primary for candidate nominations, ballot access for the general election using party labels requires the 2.5% signature requirement submitted within the 90-day window preceding the general election (Tennessee Code Annotated § 2-13-107 (a)). If a third party does rely on a primary for nominations, ballot access using party labels requires the 2.5% signature requirement to be completed 119 days prior to the date of the primary (Tennessee Code Annotated § 2-13-107 (a)(1)).

5 Following the 2000 election cycle third parties whose presidential candidates received a minimum of 5% of the total vote cast were permitted to appear on the ballot with a label in the subsequent election cycle. See Libertarian Party of Tennessee v. Goins, 793 F.Supp.2d 1064 (2010).

6 Green Party of Tennessee v. Hargett (3:11cv-692). August 17, 2016.

7 Green Party of Tennessee v. Hargett (3:11-cv-692). August 17, 2016.

8 Green Party of Tennessee v. Hargett (3:11-cv-692). August 17, 2016.

9 Write-In candidates are not included in the analysis presented.

10 Some third party candidates included in the analysis ran more than once during the time period under study.

11 Even in these election cycles not all third party candidates qualified to run with a party label.

12 In Tennessee, the only statewide constitutional office elected is the Office of Governor.

13 The only Reform Party candidates who appeared on the ballot with a label ran for the presidency in 2000. Following a legal battle over the nomination, John Hagelin and Patrick J. Buchanan were both listed under the Reform Party label on the Tennessee ballot in 2000.

14 Looking at the controls for office-holding level, non-major party vote shares for state senate elections and governor were statistically indistinguishable from the comparison category of state house elections. In reference to federal elections, however, non-major party candidates perform notably worse. The model predicts that in comparison to state house races non-major party vote shares for U.S. House elections are 4.1 percentage points lower. For statewide federal contests, the drop-off in the non-major party vote share is 3.5 points for presidential races and 3.0 points for U.S. Senate races. Overall, Model 1 explains 70% of the variance in non-major party vote shares in Tennessee from 1992 through 2016.

15 Currently, the major parties occupy all but one seat in the U.S. House of Representatives (Michigan Congressman Justin Amash switched from Republican to independent on 4 July 2019). All but two seats (those of political independents Angus King of Maine and Bernie Sanders of Vermont) in the U.S. Senate are controlled by Democrats and Republicans; and one of the independent senators, self-proclaimed democratic-socialist Bernie Sanders, ran for president in 2016 (and now 2020) as a Democrat because he wanted to win. Excluding officially nonpartisan Nebraska and current vacancies, Democrats and Republicans occupy over 99% of state legislative seats. Put another way, in the more plentiful state houses (N = 5356 seats) there are 30 members not affiliated with one of the major parties. In state senates (N = 1915 seats), there are just 3 legislators who are not Democrats/Republicans. Data are from the National Conference of State Legislatures (NCSL) website and are available at the following link: http://www.ncsl.org/Portals/1/Documents/Elections/Legis_Control_2019_August%2026th.pdf.

Additional information

Notes on contributors

M. V. Hood

M.V. Hood III is a Professor of Political Science and Director of the School of Public and International Affairs (SPIA) Survey Research Center at the University of Georgia.

Seth C. McKee

Seth C. McKee is an Associate Professor of Political Science at Texas Tech University. McKee is the author of Republican Ascendancy in Southern U.S. House Elections (Routledge 2010), the editor of Jigsaw Puzzle Politics in the Sunshine State (University Press of Florida 2015), and author of the textbook, The Dynamics of Southern Politics: Causes and Consequences (CQ Press 2019). McKee serves on the editorial boards of American Politics Research and Political Behavior, and is the Editor in Chief of Political Research Quarterly.