![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
Green parties have been quintessential issue owners since their founding. In recent national and European elections, however, these parties have begun to emphasize additional issues. This article seeks to understand when the greens expand their issue focus to talk about economic issues. It develops a theory of issue expansion which posits that the greens’ issue expansion is related to both party competition and economic context. In particular, it posits that the size of the radical-left party and the rate of unemployment influence the degree to which green parties focus on the economy in their election manifestos. It then tests the theory on a dataset of European green parties which have run in national elections between 1981 and 2017 in 14 West and East European countries. The findings have important implications for understanding party competition in general and the issue evolution of new politics parties more specifically.
Notes
Acknowledgements
Author order follows the principle of rotation. Both authors were equal contributors. We thank participants at the ‘European Elections and Attitudes towards the European Union Workshop’ hosted by the Jean Monnet Centre Montréal, Lawrence Ezrow, and the West European Politics reviewers for their comments and suggestions. All errors remain are own.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 Grant and Tilley (2019) find that mainstream parties’ accommodative strategies can increase green party vote share once they have survived several elections by increasing the salience of the environmental issue. Meijers and Williams (Citation2019) also demonstrate that an established party can only undermine a challenger party if it can apply accommodative tactics before the ‘reputational entrenchment’ of the challenger party as the issue owner (see also Meguid Citation2005).
2 In their efforts to attract voters, the greens often compete with both social democrat and radical-left parties. Below (note 12) we discuss how controlling for social democratic vote share, however, does not influence greens’ issue emphases.
3 A similar logic may be applicable to other challenger parties. That is, when an issue area adjacent to a challenger party is addressed by another party in a system, the challenger party may increase its focus on the set of issues it owns. Conversely, if an issue area adjacent to a challenger party is left unaddressed by another party, the challenger party may seek to expand its issue focus to include that area.
4 We have excluded Portugal as the Partido Ecologista os Verdes (Ecologist Party ‘The Greens’) as it is allied with the Partido Comunista Portugês (Portuguese Communist Party) in national elections under the Coligação Democrática Unitária (Unitary Democratic Coalition). We also recognize that there is variation among green parties; however, due to the limited size of the sample, we cannot disaggregate the parties into subgroups.
5 As some previous research has relied upon a narrower understanding of environmental issue emphasis (e.g. Abou-Chadi Citation2016; Spoon et al. 2014), we also ran models (see Online appendix, Table A.3) in which environmental issue emphasis is restricted to per501 (support for environmental protection). The results are similar to the models which use the broader operationalization of environmental emphasis. As further robustness checks, we also ran models in which environmental issue emphasis is defined using different combinations of these issue dimensions. All of these models have results similar to the main models (1 and 2).
6 To examine this possibility, we ran a series of models. The first of these models focused only on the esoteric and ideologically-driven indicators not originally included in our analysis (per 402, 405, 408, 409, 410, 411, 412, 413, 414, and 415—see the Manifesto Project codebook for a full description of these dimensions). Using this operationalization for economic emphasis results in the same results as found in Model 1. Lagged radical-left party vote share is statistically significant and positive, while unemployment is statistically insignificant and negative. Using this operationalization for economic emphasis also results in the interaction term having no effect in Model 2. This is what one would expect if parties do not talk about more ideologically-driven issues during an economic crisis. Further, we ran models in which all the indicators were included in a measure of economic emphasis. The results were very similar to those in which only per 402, 405, 408, 409, 410, 411, 412, 413, 414, and 415 were included. Again, this is what we would expect if parties do not talk about ideologically-driven issues during an economic crisis.
7 For histograms of all dependent variables, see Figures A1 through A3 in the Online appendix.
8 Importantly, parties discuss many issues and it is rare that one issue will completely dominant the agenda. When more than 10% of a manifesto focuses on a single set of issues, it strongly suggests that it is an issue owner.
9 Ideally, we would replace unemployment with the Gini coefficient as a robustness check; however, Gini coefficient data does not extend earlier than the mid-1990s in some countries, and no earlier than the mid-2000s for many countries. Including the Gini coefficient results in only 28 observations remaining. It is thus difficult to draw any conclusions. However, as a robustness check, we ran tests in which the change in the Gini coefficient for a country between year t and year t-1 rather than unemployment is included. The results for these truncated models largely follow the results from the main models. Most importantly, the interaction terms are statistically significant and in the expected direction.
10 While it is plausible that this distinction may be important for some research questions, in our case, we believe a radical-left party which receives 0% of the vote has the same qualitative effect on the green party’s issue emphasis as would the absence of such a party.
11 As an alternative measure of the economy, we substituted GDP growth rate for unemployment and interacted it with the lagged vote share of the radical-left party. The interaction term is positive but statistically insignificant. This is expected as changes in GDP are less likely to be felt by voters, and therefore unlikely to capture growing public salience of the economy.
12 Models in which we control for whether a green party was in government at the time of an election suggest no change in the effects of the main independent variable. Further, models in which we controlled for whether a radical-left party exists in a system, and whether a radical-left party served in government, respectively, showed nearly identical results to those reported in the main models. Models in which we included lagged vote share of the social democratic party in a system were similar to the main models of this study.
13 We have decided to not include a measure of public support for the environment as we would lose a substantial number of observations because such a question was not asked in all election studies. The green party’s vote share in the previous election does capture environmental issue salience. Further, it is possible that as green parties become more successful, they face mounting pressure to expand their issue emphases beyond traditional environmental issues. This possibility should be explored in future research.
14 When we include decade dummy variables, the results are similar to the main models.
15 As a robustness check, we ran a single-level model with country fixed effects. The results of these models are similar to the main models.
16 It is important to note that if the coefficient on radical-left party vote share was negative, this would mean that green parties discuss environmental issues less relative to economic issues.
Additional information
Notes on contributors
Jae-Jae Spoon
Jae-Jae Spoon is Professor of Political Science and Director of the European Studies Center, University of Pittsburgh. Her research focuses on electoral competition and political party strategy, primarily in Europe. Her work has been published by the Journal of Politics, British Journal of Political Science, the European Journal of Political Research, Comparative Political Studies, and the University of Michigan Press, among others. [[email protected]]
Christopher J. Williams
Christopher J. Williams is Assistant Professor in the School of Public Affairs, University of Arkansas at Little Rock. His research examines political party strategy, elections, voting behavior, public opinion, and public policy in Europe and the United States. His work has been published in the Journal of European Public Policy, Social Science Quarterly, Electoral Studies, European Union Politics, the European Political Science Review, and the Journal of Elections, Public Opinion and Parties, among others. [[email protected]]