Nothing to Lose: Prospect Theory and Political Radicalism

• In Event: Poster Session: Current and Classical Topics in Political Psychology & State Politics and Policy
  In Poster Session: Poster Session: Current and Classical Topics in Political Psychology

Abstract

What factors drive individuals to support political radicalism, and why are warnings by the political and economic establishment ineffective at preventing people from voting for radical politics? Support for both left- and right-wing political radicalism is on the rise (Rooduijn et al., 2017). Existing explanations highlight economic (Colantone & Stanig, 2018), cultural (Inglehart & Norris, 2017), and political factors (Cramer, 2016; Gidron & Hall, 2020). This paper proposes a novel framework based on prospect theory (Kahneman & Tversky, 1979) centered around the concept of loss to explain ostensibly risk-seeking political behavior and support for political radicalism among those who perceive recent economic, cultural, and social changes as negative. A two-pronged empirical strategy demonstrates the effect of loss on anti-system voting, leveraging panel data and a random forest machine-learning algorithm as well as two online survey experiments.

Research on prospect theory in political behavior usually focuses on explaining status-quo bias in voting behavior and legislative decision-making (Owen, 2011; Soroka, 2014). In this paper, I challenge the view that prospect theory always predicts risk-averse political behavior. I posit a loss effect, whereby those who believe that their economic, cultural, or political conditions have worsened are driven to risk-seeking electoral behavior, such as supporting a volatile anti-system candidate over a status-quo candidate. This occurs through two mechanisms: First, individuals' strong aversion to certain losses implies a highly negative assessment of the status quo choice. In contrast, the volatile choice appears more attractive because the certainty effect implies that people underweight outcomes that are merely probable in comparison with outcomes that are obtained with certainty. As a result of the combined effects of the strong aversion against certain losses and the discounting of the possibility of extremely high losses, decision-makers exhibit risk-seeking behavior if their expectation of the future status quo is negative.

This paper formally proves this loss effect within the framework of prospect theory and derives key hypotheses with which to test and distinguish it from theories of retrospective voting that may initially appear observationally equivalent. Empirical testing proceeds in two forms. First, using German Socio-Economic Panel (SOEP) data and a random forest machine-learning algorithm, I develop a model to predict panelists individual socioeconomic status trajectory based on socioeconomic covariate profiles in 2000. Comparing predicted to actualized status over time allows us to develop a measure of over- and under-achievement (see Kurer & Van Staalduinen, 2022) and sort respondents into gain and loss domains. In contrast to Kurer & Van Staalduinen (2022), I analyze the effect of over- and under-achievement on support for political radicalism separately to test the theoretical predictions arising from prospect theory. Regression models show that under-achievers are more likely to support radical parties, while over-achievers are more likely to support status-quo parties. Moreover, the effect magnitude of marginal underachievement is larger than marginal gains, and differences between marginal and extreme underachievement are smaller than equivalent differences in the domain of gains. Taken together, this provides convincing support for the key tenets of prospect theory.

Second, two online survey experiments fielded in the United States and Germany manipulate perceptions of societal change through a priming intervention before measuring respondents’ propensity to vote for hypothetical radical politicians using a forced-choice conjoint setup. Specifically, respondents read fabricated news articles that frame shifts as neutral, marginally negative, marginally positive, extremely negative, or extremely positive. Then, a conjoint experiment shows various randomized profiles of politicians emphasizing radical policy change versus stability. Estimating average marginal component effects (AMCEs) of the radical attribute across experimental groups produces results that mirror the panel analyses: (1) negative frames increase respondents’ propensity to vote for radical candidates; (2) the magnitude of the effect is higher in the domain of losses than it is in the domain of gains; and (3) the difference between marginal and extreme frames is less pronounced in the domain of losses than in the domain of gains.

By formally integrating prospect theory into analyses of radical voting, this project underscores overlooked psychological dynamics. It suggests volatile candidates need not convince disenchanted citizens that they will necessarily improve conditions but simply that they; it may suffice to signal volatility rather than improvement. Ultimately, prospect theory identifies micro-foundations for the rise of anti-system politics across advanced democracies.

Author

• Florian Sichart, Princeton University