The differential impact of climate interventions along the political divide in 60 countries

Abstract

A major barrier to climate change mitigation is the political polarization of climate change beliefs. In a global experiment conducted in 60 countries (N = 51,224), we assess the differential impact of eleven climate interventions across the ideological divide. At baseline, we find political polarization of climate change beliefs and policy support globally, with people who reported being liberal believing and supporting climate policy more than those who reported being conservative (Cohen's d = 0.35 and 0.27, respectively). However, we find no evidence for a statistically significant difference between these groups in their engagement in a behavioral tree planting task. This conceptual-behavioral polarization incongruence results from self-identified conservatives acting despite not believing, rather than self-identified liberals not acting on their beliefs. We also find three interventions (emphasizing effective collective actions, writing a letter to a future generation member, and writing a letter from the future self) boost climate beliefs and policy support across the ideological spectrum, and one intervention (emphasizing scientific consensus) stimulates the climate action of people identifying as liberal. None of the interventions tested show evidence for a statistically significant boost in climate action for self-identified conservatives. We discuss implications for practitioners deploying targeted climate interventions.

Fig. 1. Climate change belief, policy support, and action predicted by self-reported political ideology.

Belief in climate change (Panel A), climate policy support (Panel B), and climate action (Panel C), as a function of self-reported political ideology (measured from 0 = liberal to 100 = conservative), in the control condition (N = 4302). Vertical lines represent 95% bootstrapped confidence intervals of the means at each level of political ideology; Fitted lines represent the best fit linear regression lines.

Fig. 2. Polarization of climate change belief, policy support, and action.

The degree of polarization (where higher scores reflect greater polarization), operationalized as the absolute value of the difference between self-reported liberals’ and conservatives’ climate beliefs (Panel A), policy support (Panel B), and action (Panel C), in the control condition (N = 4302). Each country’s means and confidence intervals of these polarization scores for each outcome can be found in Figs. S1–S6.

Fig. 3. Relations between self-reported political ideology, climate belief, and climate action in the control condition (N = 4302).

Panel A: Standardized scores of climate belief (in pink) and climate action (in gray) as a function of self-reported political ideology. Panel B: Mean climate action of self-reported liberals (in dark blue) and conservatives (in red) as a function of mean climate beliefs. For both panels, fitted lines represent the best-fit linear regression lines; Error bands represent 95% confidence intervals.

Fig. 4. Impact of each intervention on climate change beliefs, policy support, and action, separated by self-reported political ideology.

Interventions’ impacts on climate beliefs (Panel A), policy support (Panel B), and action (Panel C), split by self-reported political ideology (liberals in blue and conservatives in red; N = 51,224). Vertical lines indicate the average in the control condition for each ideological grouping. Error bars represent 95% confidence intervals of the means. Upward triangles indicate significant increases, downward triangles indicate significant decreases, and circles indicate no statistically significant differences, always compared to the control.

Fig. 5. Interventions’ effects on self-reported liberals’ and conservatives’ (N = 51,224) climate beliefs, policy support, and action, compared to the control condition.

Green check marks indicate significant increases compared to control, red X marks indicate significant decreases compared to control, and empty cells indicate no statistically significant differences compared to control. The coefficients of these analyses can be found in Tables S1–S6.