Informational nudge, Eco-anxiety and Climate Policy Acceptance: Experimental Evidence from France

The intervention is based on providing information on climate change through videos of between one and two minutes within an online survey.
Respondents are allocated randomly to one of three treatment groups or to the control group. The videos were developed using available data at both the French and global levels, illustrating how the consequences of climate change—such as extreme rainfall, droughts, and heatwaves—may unfold in the future depending on the type and intensity of climate policies implemented and information regarding collective and individual co-benefits of adopting sustainable mobility and food-related behaviors.

The online experiment tests the impact of different short videos on the individuals' acceptance of a series of climate policies.

Treated respondents receive information either on the frequency and intensity of extreme climatic events in France or worldwide in the years to come if climate policies are or not implemented in the short run or on the health, environmental and financial benefits of adopting sustainable mobility and food-related behaviors.

The first treatment - Local climate change impacts - corresponds to a video providing first information on climate-related events (either drought, heatwave or floods following heavy rainfalls) that occurred in the region of residence of the respondent (its precise location, its magnitude and the extent to which the event is exceptional). The video presents subsequently information on the links between climate change and the type of event previously described. Finally, it displays projections of future events frequency under two scenarios: one in which climate change has been limited thanks to the policies implemented, and one in which it has not been limited. There is a specific video for each region of France.

The second treatment - International climate change impacts - present to respondents a video about the 2023 Canadian wildfires. Again the video is composed of two parts. The first part presents key statistics on the event, the second one links changes in frequency and magnitude of wildfires to climate change following a structure similar to the first treatment.

The respondents allocated to the third treatment - Co-benefits of climate-friendly behaviors - are shown a video presenting the benefits of two individual actions (active mobility for commuting and reduction of meat consumption) in terms of reduction in CO2 emissions, financial savings and individual health outcomes.

Respondents assigned to the control group receive no particular information before being asked about their acceptance of different climate policies.

We will estimate and compare the effects of the different treatments on the level of acceptability declared, immediately after watching the videos, by the respondents regarding a series of climate-related policies. We will also conduct a heterogeneity analysis based on individuals' eco-anxiety, individual's characteristics (socio-demographics and attitudes toward environmental issues in particular).

2025-12-22

2026-01-15

The primary outcomes are the level of acceptance (4 levels) of a series of climate policies declared by respondents within the online questionnaire (i.e. directly after watching the video for respondents allocated to one treatment group). Fourteen climate policies are considered, grouped into three categories:
i) Restrictive policies (e.g., lowering speed limits, mandating home renovations),
ii) Costly policies (e.g., raising the carbon tax, increasing taxes on polluting vehicles), and
iii) Collectively imposed behavioral policies that promote sustainable practices (e.g., requiring public and private firms to offer a vegetarian menu once a week, expanding bicycle lanes).

A binary variable indicating whether the respondent is favorable to the implementation of each policy will be computed by gathering the two highest levels of acceptance and the two lowest levels of acceptance.

An overall average acceptance score will be calculated for each respondent based on the acceptability levels reported for the fourteen public policies, reflecting their general stance toward the full set of climate policies. In addition, separate average scores will be computed for each policy type—restrictive, costly, and those promoting sustainable practices.

This online experiment aims to investigate the mechanisms through which providing information on the intensity and frequency of local or international extreme climatic events—linked to societal action or inaction in terms of implementing mitigation policies, or to the co-benefits of adopting sustainable behaviors—can increase acceptance of various climate policies. The study also examines how individual characteristics, such as eco-anxiety or environmental attitudes, influence policy acceptance. The focus is on respondents representative of the metropolitan French population.

The study will be conducted as a online survey experiment on a sample of 20,000 respondents, representative of the metropolitan French population. Participants will be recruited by an external survey firm that ensures representativeness according to quotas for gender, five age groups, and thirteen regions. Respondents will complete an online questionnaire lasting approximately 15 minutes, comprising six sections:
i) Sociodemographic information, including place of residence over the past ten years,
ii) Political orientations, beliefs about climate change, and information sources,
iii) Adopted sustainable behaviors,
iv) Mental health and eco-anxiety,
v) Exposure to the experimental video treatment (as described above), and
vi) Acceptance of public climate policies.

Random assignment to treatment conditions will allow for causal inference regarding the effects of the information provided on the acceptance of different climate policies, while also examining how individual characteristics (e.g., eco-anxiety, environmental attitudes, etc.) moderate these effects.

Hypotheses :
H1: The level of acceptance of climate policies will be highest among respondents exposed to videos highlighting the co-benefits of sustainable practices (Treatment 3), followed by those exposed to videos of local events (Treatment 1), then those exposed to videos of faraway extreme events (Treatment 2), and lowest among respondents in the control group.
H2 : The impact of treatments on the level of acceptability will be greater for individuals displaying an intermediate level of eco-anxiety.

Randomization is made at the individual level and will be coded directly within the questionnaire form by the external firm implementing the online survey.

The randomization is made at the individual level.

No

20,000 adults currently living in metropolitan France

20,000 adults currently living in metropolitan France

5,000 respondents allocated to the control group and 5,000 respondents allocated to each of the three treatments.

Based on the literature looking at the level of acceptance of similar climate policies in France, the levels of acceptance range between 39% (for urban densification) and 82% (for developping renewable energies). In order to be the most conservative possible, we consider a baseline proportion of 50% of acceptance. For the power analysis, we consider the following parameters: a Type error rate α = 0.05, a power of 1 - β = 0.8, and a relatively small effect size of 0.03. With the above-mentioned parameters 4,356 individuals are needed in each group to compare two by two each treatment arm with the control group.