Ban versus tax: how to garner support for pro-environmental transport policies, the impact of democratic innovations.

How to cite

Richard, Tanguy. 2025. "Ban versus tax: how to garner support for pro-environmental transport policies, the impact of democratic innovations.." AEA RCT Registry. AEARCTR-0017174. https://doi.org/10.1257/rct.17174-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Tanguy Richard*

* Primary investigator

Registry ID
AEARCTR-0017174
DOI
https://doi.org/10.1257/rct.17174-1.0
First registered
Last updated
Status
In development
Trial dates
2025-11-10 – 2026-02-28
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/17174

Abstract

This study explores the determinants of support for environmental policies in the transport sector. It relies on a discrete choice experiment, in which participants choose between two policies and a policy-free option. Two types of decarbonisation policy for urban transport are presented: one based on pricing and the other on a ban. In addition to the characteristics of the policies, we focus on the impact of the decision-making process on policy acceptance.

Intervention

The objective of this discrete choice experiment is to evaluate the factors that influence support for environmental urban transport policies.
We will also examine the impact of citizen participation in decision-making on policy support.
Respondents will be divided into two groups, each with a different survey design.
In both cases, respondents will be presented with ten choice sets, in which they will be asked to choose between a pricing policy, a ban policy and the absence of a policy. The policies under consideration are Low Emission Zones, which are urban zones where vehicles that do not meet an emission standard are either banned or have to pay a fee to enter.

The two survey designs differ in the attributes used to describe the policies; only one of the designs includes the decision-making process as an attribute.
The other four attributes are:
- The daily cost to enter the urban zone (for the pricing policy only).
- Possible investment in the public transport sector
- Possible adaptations to prevent an impact on low-income households.
- The level of emission reduction.

The main discrete choice experiment will be followed by a set of follow-up questions based on the psychosocial literature.

This discrete choice experiment will result in two articles.
1) The first will analyse the four-attribute sample data (excluding the attribute concerning the decision-making process) and focus on the difference in preference between pricing and ban policies.
2) The analysis of the full sample (including both the four- and five-attribute samples), focusing on the effect of the decision-making process on preferences.

The intervention will be conducted online with the help of a professional company, using a representative sample of individuals residing in France.

Experimental design

A discrete choice experiment consists of repeated choices between different alternatives, which are described through a set of attributes. The levels of the attributes differ between choices, allowing us to elicit respondents' preferences for attributes and their willingness to trade off one attribute for another.

We use a labelled discrete choice experiment in which the three alternatives are: (i) a pricing policy, (ii) a ban policy and (iii) a no-policy alternative.
The attributes used to describe the alternatives are:
- Daily cost (only for the pricing policy): the daily cost of entering the zone
- Investment in the public transport sector: Optional investment to improve public transport
- Policy measure for low-income households: Optional measure to reduce or negate the impact of the policy on low-income households.
- Pollution reduction: The expected level of emission reduction.
Citizen involvement: (Only presented to one half of the sample) corresponds to the way citizens were involved in the decision-making process.

The questionnaire follows the sequence:
- Initial psychosocial questionnaire (also used by the professional company to ensure a representative sample).
- Basic information about pollution in the transport sector and low emission zones.
- [For the part of the sample with the attribute 'citizen involvement'] Additional information was provided about the different ways citizens can be involved in the decision-making process.
- Information that the questionnaire would be shared with policymakers (to mitigate hypothetical bias).
- The discrete choice experiment: ten choice situation, using a best-best DCE design.
- Additional follow-up questions (information relating to their behaviour when choosing cards, evaluation of psychosocial constructs and information about their commuting behaviour and vehicle ownership).

The choice cards were designed using NGene. The two card designs were generated independently. We used NGene to design 40 cards, divided into four blocks of 10.

The survey is conducted online by a professional company, using a representative sample of individuals residing in France.

Design details

Randomization method
The professional company is responsible for the random allocation of respondents to the four or five attributes design and, within the design, to a specific block of ten cards.
Randomization unit
Randomisation is performed at the level of the participant.
Clusters
We will initially consider 2,000 participants (1,000 per design). We will conduct a power analysis using the data collected during the pilot to adjust this number.
Observations
We have ten choice cards for each participant, with each card granting a first and second choice. With 2,000 participants, this corresponds to 2,000 × 10 × 2 = 40,000 observations. This number may change following analysis of the pilot data.
Arms
The sample will be divided equally between each design (1000 participants)
Primary outcomes
The main outcomes consist of the different choices made in the discrete choice experiments. We used a best-best design, in which respondents chose between three options: (i) a policy banning polluting vehicles; (ii) a policy requiring a fee for polluting vehicles; and (iii) no policy. After making their first choice, respondents are presented with the two remaining options and asked to declare their second preference. These choices are used as variables to determine preferences in the statistical analyses.
Secondary outcomes
- Information relating to their behaviour when choosing cards (attribute non-attendance and the systematic choice of the status quo). - Evaluation of psychosocial constructs (environmental concern, perceived legitimacy of low emission zones, perceived problem-solving potential of democratic innovations). - Information about their commuting behaviour and vehicle ownership. - Socio-demographic variables

Cited by

No citation data available for this registration yet.

Keywords