Back to History Current Version

Consumer Preferences for Transparent Carbon-Neutral Labels: A Choice Experiment

Last registered on April 02, 2024

Pre-Trial

Trial Information

General Information

Title
Consumer Preferences for Transparent Carbon-Neutral Labels: A Choice Experiment
RCT ID
AEARCTR-0012520
Initial registration date
March 27, 2024

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
April 02, 2024, 11:18 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Primary Investigator

Affiliation
Eawag: Swiss Federal Institute of Aquatic Science and Technology & University of St.Gallen

Other Primary Investigator(s)

Additional Trial Information

Status
In development
Start date
2024-04-05
End date
2024-07-30
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This paper focuses on the impact of transparency in carbon-neutral labeling on consumer preferences and willingness to pay. In current markets, there is a proliferation of various climate labels representing different concepts but overlapping characteristics, which may lead to consumer confusion. Furthermore, theoretical evidence suggests that a proliferation of labels might promote those labels with lower environmental quality. Therefore, this study aims to empirically understand whether providing information on the percentage of CO₂ offsetting and CO₂ reduction influences consumer preferences for carbon-neutral labels. I conduct a discrete choice experiment among UK tea consumers through an online survey. I use a split sample approach, comparing consumers' willingness to pay for standard carbon-neutral labels and “transparent” carbon-neutral labels, which additionally disclose the percentage of CO₂ offsetting and CO₂ reduction. Furthermore, I examine the effect of various factors, including concerns about carbon offsetting, trust, and confusion, on consumers' preferences for standard and transparent carbon-neutral labels.
External Link(s)

Registration Citation

Citation
Ozdemir Oluk, Begum. 2024. "Consumer Preferences for Transparent Carbon-Neutral Labels: A Choice Experiment." AEA RCT Registry. April 02. https://doi.org/10.1257/rct.12520-1.0
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
The intervention is "the transparency of the carbon-neutral label," which varies across three subsamples, where transparency refers to the percentage of CO₂ offsetting and CO₂ reduction.
Intervention Start Date
2024-04-05
Intervention End Date
2024-07-30

Primary Outcomes

Primary Outcomes (end points)
The primary outcomes are:
 Respondents' choices between the three choice alternatives (two tea product alternatives and a 'none of the two' option) on each choice card,
 Marginal WTP (MWTP) estimates for carbon neutral label from each subsample.
Primary Outcomes (explanation)
Respondents' choices between the three alternatives on each choice card will be coded as binary variable, where 1 indicates the chosen alternative, and 0 otherwise. Multinomial logit (MNL) and random parameters logit models (RPL) will be used to estimate the average MWTP for a carbon-neutral label from each sub-sample.
Based on the MNL model and the RPL model (where I randomize the carbon-neutral label parameter, but not the price parameter), the MWTP is obtained by dividing the negative of the carbon-neutral label's coefficient by the price coefficient. Based on the RPL model (where I randomize both the carbon-neutral label and price parameters), the MWTP is obtained through simulation.

Secondary Outcomes

Secondary Outcomes (end points)
The secondary outcomes are the MWTP estimates for organic and ethical trade labels from each subsample.
Secondary Outcomes (explanation)
The MWTP for organic and ethical trade labels are obtained following the same approach as the MWTP for the carbon neutral label.

Experimental Design

Experimental Design
The respondents are asked to consider 80 teabags of tea (or equivalently 200 grams). The main survey includes 1,200 tea drinkers, with 150 participants for pre-testing. The discrete choice experiment (DCE) includes the following choice attributes: carbon-neutral, organic, and ethical trade labels, and price. A split-sample approach is used to understand the effect of more transparent compared to standard carbon-neutral label on consumers preferences. The survey is hypothetical and the participants are not required to pay for their choices.
Experimental Design Details
This study aims to understand consumer preferences and their WTP for carbon-neutral labels, focusing on the distinctions between transparent and standard labels. The main research questions include: (i) Are consumers willing to pay for carbon-neutral labels? (ii) Do consumers value transparent carbon-neutral labels more than standard ones, and if so, by how much? (iii) How do consumers' preferences for CO₂ reductions differ from CO₂ offsets? To gain insights into how much consumers can internalize the climate externality through carbon-neutral labels, the study further explores the following question: (iv) How much are consumers willing to pay to offset or reduce one tonne of CO₂ through carbon-neutral labels, and how does this compare to the social cost of carbon estimates?
The survey begins with an informed consent to participate in the survey and screening questions about age, tea consumption and purchasing habits. Participants under the age of 18, or those who never consume or purchase tea, are screened out. The DCE focuses on 80 teabags or a 200-gram box of tea or with the following attributes: carbon-neutral label, organic label, ethical trade label, and price. The sustainability labels are designed for the purpose of the survey and can take two levels, either the tea product has the label or does not have it. The price levels for the pre-test take seven levels ranging from £0.90 to £6.90 with £1 increments. However, based on participants' feedback in the pre-test, the price levels for the main survey will be adjusted if they are considered too high or too low.
I use a split-sample approach with three different samples. All samples receive identical surveys and choice experiment designs. The only difference is the type of the carbon-neutral label. The first sample is shown a standard carbon-neutral label, stating ‘CO₂ neutral.’ Samples 2 and 3 are shown a transparent carbon-neutral label with additional text stating the percentage of GHG offsetting and GHG reduction. Sample 2 is shown a carbon-neutral label indicating a 50% emission offset and a 50% emission reduction, while Sample 3 sees a carbon-neutral label indicating a 100% emission offset and no (0%) emission reduction.
Ngene software is used for generating DCE design, consisting of 16 choice tasks, with various combinations of attribute levels. There are two blocks, which means that each participant sees 8 choice tasks. There are three choice alternatives in each choice task: two tea product alternatives and “none of the two'' alternative. For the pre-test, a D-efficient design with a Multinomial Logit (MNL) model is used, and the parameter distributions from the pre-test are used for the final design using a Random Parameters Logit (RPL) model.
Randomization Method
The randomization will be done at the individual level by the survey company, stratifying by age, gender, and education.
Randomization Unit
Randomization unit is the “individual.”
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
150 individuals for the pre-test, 1,200 individuals for the main survey
Sample size: planned number of observations
150 individuals for the pre-test, 1,200 individuals for the main survey
Sample size (or number of clusters) by treatment arms
For the pretesting: 50 individuals for each of the three sub-samples; for the main survey: 400 individuals for each of the three sub-samples.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
I find that the MDEs are 0.18 pounds for the difference between samples 3 and 1, and between samples 3 and 2. Furthermore, the MDEs for the difference between samples 1 and 2 is 0.36 pounds. All differences correspond to 18% of the standard deviation. Please refer to Appendix B of the attached proposal in the section "Supporting Documents & Materials" for the details of the power calculation.
Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
The Ethics Committee of University of St.Gallen
IRB Approval Date
2024-03-01
IRB Approval Number
Exempt (N/A)

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials