Participation of pesticide retailers in the agroecological transition

Last registered on February 16, 2024
View Trial History

Pre-Trial

Trial Information

General Information

Title
Participation of pesticide retailers in the agroecological transition
RCT ID
AEARCTR-0012969
Initial registration date
February 13, 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
February 16, 2024, 3:37 PM EST

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

Locations

Country
France
Region

Primary Investigator

Name
Nahomy Gallo Aguilera
Affiliation
Grenoble Applied Economics Laboratory
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Adelaide Fadhuile
PI Affiliation
Grenoble Applied Economics Laboratory
Contact Investigator
PI Name
Julie Subervie
PI Affiliation
INRAE
Contact Investigator
PI Name
Maud Blanck
PI Affiliation
INRAE
Contact Investigator

Additional Trial Information

Status
On going
Start date
2023-09-01
End date
2024-03-31
Keywords
Agriculture, Behavior, Environment & Energy
Additional Keywords
Public Policies, Nudge, Agroecological Transition, Pesticides
JEL code(s)
Q13, Q18, C93
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
In France, as in many other countries, sellers of phytosanitary products occupy a strategic place in the agrifood sector since they often play an advisory role to their buyers, in addition to their activity of selling inputs. For several years, the public authorities have put in place a system of points (so-called certificates) offered to retailers who make the effort to reorient their activity towards the sale of more environmentally friendly products and the promotion of actions in favor of the environment, such as putting farmers in contact with providers of alternatives to pesticides. Unfortunately, retailers are struggling to collect a sufficient number of certificates to bring about a significant change in farmers' practices. Certain green actions are very rarely carried out, although they yield a lot of certificates, are inexpensive to implement and have a proven positive effect on the environment. This study aims to change retailers' aversion for a key action of the scheme, namely collecting a new minor crop from farmers. In a randomized controlled trial (RCT), we nudge retailers, with the aim to improve their understanding of the certificates system and to remind them of the environmental gains associated with the introduction of a new minor crop into the landscape. Half of the resellers who volunteer to participate receive the nudge, while the other half do not. Next, both groups participate in a discrete choice experiment (DCE) to provide estimates of their preferences for a number of green actions (including collecting a minor new crop). The experiment aims to evaluate the impact of the nudge on responses to the DCE, that is to say, the impact on the retailer's willingness to collect a new minor crop of great importance for the environment.
External Link(s)

Registration Citation

Citation
Subervie, Julie et al. 2024. "Participation of pesticide retailers in the agroecological transition ." AEA RCT Registry. February 16. https://doi.org/10.1257/rct.12969-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
The intervention takes the form of a webinar during which an oral presentation of a few minutes with an illustrated slideshow is made to the treated group. The presentation deals with the benefits of the introduction of a new minor crop into the agricultural landscape. It includes a variety of nudges. Retailer preferences for green actions (including the promotion of minor crops) are measured immediately after the intervention, using a discrete choice experiment (DCE).
Intervention Start Date
2024-02-19
Intervention End Date
2024-02-20

Primary Outcomes

Primary Outcomes (end points)
The main outcome is retailer willingness-to-accept (WTA) for implementing the requested action, namely the collection of a new minor crop from the farmers. Willingness-to-accept is derived from the estimation of a mixed logit model. The intervention is expected to decrease the cost that retailers believe they have to bear when they engage in the collection a new minor crop from the farmers.
Primary Outcomes (explanation)
The main outcome is retailer willingness-to-accept (WTA) for implementing the requested action, namely the collection of a new minor crop from the farmers. Willingness-to-accept is derived from the estimation of a mixed logit model. The model is estimated using data collected from retailers during a discrete choice experiment (DCE), where they are offered to choose between scenarios that include both actions (including collecting a new minor crop) and rewards. In our framework, the estimated WTA for collecting a new minor crop is actually a measure of the opportunity cost of doing it. Since all or part of this cost is in reality due to a lack of understanding of the costs and benefits of the requested action, the intervention is expected to reduce, or even cancel, this cost. In practice, the intervention is expected to significantly decrease the estimate of retailer WTA as estimated through the mixed logit model.

Secondary Outcomes

Secondary Outcomes (end points)
The intervention is also expected to decrease the cost that retailers believe they have to bear when they engage in other green actions included in the DCE.
Secondary Outcomes (explanation)
The intervention is likely to remove obstacles common to all of the actions requested in the certificates system. Typically, participants could believe (rightly) that all actions, and not just the action relating to the collection of a new crop, can generate an environmental gain without incurring high costs for their implementation. In this case, the intervention would have the effect of modifying all other coefficients (all other WTA) of the mixed logit model.

Experimental Design

Experimental Design
The RCT includes one treatment group and one control group. Both groups participate in the DCE, which allows the estimation of the WTA. The treatment group receives the intervention (they attend the nudges webinar) before participating in the DCE, while the control group receives the intervention after playing the DCE.
Experimental Design Details
The sample includes about 130 retailers volunteering to participate in a research program that deals with the reduction of the use of pesticides. The participants come from two French retailer networks: Coopération Agricole and Fédération du Négoce Agricole. Each volunteer participant is asked to sign an FPIC form specifying that she understands that her participation in the experiment will not change her obligations with regard to the current certificates scheme. The protocol includes two stages, organized into webinars.
During the first webinar, all retailers are invited to participate in a DCE, which serves as a baseline for the analysis and the results of which are presented during a second webinar, organized a few weeks later. After the first webinar, participants are randomly assigned to the two treatment arms. This is a stratified randomization according to two dimensions: the network from which the participant comes (Coopération Agricole or Fédération du Négoce Agricole) and the size of the company she represents (large or not). Stratified randomization is performed using computer software.
During the second webinar, both groups are presented with the main results of the DCE run during the baseline session. Next, participants are assigned to two separated virtual rooms, according to the random assignment. Treated group attends the nudges webinar and are invited to play the DCE immediately afterward. Control group play the DCE first.
Randomization Method
This is a stratified randomization according to two dimensions: the network from which the participant comes (Coopération Agricole or Fédération du Négoce Agricole) and the number of certificates that retailers must collect under the mandatory current scheme (large or not). Stratified randomization is performed using computer software.
Randomization Unit
Retailers (individuals).
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
No clusters.
Sample size: planned number of observations
Total number of retailers is 130.
Sample size (or number of clusters) by treatment arms
65 retailers in the treated goup and 65 retailers in the control group.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
With a sample of 130 subjects, 65 per group, we will detect a difference of 0.12 between the control-group mean and the treated-group mean, given a standard deviation of 0.25, with 90% power using a 5%-level two-sided two-sample means t test
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan

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