Reducing pesticides in Bordeaux vineyards

How to cite

Raineau, Yann; Giraud-Héraud, Eric. 2020. "Reducing pesticides in Bordeaux vineyards." AEA RCT Registry. AEARCTR-0006830. https://doi.org/10.1257/rct.6830-1.0

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

Yann Raineau* , Eric Giraud-Héraud

* Primary investigator

Registry ID
AEARCTR-0006830
DOI
https://doi.org/10.1257/rct.6830-1.0
First registered
Last updated
Status
Completed
Trial dates
2016-01-01 – 2019-12-31
Countries
France (Nouvelle-Aquitaine)
JEL codes
Q12, Q52
Registry record
http://www.socialscienceregistry.org/trials/6830

Abstract

We study the comparative effect of two types of social comparison nudges, mobilized in an empirical framework of support to winegrowers in limiting the overdose of phytosanitary products for the protection of their vines. Three groups of winegrowers from the same Bordeaux region were randomly constituted: two action groups each receiving one type of nudge, and a control group. Their phytosanitary practices are monitored for several years, making it possible to measure the effect of these interventions, but also to analyze the real basis of their functioning.

Intervention

Within the framework of this partnership project carried out with a wine cooperative, postal mails were co-designed with the technical service of the cooperative and then sent by this service (in order to ensure the credibility of the information received). 165 letters were sent on April 8, 2015, 82 containing the social comparison nudge modality A (complete information on the distribution of the group), and 83 containing the social comparison nudge modality B (information limited to individual positioning and the group average). No further mail is sent afterwards.

Experimental design

From 2015 to 2019, we collect data concerning the phytosanitary practices of a group of winegrowers belonging to the same cooperative in Bordeaux region, randomly divided into three subgroups A, B and C. Before chemical protection starts at spring 2016, each individual in groups A and B receives a letter informing him/her of his/her intensity of use of phytosanitary products for the previous year 2015, expressed through a new synthetic indicator, compared to the average intensity of the whole group A+B+C. The letters addressed to group B also present the complete distribution of all the different levels of intensity. No further action is taken afterwards. The results are obtained at the end of each crop year, with a follow-up from 2016 to 2019.

Design details

Randomization method
To conduct randomization, individuals' assignment within one of the three groups was conducted according to a stratification scheme by the observed variable, the chemical-use intensity indicator (TFI). We adopted the principle of pairwise matching, presented by Duflo, Glennerster, and Kremer (2007). In our case, because we wanted three groups instead of two (two action groups and a control group), we proceeded to an assortment by trios and not by pairs. Thus, once ordered by the value of their TFI in 2015, trios were formed and, within these trios, each winemaker was randomly assigned to one of groups 1, 2 or 3.
Randomization unit
Individual.
Clusters
247 individuals.
Observations
247.
Arms
82, 82 and 83.
Primary outcomes
Three groups of winemakers were formed in this study: two groups each receiving a type A or B social comparison nudge that compared (in a different way from A to B) the practices of the targeted individual with those of the rest of the group, and a third control group. At the end of the experiment, we find that only one of the two nudges manages to change the group practices: group B (limited information). This leads us to a better understanding, by comparing the formulation of the two nudges, of the necessary element for their efficiency. This element consists of an incomplete level of information, leaving room for interpretation bias. We also note the disappearance of the nudge effect after a few years (monitoring of practices until 2019).

Cited by

No citation data available for this registration yet.

Keywords