Accelerating Varietal Turnover: Appealing to Smallholders as Both Producers and Consumers

How to cite

Trachtman, Carly; Abate, Gashaw; Aredo, Samson Dejene; Kannan, Samyuktha; Krahn, Anne; Kramer, Berber; Ma, Ning; Mulwa, Chalmers; Nabwire, Leocardia; Okello, Julius Juma; Patil, Vikram; Ragasa, Catherine; Campenhout, Bjorn Van; Veettil, Prakashan Chellattan; Waweru, Carol. 2023. "Accelerating Varietal Turnover: Appealing to Smallholders as Both Producers and Consumers." AEA RCT Registry. AEARCTR-0012728. https://doi.org/10.1257/rct.12728-2.0

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

Carly Trachtman* , Gashaw Abate , Samson Dejene Aredo , Samyuktha Kannan , Anne Krahn , Berber Kramer , Ning Ma , Chalmers Mulwa , Leocardia Nabwire , Julius Juma Okello , Vikram Patil , Catherine Ragasa , Bjorn Van Campenhout , Prakashan Chellattan Veettil , Carol Waweru

* Primary investigator

Registry ID
AEARCTR-0012728
DOI
https://doi.org/10.1257/rct.12728-2.0
First registered
Last updated
Status
On going
Trial dates
2023-01-01 – 2025-01-31
Countries
Private
JEL codes
O13, O33
Registry record
http://www.socialscienceregistry.org/trials/12728

Abstract

Smallholder producers throughout the developing world continue to produce old crop varieties, despite the availability of newer, likely preferable alternatives. One explanation is that learning about new varieties through experimentation is risky, as newer varieties can be more expensive, and farmers may be unsure of how these varieties will perform on their land. As self-consumption is also prevalent, households may face additional consumption-related risks when producing a new variety, such as dislike of the taste. In this project, we compare interventions that address production-side risks (free seed trial packs) and consumption-side risks (free samples of the same variety's end-product, and a cooking demonstration) to adoption of new varieties. We use a randomized controlled trial with a 2x2 matrix treatment design, where farmers either receive one intervention, both, or neither. We implement this across six settings (Ethiopia, India, Kenya, Nigeria, and two in Uganda) considering a variety of crops.

Intervention

Production Risk Intervention: Farmers are given a free trial pack of seeds of one of the improved varieties being promoted in their setting. Trial packs are distributed at the beginning of the relevant agricultural season, such that farmers can still incorporate cultivating the trial pack as part of their production for the season. The trial pack is only meant to cover a portion of a farmer's agricultural land, but large enough for farmers to learn and experiment with how the variety works.

Consumption Risk Intervention: Farmers will be given a ``free sample'' of the crop itself rather than of the seeds of the crop. Similarly to with the trial pack, the sample is only intended to allow families to learn about the taste and consumption attributes of the new variety without having to necessarily invest in cultivating or purchasing the variety. In practice, the samples in question usually constitute enough to make 1-3 meals for the household, with some variation between settings. In some country sites, tasting experimentation was facilitated in a formal setting with individuals or groups, guided by enumerators. In other cases, households were simply given the free sample encouraged to cook with and taste it, and compare its taste to other varieties they usually eat.

Experimental design

The basic evaluation design to be implemented across all 6 settings (in 5 countries) is a randomized controlled trial (RCT), randomized at the cluster level. Hence identification of treatment effects comes from random assignment. The design, is a 2x2 matrix design, in which clusters are assigned to receive the Production
Risk Intervention, the Consumption Risk Intervention, both, or neither (Control Group).

Design details

Randomization method
Done in office by a computer
Randomization unit
Village clusters
Clusters
Across all sites, there will be a total number of 809 clusters. (Ethiopia: 120, India: 120, Kenya: 157, Nigeria: 172, Uganda 1: 120, Uganda 2: 120)
Observations
Total: 10,636 (Ethiopia: 1920, India: 2040, Kenya: 3140, Nigeria: 1376, Uganda 1: 1200, Uganda 2: 960)
Arms
189 clusters in pure control, 215 getting the production risk treatment only, 192 getting the consumption risk treatment only, and 213 clusters getting both treatments (across all settings)
Primary outcomes
A binary indicator of whether a household reports growing the variety for which they received a trial pack/kit in the subsequent season

Cited by

No citation data available for this registration yet.

Keywords