The Impact of Contract Farming: Experimental Evidence from East Africa

How to cite

Casaburi, Lorenzo; Willis, Jack. 2023. "The Impact of Contract Farming: Experimental Evidence from East Africa." AEA RCT Registry. AEARCTR-0011717. https://doi.org/10.1257/rct.11717-1.0

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

Lorenzo Casaburi* , Jack Willis

* Primary investigator

Registry ID
AEARCTR-0011717
DOI
https://doi.org/10.1257/rct.11717-1.0
First registered
Last updated
Status
On going
Trial dates
2022-12-01 – 2024-07-31
Countries
Rwanda (), Uganda ()
Registry record
http://www.socialscienceregistry.org/trials/11717

Abstract

In the experiment, we study the impact of linking smallholder farmers in value-chain contracting arrangements that provide iron-fortified beans. The partner contract farming companies first identify a set of potential villages to target for expansion into and the randomization then selects a subset of such villages for actual expansion. In a first random sample of villages, farmers receive a crop purchase guarantee and the provision of agricultural inputs (fertilizer and bean seeds) on credit. A second group receives a crop purchase guarantee and the possibility to buy agricultural inputs for cash. A final group of villages is allocated to control.

Intervention

We collaborate with three contract farming companies interested in expanding their business operations to new villages. Within each village involved in the research, the contract farming companies recruit an average of 20 farmers whom they intend to establish contracts with if the village is eventually assigned into treatment. Then, villages are randomized into three groups:
- Control group: farmers in 1/3 of villages will not be contracted.
- Treatment 1: farmers in 1/3 of villages will be offered agricultural inputs on credit and a commitment to purchase their beans at harvest.
- Treatment 2: farmers in 1/3 of villages will be offered agricultural inputs for cash and a commitment to purchase their beans at harvest.

Experimental design

The contract farming partner companies first identify a group of previously untapped villages where they would be interested to expand their business operations. Within each village, they further recruit an average of 20 farmers they would like to contract with if the village is eventually assigned into treatment. Then, villages are randomized into three groups:
- Control group: farmers in 1/3 of villages will not be contracted.
- Treatment 1: farmers in 1/3 of villages will be offered inputs on credit and a commitment to purchase their beans at harvest.
- Treatment 2: farmers in 1/3 of villages will be offered inputs for cash and a commitment to purchase their beans at harvest.

A listing exercise was conducted in September 2022. The baseline survey took place between December 2022 and January 2023. We will do at least two rounds of follow-up surveys (August 2023 and February 2024), more if funding allows.

Design details

Randomization method
Randomization on a computer.
Randomization unit
Randomization at the village-level.
Clusters
67 villages in Uganda and 40 villages in Rwanda.
Observations
Approximately 20 farmers per villages (i.e., 1340 farmers in Uganda and 800 farmers in Rwanda). We may expand the sample in future stages of the research.
Arms
35-36 villages per treatment arm
Primary outcomes
- Contracting outcomes: contract take-up, determinants of take-up - Cultivation choices: amount of land cultivated, cultivation of biofortified beans, other crop choices - Agricultural inputs: use and cost of seeds, fertilizer, herbicides, and pesticides, for bean cultivation and farm-level outcomes - Agricultural labor: indicators of household and hired labor, for bean cultivation and farm-level outcomes - Agricultural production: indicators of quantities harvested, revenues, profits and productivity, for bean cultivation and farm-level outcomes
Secondary outcomes
- Migration patterns - Food security indicators - Household labor and business activities indicators: i. Agricultural labor activities on other people's farms ii. Non-agricultural labor activities - Credit, loans and savings indicators - Household assets indicators - Mental well-being indicators - Media consumption and community engagement indicators

Cited by

No citation data available for this registration yet.

Keywords