Measuring Spillover Impact of a Market Access Program of Avocados in Kenya

How to cite

Naso, Pedro; Sarr, Mare; Mensah, Edouard. 2023. "Measuring Spillover Impact of a Market Access Program of Avocados in Kenya." AEA RCT Registry. AEARCTR-0011741. https://doi.org/10.1257/rct.11741-1.0

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

Pedro Naso* , Mare Sarr , Edouard Mensah

* Primary investigator

Registry ID
AEARCTR-0011741
DOI
https://doi.org/10.1257/rct.11741-1.0
First registered
Last updated
Status
In development
Trial dates
2023-01-04 – 2023-12-08
Countries
Kenya (Mount Kenya Region)
JEL codes
Q10 Q12 O13 O12
Registry record
http://www.socialscienceregistry.org/trials/11741

Abstract

This project studies the economic impact of a certification program to avocado farmers in the Mount Kenya region. A large non-profit organization in the area allows farmers to get their Hass avocado crops certified ahead of the avocado season. Upon receiving certification, farmers enter into an agreement with the NGO, which ensures that the NGO will purchase their avocados at a higher price. The NGO then exports these Hass avocados to Europe. We partnered with this NGO to estimate the direct and indicret impact of this certification program. We are interested in estimating the differences in prices, revenue and profits between (i) farmers with and without certification, and (ii) uncertified farmers in areas with certification compared to those in areas without certification. We propose that the certification program could be boosting the earnings of all avocado growers in the Mount Kenya region - not solely those with certification. Our identification strategy relies on the fact that the latest expansion of the certification program was carried out randomly at the subcounty level.

Intervention

A prominent NGO in the Mount Kenya region provides a certification program to farmers growing Hass avocados. If a farmer gets certified, which requires some basic modifications in farm organization and hygiene practices, they can sell their Hass avocado yield to the NGO at a higher price. This certification initiative began in 2022 in 9 sub-counties around Mount Kenya, and the NGO is gradually broadening the locations offering certification. The most recent expansion of the program was conducted randomly at the sub-county level, which allows us to determine the causal impact of the certification program. We evaluate two types of effects, direct and indirect, by comparing:
(i) Certified and non-certified farmers (Direct Impact);
(ii) Non-certified farmers in areas with and without certification (Indirect Impact).
Our research focuses on three variables: the price farmers charge for avocados, avocado revenue, and avocado profits. We propose that the certification program may increase avocado prices in the region, both in sub-counties where certification is available and in those where it is not. Furthermore, we believe that all avocado varieties (Hass, Fuerte and local varieties) in the region might see a price rise.

Experimental design

In partnership with a large NGO that operates in the Mount Kenya region, we randomize the expansion of a certification program for Hass avocado farmers at the sub-county level. In treated sub-counties, farmers growing Hass avocados can join the certification program, which calls for some elementary changes in farm organization and cleanliness measures. After obtaining certification, these farmers can sell their Hass avocados to the NGO at a price above the regular market rate. We compare these treated farmers with three types of farmers:
(i) Uncertified farmers based in sub-counties where there is certification (control group 1). These are also non-elegible for certification;
(ii) Uncertified farmers who are elegible for certification but are based in sub-counties where the NGO does not offer certification (control group 2);
(iii) Uncertified farmers who are not elegible for certification and are based in sub-counties where the NGO does not offer certification (control group 3).
We then estimate direct impact comparing certified farmers with control group 2; and spillover impact comparing control groups 1 and 3. We focus on three variables of interest: average avocado prices, avocado revenue, and avocado profit. These are measured for the most important varieties of avocado in the region: Hass, Fuerte and local varieties.

Design details

Randomization method
Randomization was done in office by a computer.
Randomization unit
Cluster: sub-counties in the Mount Kenya Region
Clusters
We will have 32 clusters: 16 treated (where there is the possibility of certification) and 16 control (where there is no possibility of certification).
Observations
45 farmers per cluster, which is equivalent to 1,440
Arms
16 treatment clusters and 16 control clusters
Primary outcomes
Avocado prices, avocado revenue (prices times quantity), and avocado profits (revenue minus costs of production). We will estimate these outcome variables for all farmers we survey and for all varieties of avocado (Hass, Fuerte and local varieties).
Secondary outcomes
We are also interested in brokers activity in the region. So, we will also work with the variables: prices offered by brokers, frequency of brokers visit, and number of different brokers.

Cited by

No citation data available for this registration yet.

Keywords