Judging a book by its cover: an experiment about information asymmetry and trustworthiness in the Nigerian seed market

How to cite

Steeg, Emily ter; Aju, Stellamaris; Bulte, Erwin; Grimm, Michael; Berg, Marrit van den. 2024. "Judging a book by its cover: an experiment about information asymmetry and trustworthiness in the Nigerian seed market." AEA RCT Registry. AEARCTR-0014532. https://doi.org/10.1257/rct.14532-1.0

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

Emily ter Steeg* , Stellamaris Aju , Erwin Bulte , Michael Grimm , Marrit van den Berg

* Primary investigator

Registry ID
AEARCTR-0014532
DOI
https://doi.org/10.1257/rct.14532-1.0
First registered
Last updated
Status
On going
Trial dates
2024-06-26 – 2025-05-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/14532

Abstract

Smallholder farmers often underinvest in agricultural inputs, such as quality seed of improved varieties. One reason for this underinvestment is that farmers perceive the quality of agricultural inputs sold by agrodealers as inconsistent, leaving them exposed to the risk of purchasing ‘bad seed’. In the maize seed market, two types of packaging are common: unbranded (see-through) bags and branded bags. Farmers are aware that seed in unbranded bags is repacked by the agrodealer. In the Nigerian tomato seed market, we find three types of packaging: branded cans, branded pouches and unbranded pouches. Farmers strongly prefer the cans because they consider these to be tamper-proof. Thus, the different types of seed packaging represent varying levels of information asymmetry for farmers. Information asymmetry can lead to a lack of trust among farmers, causing them to avoid buying seed altogether, or to only purchase seed sold in cans or pouches from well-known brands. Alternatively, farmers may rely on signals from agrodealers and packaging to assess trustworthiness and quality as a way to cope with information asymmetry. This study has three objectives. First, we identify characteristics that influence farmers' perceptions of agrodealers' trustworthiness and the quality of their products. Second, we examine how the presence or absence of these signals impacts farmers' Willingness to Pay (WTP) for seed. We compare WTP for two types of maize seed (unbranded bags versus branded bags) and two types of tomato seed (branded cans versus branded pouches), exploring whether these signals play a smaller role in markets with less pronounced information asymmetry, and a larger role where asymmetry is more pronounced. Finally, we investigate whether the signals observed by farmers have predictive value regarding the quality of the seed sold by agrodealers.

Experimental design

At the core of our research design are two series of Becker-DeGroot-Marschak (BDM) auctions with farmers. BDM auctions are incentive compatible as they confront participants with real purchase decisions. With these auctions, we measure the impact of signals of trustworthiness sent by agrodealer appearance and packaging on farmers’ WTP for seed, and how the value of signalling is mediated by the extent of asymmetric information in the relevant market.

Design details

Randomization method
Randomisation is done by a computer.
Randomization unit
During the BDM auctions, for each farmer, one agrodealer is randomly selected from each the three 'trustworthiness' bins.
Clusters
There are no clusters.
Observations
With 500 farmers in our samples (stage 3 and 4) each making 12 bids, we have 6000 total observations for the mixed crop models, and 3000 observations for each of the crop-specific models.
Arms
Farmer sample 1: 116 Farmer sample 2: 289 Farmer sample 3: 500 Farmer sample 4: 500 Agrodealer sample 1: 78 Agrodealer sample 2: 45
Primary outcomes
We contribute to this body of literature by studying how informal quality signals regarding agrodealers and seed packaging impact Willingness to Pay (WTP) among Nigerian smallholder farmers for maize and tomato seed. Our key outcome variables are 1) general WTP and 2) crop-specific WTP for seed.

Cited by

No citation data available for this registration yet.

Keywords