Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Rural Finance and Commitment Mechanisms in Agricultural Input Decisions
Last registered on January 02, 2019


Trial Information
General Information
Rural Finance and Commitment Mechanisms in Agricultural Input Decisions
Initial registration date
April 07, 2017
Last updated
January 02, 2019 4:24 PM EST
Primary Investigator
Northwestern University
Other Primary Investigator(s)
Additional Trial Information
In development
Start date
End date
Secondary IDs
The proposed experiment investigates the role that credit and access to early commitment mechanisms can have in increasing fertilizer adoption. The context in West Africa - and especially the Sahel - is one with large market inefficiency between agricultural input dealers, agricultural credit providers and farmers. In the seminal Duflo et al. (2012) study of commitment mechanism and fertilizer adoption, early offers of hard commitment mechanism (full payment at commitment) to adopt fertilizer with guaranteed delivery during the planting season were found to induce higher take-up of fertilizer than subsidized or market-priced offers at planting season. Two mechanisms potentially explain these results: time inconsistent preferences by farmers or the effect of credit constraints during the planting season, when farmers may have exhausted agricultural income from the previous year’s harvest. The proposed experiment will disentangle the relative importance of commitment mechanisms and credit in increasing fertilizer adoption in rural Mali by partnering with a national microcredit provider and a national network of agro-input dealers to provide agricultural credit on different commitment terms via agricultural input fairs organized throughout the agricultural season.
External Link(s)
Registration Citation
Dillon, Andrew. 2019. "Rural Finance and Commitment Mechanisms in Agricultural Input Decisions." AEA RCT Registry. January 02. https://doi.org/10.1257/rct.2032-2.0.
Former Citation
Dillon, Andrew. 2019. "Rural Finance and Commitment Mechanisms in Agricultural Input Decisions." AEA RCT Registry. January 02. http://www.socialscienceregistry.org/trials/2032/history/39842.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
A national agro-input dealer network will organize input fairs at different times in the agricultural season. Soro Yiriwaso (SY), a national microfinance institution, will facilitate purchases at input fairs by offering different commitment mechanisms to access agricultural input credit. The commitment will vary between hard and soft by using the deposit amount (down payment) as the distinguishing feature in making the commitment. A variation of the proposed intervention has already been successfully implemented in Burkina Faso in 2014 by Dillon et al. The early commitments will vary between hard and soft by using the deposit amount as the distinguishing feature. For half of the groups, SY will also offer a new financial product called “input credit.”
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
1. Does early commitment induce higher take-up of agricultural inputs relative to late commitment?
2. Do hard or soft early commitment mechanisms induce higher take-up of inputs?
3. Does the provision of a credit scheme at planting season induce higher take-up relative to early commitment mechanisms, providing evidence that credit constraints rather than mental accounting is the binding constraint?
4. Are there differences in treatment response by gender?
Primary Outcomes (explanation)
The main outcome of interest for this research is take-up of agricultural inputs and specifically fertilizer. A village baseline census and follow-up administrative data (collected at time of input delivery) will be the primary survey instruments used to collect demographic, production, and credit information which will be accompanied by administrative data from our implementation partner. Take-up will be computed as the fraction of farmers purchasing agricultural input with respect to the total population. In parallel, outcomes related to investment and production will be tracked, such as input purchases (from the fair and from other vendors), land cultivated, value of inputs, labor and (reported) value of harvest. Data will be collected for male and female farmers. Finally administrative data on credit history, the credit contract, repayments and default rates will be collected by our partner SY which will be linked to the census household survey. Administrative data on purchase orders and delivery of agricultural inputs will be collected by our partner UNRIA, which have a tracking system already tested and available.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The study will test variations of timing of agricultural input fairs combined with variations of the credit contract in the form of the up-front payment required. The commitment will be a down payment that the farmers make at the end of the harvest season, for inputs to be delivered six months later. The products characteristics are defined by whether there is (i) an upfront payment with a soft commitment of 5%, (ii) hard commitment of 50% or (iii) no upfront payment, and whether the payment upon delivery of the product is financed (i) through a loan or (ii) paid directly by the farmer.
In Group #1, SY will facilitate interaction between farmers and input dealers by accompanying the supplier as they organize the input fair. The early input fairs occur just after harvest (January 2018). The opportunity to purchase agricultural inputs is then offered to farmers, with the possibility of financing their purchase with a loan. The farmer would pay a 5% deposit on the day of the fair. The balance of the purchase will be financed by a loan activated upon delivery of the inputs by the dealer at the beginning of the planting season. SY evaluates the applicant for creditworthiness when the farmer purchases the inputs and upon approval, it places the deposit amount in a blocked account. The blocked account will accrue a nominal interest rate and will be fully transferred to the input dealer upon delivery, along with the balance payment financed by credit. If the farmer reneges on the purchase, SY will transfer the farmer’s deposit to the agro-input dealer.
In Group #2, a similar organization will be used, with the exception that the farmer deposit will be 50% of the purchase order. In Group #3, the input fair takes place at the beginning of the planting season, approximately 5 months after the harvest. During the fair, SY offers credit contracts for purchasing fertilizer at market price. The purchase value will become a loan and SY will make a payment directly to the input dealers after execution of the credit contract. The treatments in groups #4, #5 and #6 will occur without offering credit to farmers, as a comparison to the treatments implemented with credit availability.
Experimental Design Details
Randomization Method
The randomization will be done in office by a computer
Randomization Unit
The randomization unit will be at the level of villages.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
In total, there will be 140 villages.
Sample size: planned number of observations
We are planning to observe 9240 households. This has been decided on the base of 60 households per village with 10% of attrition.
Sample size (or number of clusters) by treatment arms
The sample size amounts to 140 villages, 20 in each treatment group (groups #1-#6) and 20 in the control group (group #7). In each village there will be around 60 households observed.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
The study’s power calculations focus on the tradeoffs between study power and the number of clusters, as the study design will take a baseline census of all agricultural households in each cluster. We focus on the primary outcome of interest for the study, adoption and purchase of fertilizer. Given standard assumptions of power set to 0.8 and statistical significance of 0.05, the number of clusters required to achieve this level of power is estimated using Optimal Design to identify differences in mean take-up rates between each of the study’s six interventions . Each comparison between two treatments is adequately powered when the number of clusters is set to 40 villages (20 villages per treatment), with the exception of differences between group #2 and #4 where we anticipate similar take-up rates. Therefore, the study design has set the number of clusters per treatment arm to be 20 villages per treatment with anticipated 60 households per cluster given our budget constraint which adequately powers the mean differences between treatment groups.
Supporting Documents and Materials

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
IRB Name
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)