Back to History Current Version

Improving agricultural productivity and resilience with cellphone imagery to scale climate-smart crop insurance

Last registered on April 06, 2021


Trial Information

General Information

Improving agricultural productivity and resilience with cellphone imagery to scale climate-smart crop insurance
Initial registration date
April 05, 2021

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
April 06, 2021, 6:21 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated
April 06, 2021, 8:13 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.


Primary Investigator

International Food Policy Research Institute

Other Primary Investigator(s)

PI Affiliation
Kenyan Agricultural and Livestock Research Institute
PI Affiliation
ACRE Africa
PI Affiliation
Wageningen University

Additional Trial Information

On going
Start date
End date
Secondary IDs
Weather hazards such as erratic rainfall cause significant hardship for smallholder farmers in Kenya. Climate change is expected to further exacerbate farmers’ vulnerability to extreme weather. The anticipation of possible losses discourages farmers from making productivity-enhancing investments, trapping them in low-risk yet low-return agriculture. Agricultural insurance complemented with other risk-reducing practices is often cited as a sustainable approach to unlock investments in agriculture for smallholders, improving their resilience and productivity. Thus far, the number of successful insurance schemes targeting smallholders is limited, however, because of high monitoring and verification costs of traditional insurance, combined with low demand for index-based insurance—designed to eliminate the need to verify losses—mainly due to poor trust and basis risks (i.e. the imperfect correlation between farmers’ actual losses and insurance payouts). This project aims to overcome these problems through climate-smart picture-based insurance (PBI), which uses cellphone imagery to verify losses, observe management practices and promote the adoption of productivity-enhancing yet resilient technologies through bundling with stress-tolerant seeds. Ground pictures help reduce monitoring costs, minimize basis risks and create synergies with climate-smart resilience technologies. By taking pictures of insured crops, farmers engage directly in the insurance process, improving trust and tangibility. We measure the impacts of this approach by means of a cluster randomized trial in 7 counties from different regions in Kenya. We randomly assign local village entrepreneurs who provide farmers in their communities with insurance and seeds into one of three treatment arms: a) no insurance marketing efforts (control), b) marketing of the climate-smart PBI product (treatment), and c) marketing of a comparable index-based product that does not rely on smartphone pictures for claims settlement (placebo). We cross-randomize whether the village entrepreneurs provide only regular seeds, or also stress-tolerant cultivars that offer partial protection from extreme weather events.
External Link(s)

Registration Citation

Cecchi, Francesco et al. 2021. "Improving agricultural productivity and resilience with cellphone imagery to scale climate-smart crop insurance." AEA RCT Registry. April 06.
Sponsors & Partners

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

Request Information
Experimental Details


The project tests two interventions to improve agricultural risk management:

1. A new picture-based crop insurance (PBI) product that settles insurance claims based on visible damage in smartphone images of a farmer's crops, taken by so-called champion farmers from sowing to harvest, in order to reduce basis risk in crop insurance relative to a remotely measured weather index (provided free of charge to project farmers in the first four seasons of the project).

2. Marketing and other promotions by champion farmers of stress-tolerant varieties for targeted crops (maize, sorghum and green grams). Farmers in both treatment and control received trial packs of these varieties in the first and third season to raise awareness. The champion farmers are aggregating demand for these varieties in the treatment arm during the third and fourth season of the project.
Intervention Start Date
Intervention End Date

Primary Outcomes

Primary Outcomes (end points)
Insurance coverage; adoption of targeted stress-tolerant varieties; investments in other productive inputs and technologies; and labor use (both family and hired labor).
Primary Outcomes (explanation)
Investments in other productive inputs and technologies will include fertilizer, pesticides and other chemicals that are being recorded in our questionnaires.

Secondary Outcomes

Secondary Outcomes (end points)
Agricultural productivity; income per acre from target crops; seed and insurance demand; food consumption (diversity); resilience indicator based on repeat measures of food consumption; and women's empowerment in agriculture.
Secondary Outcomes (explanation)
To measure food consumption, we will construct food consumption scores and household-level dietary diversity scores. For the resilience indicator, we will use a method developed by Upton et al for food consumption scores (see the article "Caveat utilitor: A comparative assessment of resilience measurement approaches"). For measuring women's empowerment in agriculture, we will use the Pro-WEAI indicator developed by IFPRI (

Experimental Design

Experimental Design
Champions are randomly assigned to one of the following treatment arms:
1. Control: champion sends in images of crops for participating farmers but does not provide insurance
2. Weather index-based insurance: champion sends in images of crops for participating farmers and provides these farmers with weather index-based insurance for the crops that are visible in the cellphone images.
3. Picture-based insurance: champion sends in images of crops for participating farmers and provides these farmers with picture-based insurance that settles claims based on visible damage to the crops shown in the cellphone images.

We cross-randomize the type of seeds that champions provide: either regular seeds only, or also stress-tolerant seeds (in addition to any insurance products that they may offer). Specifically, within each treatment arm, 50% of champions are randomly assigned to provide regular seeds, with the remaining 50% also providing stress-tolerant varieties.
Experimental Design Details
Randomization Method
Randomization done in office by a computer (in STATA)
Randomization Unit
We randomize at the level of the champion farmer (village entrepreneurs who are hired by ACRE Africa to provide services such as insurance products and seeds to farmers in their communities).
Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters
We are planning to include 191 champion farmers.
Sample size: planned number of observations
We are planning to include about 15 farmers per champion farmer/cluster (total sample size of 2,880 farmers) in the survey (including the champion farmer him/herself).
Sample size (or number of clusters) by treatment arms
We have the following numbers of champion farmers by treatment:
- 80 champions control (36 providing regular seeds only, 44 also providing stress-tolerant varieties of seeds);
- 40 champions with WBI (29 providing regular seeds only, 11 also providing stress-tolerant varieties of seeds);
- 71 champions with picture-based insurance (35 providing regular seeds only, 36 also providing stress-tolerant varieties of seeds).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
We are targeting a minimum detectable effect size of 0.3, which is a medium effect size. Assuming 2,880 households from 190 champions, an attrition/non-response rate of 13.3%, an intra-cluster correlation of 0.24, a type-I error of 0.05, two-way hypothesis testing, and a power equal to 0.8 (type-II error of 0.2), we obtain the following minimum detectable effect sizes (MDES) in different types of comparisons across treatment arms: Comparison of treatment arms # clusters # obs MDES k1/k2 n1/n2 No insurance vs. PBI 80/71 1,040/923 0.2496 No insurance vs. WBI 80/40 1,040/520 0.2964 PBI vs. WBI 71/40 923/520 0.3026 No insurance vs. Insurance (any) 80/111 1,040/1,183 0.2245 No STV vs. STV Sales 100/91 1,300/1,183 0.2217 STV only vs. PBI+STV 44/36 572/468 0.3440

Institutional Review Boards (IRBs)

IRB Name
Maseno University Ethics Review Committee
IRB Approval Date
IRB Approval Number
IRB Name
International Food Policy Research Institute
IRB Approval Date
IRB Approval Number


Post Trial Information

Study Withdrawal

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

Request Information


Is the intervention completed?
Data Collection Complete
Data Publication

Data Publication

Is public data available?

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials