Information Asymmetries in Crop Insurance: Evidence from the Philippines
Last registered on June 21, 2018


Trial Information
General Information
Information Asymmetries in Crop Insurance: Evidence from the Philippines
Initial registration date
June 21, 2018
Last updated
June 21, 2018 10:59 AM EDT
Primary Investigator
University of Maryland, College Park
Other Primary Investigator(s)
Additional Trial Information
Start date
End date
Secondary IDs
Asymmetric information imposes costs on a wide range of markets and may explain why some important markets, such as most agricultural insurance markets, have failed to develop. This trial uses a two-level randomization and incentivized preference elicitation to study asymmetric information in crop insurance in the Philippines. A total of 569 farmers participated in the experiments, which were conducted in 2010-12. Farmers were asked which plot in their portfolio they preferred to be insured and insurance was randomly allocated to farmers and plots in a way that provided for incentive compatibility in the plot choice decision and induced within- and across-farm variability in insurance coverage. The data generated by the choice and randomized experiments were combined with detailed farm- and plot-level survey data, geospatial data on plot and residence location and administrative data from the insurance provider.

In this first RCT of moral hazard in non-health insurance I find evidence of moral hazard in preventing damage from pests and crop diseases and this effect is particularly strong among farmers who report high trust in the insurance company: for this group harvest losses from this cause double on insured plots. Farmers prefer insurance on plots that are at risk of floods and crop diseases, a sign of classic adverse selection, and plots that are far away from home, a sign of selection on anticipated moral hazard behavior, resulting in 72% higher payouts on preferred plots.
External Link(s)
Registration Citation
Gunnsteinsson, Snaebjorn. 2018. "Information Asymmetries in Crop Insurance: Evidence from the Philippines." AEA RCT Registry. June 21.
Former Citation
Gunnsteinsson, Snaebjorn. 2018. "Information Asymmetries in Crop Insurance: Evidence from the Philippines." AEA RCT Registry. June 21.
Experimental Details
Crop insurance from the Philippines Crop Insurance Corporation.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Harvest losses due to (1) typhoons or floods, or (2) pests and crop diseases. Insurance payouts.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Harvest, fertilizer use.
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The randomization was done in two levels. First, farmers were randomized into treatment groups. Second, plots of farmers in the treatment groups were randomized to be covered by insurance or not. Before randomization farmers were asked to rank their plots in order of preference for insurance and the randomization was done in a way to provide incentive compatibility in the farmers' first choice.
Experimental Design Details
Randomization Method
By computer.
Randomization Unit
Level 1: Farmer/farm; Level 2: Plot
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
Total of 569 farmers were ever registered in the trial. Some farmers participated in multiple rounds of the experiment and we had a total of 838 farmer-season observations in the full scale experiments (excluding the pilot).
Sample size: planned number of observations
We had a total of 2,399 plots in the full scale experiments.
Sample size (or number of clusters) by treatment arms
About 2/3 of farmers were in treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
Yale University IRB
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Is data collection complete?
Data Collection Completion Date
Final Sample Size: Number of Clusters (Unit of Randomization)
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
Data Publication
Data Publication
Is public data available?
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers