In collaboration with the seed companies and Tanzanian extension program, we selected six maize-growing districts in Tanzania which have high drought prevalence. We then match all candidate villages into groups of three (triad) based on agro-ecological characteristics. Using Stata, we randomly select five triads per district. Randomly assign treatment status within triad. Randomly select household from village-level roster.
In the first year (2015-2016), treatment was uniform across both treatment arms: an intensive information and marketing campaign for drought-tolerant maize varieties. In addition to providing small packs (100 grams) of seeds to households in treatment villages, we will provide trial packs (two kilograms) to farmers randomly selected to be study participants. In the second and third years (2016-2018), treatment was split between two arms. In one group of communities, the marketing campaign for drought-tolerant maize varieties was continued, and seeds were made available to a village seller. In the other group of treatment communities, seeds sold were bundled with a satellite based index insurance.
The insurance product premium was 20% of the price of the seed and covered the replacement of the seed in the case of insurance payout. The insurance is indexed on precipitation and area yields and has two triggers. The first trigger is determined by rainfall in the forty days following planting. The second trigger is area maize yields estimated using NDVI.
The insurance was offered by UAP Insurance. Partner seed companies collected lists of farmers purchasing insured seed and provided these farmers with vouchers indicating their use of insured seed. The partner seed companies then remitted the insurance premiums to UAP Insurance. In the event of an insurance payment, the seed company would replace farmers' seed for the following planting season.
It is not necessary for farmers to report drought losses nor for the insurance company to send a claims adjuster to verify losses because the insurance is based on an area level index, not individual outcomes. The indices were monitored by the research team, which provided them to the insurance company. Results of the indices were communicated to farmers through community leaders.