Willingness to pay for index-based weather insurance by rice farmers in Vietnam

While the number of weather insurance programs for farm households has increased in low- and middle-income countries, demand for insurance remains at low levels globally. Demand for agricultural insurance is also low among rice farmers in Vietnam, who are increasingly exposed to extreme weather events, including heavy rainfall. Among the possible reasons for the low uptake of weather insurance in Vietnam and elsewhere is the price of insurance premiums.

The project outlined here aims to investigate the willingness to pay (WTP) and realized purchases of farming households in Vietnam for parametric insurance against heavy rainfall offered by Igloo Insurance. The insurance product is tailored to rice farmers, who can purchase a 30-day insurance coverage for a premium of 55,000 VND (around 2,25 USD, including VAT) per hectare of land (at the time of the study). The insurance product provides coverage against the risk of excessive rainfall. If a heavy rainfall event occurs (i.e., when the amount of rainfall in the farmer’s district exceeds one of three pre-defined thresholds over a five-day period), insured farmers receive a payout within 5 days of the event.

We employ the Becker-DeGroot-Marschak (BDM) method to elicit the WTP among participants of the first wave of the Vietnam Rice Farmer Panel Survey, implemented by RWI – Leibniz Institute for Economic Research and the Mekong Development Research Institute (MDRI) between January and May 2024. The sample of the survey encompasses ca. 3,000 rice-farming households. Survey interviews were carried out in person, usually at the households’ premises, by MDRI enumerators.

All survey respondents received information on Igloo’s parametric rainfall insurance in the form of a 5-minute video clip and a leaflet. MDRI enumerators presented the video to the survey respondents on a tablet. The video contained information about the insurance product (the index, triggers, payout conditions, and how to buy insurance) as well as the WTP module of the Vietnam Rice Farmer Panel Survey. The leaflet contained information about Igloo and the insurance product (index, triggers, payout conditions, and how to buy insurance).

Two randomized interventions will be implemented as part of this WTP study, which allow us to examine the impact of price information and premium discounts on the WTP for weather insurance among rice farmers in Vietnam.

Information treatment on real-world pricing: To study the effect of knowledge of real-world pricing on farmers’ WTP, half of the survey respondents are provided with information on the market price of Igloo’s parametric rainfall insurance before eliciting their WTP. The following sentence is read out by the survey enumerator before the first question on the WTP is asked: “Before we start, we would like to inform you that the market price of this insurance is 55,000 VND for covering one hectare for one month.”

Randomized premium discount voucher treatment: To study the effect of pricing on realized insurance purchases, survey respondents are offered vouchers to purchase Igloo’s parametric rainfall insurance at a discounted price, with the value of the discount being randomized. The discounted prices range from 2,500 to 40,000 VND per month and hectare and are determined by a dice roll by the survey respondents.

The voucher can be used to purchase Igloo’s parametric rainfall insurance at a discounted price. Recipients of the voucher can decide the length of the period they want to purchase insurance for, choosing between 30 days, 60 days, 90 days, 120 days, and 150 days. Recipients can use this voucher to purchase an insurance policy for up to five hectares of rice area.

Recipients can redeem their voucher to purchase Igloo’s parametric rainfall insurance by calling a hotline. In addition, survey respondents who receive a voucher are contacted by Igloo’s insurance agents after the survey via phone calls to facilitate the redeeming of the voucher.

Data on farmers’ realized purchase behavior is planned to be sourced from Igloo’s customer database in September 2024, with further updates planned for December 2024.

2024-01-08

2024-05-13

Stated value of willingness to pay for Igloo’s parametric rainfall insurance.
Realized purchases of Igloo’s parametric rainfall insurance.

The WTP study builds on the Vietnam Rice Farmer Panel Survey, which is implemented in four provinces of Vietnam’s Mekong Delta (Long An, Dong Thap, An Giang, and Can Tho). The survey is conducted among about 3,000 rice-farming households in 200 villages. Data from the first wave was collected between January 15 and May 13, 2024. The BDM WTP module of the Vietnam Rice Farmer Panel Survey is implemented among all households that confirmed they understand Igloo’s parametric rainfall insurance product and the rules of the WTP exercise after watching an explanatory video and receiving additional explanation from the enumerator (if necessary).

Information treatment: The assignment of whether a survey respondent receives information on the real-world pricing of Igloo’s parametric rainfall insurance product is determined by a random number between 0 and 1 generated by the survey software during the survey interview. Households with a random number less than (or equal to) 0.5 receive the information.

Randomized premium discount voucher treatment: The random offer of premium discount vouchers is implemented among all survey respondents who stated a positive WTP for one hectare-month coverage of Igloo’s parametric rainfall insurance. All respondents who stated a WTP greater than zero were asked by the enumerator to roll two dice. Each possible sum of dice is associated with a price offer between 2,500 VND and 40,000 for covering one hectare for one month with Igloo’s parametric rainfall insurance (see Table 1 in additional materials). Depending on the sum of the dice, the enumerator will offer the respondent a voucher to purchase insurance at a discounted price, provided their stated WTP is equal to or greater than the offer associated with the sum of the dice.

Information treatment: Random number generated by survey software during the survey interview.
Randomized premium discount voucher treatment: Dice rolling during the survey interview.

Survey respondent

No

Ca. 3,000 survey respondents

Ca. 3,000 survey respondents

Information treatment: Ca. 1,500 survey respondents control, ca. 1,500 survey respondents treatment

Randomized Premium Discount Voucher Treatment: Random price offers are made according to the sum of two dice that are rolled by survey respondents who stated a positive willingness to pay. Eleven price offers are possible. The probability of receiving the respective price offers depends on the probability of rolling the respective sum of two dice (see Table 1). Whether a survey respondent receives a premium discount voucher (with the value of that discount determined by the dice roll) depends on whether the respondent’s stated willingness to pay is equal or larger to the random price offer (in that case, the survey respondent receives the voucher) or smaller than the random price offer (the respondent does not receive a voucher).