2008年-世界发展银行全球_Patterns_of_Rainfall_Insurance_Participation_in_Rural_India_28页_195kb

Summary of "Patterns of Rainfall Insurance Participation in Rural India"

Core Content

This article analyzes the factors influencing the participation of smallholder farmers in a rainfall insurance product in rural India. The study is based on a household survey conducted in 2004 in Ananthapur and Mahbubnagar districts of Andhra Pradesh. The insurance product, developed by ICICI Lombard with World Bank support, is marketed by BASIX, a microfinance institution, and pays out based on rainfall levels during the three phases of the monsoon season.

Main Viewpoints

• Insurance Participation Trends:

  • Insurance take-up decreases with basis risk (the mismatch between insurance payouts and actual income fluctuations).
  • It increases with household wealth and decreases with binding credit constraints.
  • These patterns align with a neoclassical model that incorporates borrowing constraints.

• Contradictions to Benchmark Model:

  • Risk-averse households are less likely to purchase insurance, contrary to the neoclassical prediction.
  • Familiarity with the insurance vendor (e.g., being a BASIX customer) and participation in village networks significantly influence insurance take-up.
  • These findings suggest that uncertainty about the product and lack of understanding may deter participation, especially among less informed or less trusting households.

• Product Characteristics:

  • The insurance is based on rainfall data from local weather stations.
  • It covers three monsoon phases: sowing, podding/flowering, and harvest.
  • The payout is determined by rainfall thresholds, with a fixed amount paid per millimeter of shortfall, up to a maximum lump-sum payment.
  • The product is relatively inexpensive, costing between 150-250 rupees per acre, and is accessible to farmers with modest incomes.

• Marketing Strategy:

  • BASIX used a combination of opinion leaders and existing customers to promote the insurance.
  • Villages were selected based on the presence of existing customers, crop area, and proximity to rain gauges.
  • Insurance was not sold in all suitable villages due to time constraints and limited marketing capacity.

• Empirical Evidence:

  • Only 4.6% of households purchased the insurance in 2004, reflecting the novelty of the product.
  • The survey included 752 households, with a low nonresponse rate, indicating high survey participation among both purchasers and nonpurchasers.

Key Information

Insurance Design and Pricing

• The 2004 insurance policies were tailored to castor and groundnut, the two main cash crops in the region.
• Premiums were calculated based on historical rainfall data and included expected payouts, standard deviation, administrative costs, and government taxes.
• The payout structure was designed to provide compensation for rainfall below certain thresholds, with a maximum payout per phase.

Example: Narayanpet Mandal

• Premium per acre: Rs 200.
• Phase 1: 60 mm threshold, 10 rupees per mm shortfall.
• Actual rainfall: 12 mm in phase 1, resulting in a payout of Rs 1,500 per acre.
• Phase 2: 100 mm threshold, 15 rupees per mm shortfall.
• Actual rainfall: 84 mm, resulting in a payout of Rs 240 per acre.
• Phase 3: 75 mm threshold, 15 rupees per mm shortfall.
• Actual rainfall: 177 mm, resulting in no payout.

Marketing and Distribution

• BASIX conducted marketing meetings and individual visits to promote the insurance.
• The marketing process involved selecting a subset of villages due to time and resource constraints.
• Factors influencing take-up included the presence of opinion leaders, the number of rainy spells, and the availability of liquid assets.

Theoretical Framework

• The study tests four hypotheses based on economic theory:
  1. Insurance participation is higher with higher risk aversion, lower basis risk, and larger insured risk.
  2. Insurance participation is higher when households expect lower future rainfall.
  3. Insurance participation is higher when households are less credit constrained.
  4. Insurance participation is lower when trust in the vendor is weak and information networks are poor.

Implications

• The findings highlight the importance of trust, familiarity, and social networks in the adoption of new financial products.
• They also suggest that behavioral factors, such as risk perception and cognitive ability, play a significant role in insurance decisions.
• The study underscores the need for better understanding and communication of insurance products in developing markets.

Conclusion

The research provides insights into the barriers to microinsurance participation in rural India, emphasizing the role of both economic and behavioral factors. It calls for further studies using randomized experiments to isolate the effects of information and marketing on insurance take-up. The results are relevant for designing more effective insurance products and marketing strategies in similar contexts.