Implementing a Pilot Program for Index Based Weather Risk Management: The Case of Malawi

1. Implementing a Pilot Program for Index Based Weather Risk Management: The Case of Malawi Erin Bryla, Joanna Syroka, Shadreck Mapfumo Commodity Risk Management Group, The World Bank December 5, 2007

2. COMMODITY RISK MANAGEMENT GROUP • Focus on Agriculture • Provides technical assistance on: • Market based price risk management instruments • Index based weather insurance • Working to improve access to risk management opportunities by: • Researching risk management alternatives • Introducing new products through pilot programs • Working with regulators and governments • Disseminating best practices and lessons learned • Works with: • Banks, microfinance organizations, other financiers • Insurance companies • Ginners, processors • Cooperatives and producer associations • Governments

3. INDEX BASED WEATHER INSURANCE IN DEVELOPING MARKETS • Traditional crop insurance for smallholder economies is extremely challenging • Main constraints for traditional products, based on individual loss assessments: • Poor rural insurance infrastructure and capacity • Operationally difficult for small farmer agriculture • Loss adjustment, availability of farm level data • Moral hazard • Adverse selection due to asymmetric information • High monitoring and administrative costs • Agricultural production risk needs reinsurance • Covariant risks (e.g. drought) are an inherent characteristic • Multi-peril Crop Insurance • High Administrative Costs • Moral Hazard • Adverse Selection • Index-Based Weather Insurance • Rainfall is a proxy for damage • Objective triggers and structured rules for payouts • Improved correlation between need and provision VS

4. WHY WEATHER INSURANCE IN MALAWI • Malawi Facts: • Population: 12 million; 165th Human Development Index • Food crop - maize; export crop - tobacco, tea, cotton, coffee, sugar • One rainy season, November-April • 45% of GDP comes from agriculture • 87% of total employment is ag • 64% of rural income is from agriculture • Drought and Lending (2004/2005): • Recovery rates for ag lenders in the range 50-70% • Ex. One big bank that lost $110,000 to smallholders farmers stopped lending to those farmers • Major government and donor lending program lost 50% of value over 5 years due to drought-blamed default and the program was discontinued. • Microfinance institutions stopped lending

5. PILOT OBJECTIVES In 2004 the National Smallholder Farmers Association of Malawi (NASFAM) wanted to expand its operations and grow the Malawi groundnut market domestically and for export, but high risks from drought and high loan default rates deterred financing institutions from providing loans. The Pilot Program aimed to: • Gives farmers the ability to mitigate drought risk • Secure access to finance and inputs for improved production • NASFAM training and higher quality seed will improve long-term production and revenues • Protects both producer and loan provider from weather-related production risks • Allowing banks to expand their lending portfolios in a managed way • Gives insurers the opportunity to re-enter rural markets • No regulatory impediment and reinsurance potential • Little (and bad!) experience with traditional agriculture insurance • Opportunity for NASFAM to expand its operations and grow the Malawi market domestically and for export

6. Historical weather data & weather stations Length of historical record - 30 years or more Quality controlled, cleaned, enhanced Reliable ongoing collection and reporting procedures Third-party settlement data Ability to index risk Basis Risk or “How good is this insurance?” Mismatch between coverage and actual result How can it be minimized? Yield data that shows strong correlation with weather event Local insurance company willing to intermediate product PRE-CONDITIONS

7. Malawi Met Office data excellent: over 30 years, few gaps, 21 primary synoptic stations (red) In addition over 200 rain gauges around the country which can be leverage. CRMG/Met Office piloting the installation of new automatic weather stations (green) and creating historical synthetic data, opening new agricultural areas to weather insurance for this year Securing donor funds to scale-up infrastructure pilot significantly next year Weather Infrastructure • Dense, high-quality network

8. Standardized Contract Design • Balance simplicity that farmers and stakeholders can understand, with the complex dynamics that characterize water stress impact on crop yields: • Easy to communicate to farmers and stakeholders • Performs well from agro-meteorological perspective • Provides required protection for all stakeholders at an affordable level • Captures local conditions and environment • Simple to replicate to other locations and crops so that programs are scalable • Local ownership, so programs are sustainable

9. Payout ($)  Payout ($)  Payout ($)  Deficit Rainfall (mm)  Deficit Rainfall (mm)  Deficit Rainfall (mm)  PHASE 1 Sowing & Establishment PHASE 2 Growth & Flowering PHASE 3 Yield Formation to Harvest Cropping Calendar*  Sowing Window & Dynamic Start Date * Cumulative 10-day rainfall is capped to prevent excessive rainfall impacting the phase-wise total Standardized Contract Design • Given a target premium and set pricing guidelines,and a required maximum payout per phase, red dots are calibrated to a simple crop water-balance model – the FAO WRSI – cross-checked against historical yields and local experts (blue dots) to minimize farmer income Value-at-Risk and maximize payout correlation to yields

10. THE PILOT • Loans to cover seed, insurance premium and interest: • Opportunity International Bank of Malawi • Malawi Rural Finance Corporation • Policies: • Insurance Association of Malawi (seven companies pooled the risk) • Premium: 6-7%, Max Payout per farmer: Loan Size given by bank • Seed & Product Distributor: • NASFAM: Groundnut in 2005, Groundnut & Hybrid Maize in 2006 • Participants: • Farmers all members of NASFAM clubs • 2005: 900 farmers, 4 weather stations, sum insured $35,000 • 2006: 2500 policies, 5 weather stations, sum insured $110,000 • Insurance Payout Payment details: • Payout: channeled from insurance company directly to the bank; • No Payout: farmers benefit from selling the higher value production

11. Insurance Insurance MET OFFICE MET OFFICE Step 7 Step 7 Association of Association of S S Malawi Malawi t t e e p p Step 3 Step 3 8 8 Step 6 Step 6 MRFC/ MRFC/ NASFAM NASFAM Step 10 Step 10 OIBM OIBM S S t t Step 5 Step 5 Step 2 Step 2 Step 4 Step 4 e e p p 9 9 Step 11 Step 11 CLUB CLUB Step 1 Step 1 ORGANIZATION OF THE PILOT PROGRAM

12. EXPANDING THE PILOT: A PORTFOLIO APPROACH • Pilot program reinforced that weather insurance alone cannot solve agricultural lending problems • Initial pilot programs faced: • Side-selling, due to a nascent agricultural supply chain for groundnut • Leading to non-weather related defaults • Operational challenges in coordinating stakeholders • Therefore for 2007 pilot efforts: • Focus on established agricultural supply chains, e.g. tobacco • 70% of current loan portfolios • Cash crop loans also have a maize component that can also be insured • Look to economies of scale and critical diversification for insurers • Reduce premiums per policy • Increase volumes to attract reinsurers • Looks at a portfolio level approach rather then an individual farmer policy • Comfort for lenders when considering expanding their portfolio and reducing the risk (and therefore cost) of lending • Tie-in with emerging contract farming relationships in Malawi • Will be expanded to individual farmers in 2008

13. MAJOR PILOT ACHIEVEMENTS • Unlocking credit facilities for smallholder farmers. • 1800 farmers formerly excluded from financial markets. • Before pilot OIBM did not lend to agriculture, but now is using lessons learnt from project to expand lending book. • Four other banks promised to unlock more than USD 10 million of credit if weather risk is insured. • Access to high yielding seeds and fertilizers. • Farmers interviewed indicated that they got an average of SIXTY 50 kg bags (by using hybrid maize seed) as opposed the usual 20-25 bags. • Improved sustainability for contract farming operations • Sustainability of credit markets: • Quantification of weather exposure • Can expand lending in a managed way • In case of another 2004/5 drought loans will be paid off • Banks have flexibility to protect themselves either through marketing insurance directly to farmers or through a portfolio approach • Expansion of the risk retention capacity through risk transfer to the international market

14. DATA REMAINS A CHALLENGE FOR GROWTH • Basis Risk • Effectiveness of the index is driven by availability of weather (rainfall) data • Without local data, basis risk events are more difficult to control • Contract Pricing • Without weather data insurers and reinsurers will need to increase the cost of uncertainty for the weather insurance product as well as the risk margin • Increasing Participation • Without more stations increasing the number of participating farmers will be challenging • Installing a new station in 2006/7 added 672 farmers • Upgrade of 50 other stations will capture 300,000 farmers • Need to leverage the existing rain gauge network • Remote Observation and Satellite Data • Merging remote data with ground data can expand the market for index based products • Satellite rainfall data is only as good at the underlying ground data • How can existing data be harnessed to expand the market for index based products • In many cases data exists, it is just not accessible. • What will the market take? • How to make more data available? • Investment in new infrastructure

15. WHAT COMES NEXT … • Training and Capacity Building Program: • Local capacity to design contracts • Technical training for the insurance industry • Training to other agriculture sector actors on products and program management • Piloting • Risk assessment techniques for financiers • Investment: • More new weather stations and better data • Currently approximately 110,000 tobacco growers can be serviced by existing network • An investment of $1.2 million can increase this to 280,000, 13% of which are currently receiving financing • Broader risk management activities: • Linkages to other institutional arrangements, e.g. biometrics, credit bureau, price risk management • Government-level weather risk management • Work with regulator on comprehensive regulatory framework