Index-based and loss assessment-based weather insurance contracts: toward a two-level group-individual approach
Weather risks are a major cause of under-investment in agriculture and of losses of past investments and diminished livelihoods with eventual serious hardships when uninsured shocks occur. Climate change makes protection against weather shocks increasingly important. Yet, conventional insurance contracts based on loss assessment has had very limited success due to difficulties in implementation and high costs, particularly for small farmers. Index-based weather insurance offers a new promise in facilitating implementation and reducing costs. Yet, it has had limited uptake without extensive subsidies. New approaches are being explored to increase uptake among smallholder farmers by reducing basis risk, interlinking index insurance with the purchase of inputs, or offering interlinked credit-insurance contracts.
Another approach that is starting to be explored is the design of two-level contracts that combine index-based insurance at the level of a producer group, such as a cooperative, with traditional loss assessment-based insurance internally to the group. This offers promise as it can cumulate the advantages of the two approaches while avoiding their pitfalls: reduce basis risk at the level of the group as a whole, and capitalize on direct observability of damages and moral behavior based on repeated games among members of the group. Yet, design of the corresponding contracts and incentives for contractual parties is yet to be developed. Specific experiences with implementation need to be constructed and analyzed. The role of foreign assistance in helping support the start-up of such schemes needs to be explored.
This is the objective of this FERDI workshop to be held in Clermont-Ferrand on June 21-22, 2011.
Mardi 21 Juin 2011
Patrick Guillaumont , Emeritus Professor, Université d’Auvergne, Président, Ferdi
Alain de Janvry, Workshop coordinator, Senior Fellow Ferdi and Professor at the University of California at Berkeley
Michael Carter, University of California at Davis and I4 Project : “The impact of interlinked index insurance and credit contracts on financial market deepening and small farm productivity”
Stefan Dercon, Oxford University : “Reinsuring the poor : Groups, indices, and samples” (presentation)
Elisabeth Sadoulet, Ferdi and Professor, University of California at Berkeley : “Review of group-insurance issues and research questions” (paper)
Daniel Clarke, Oxford University : “Reinsuring the Poor: Group Microinsurance Design and Costly State Verification”
Ethan Ligon, University of California at Berkeley : “Risk management in the cooperative contract: Intra-coop asset heterogeneity and fluctuating membership”
Vianney Dequiedt, CERDI, Université d’Auvergne : “Group insurance against common shocks”
conduite par Jean-Philippe Platteau, Université de Namur
Mercredi 22 Juin 2011
Catherine Guirkinger, Université de Namur : “Group Insurance: The Case of Cotton Producers in Mali” (presentation)
Xavier Giné, The World Bank : “Microinsurance : a case study of the Indian rainfall index insurance market”
Ruth Vargas-Hill, IFPRI : “Offering rainfall insurance to informal insurance groups: evidence from a field experiment in Ethiopia” (présentation)
Craig McIntosh, University of California at San Diego, and Alexander Sarris, University of Athens : “The role of group contracts in interlinking credit with insurance: An application to Ethiopia”
Jing Cai, University of California at Berkeley : “Role of social networks in insurance take-up in China”
René Gommes, Joint Research Center, Ispra, European Commission : “Calibrating rainfall index insurance payoffs” (présentation)
Round table and conclusions
Alain de Janvry, Ferdi and University of California at Berkeley
Patrick Guillaumont, Professeur émérite, Université d’Auvergne, Président, Ferdi
Jan Gunning, Free University of Amsterdam
Michael Carter, University of California at Davis and I4 Project
Stefan Dercon, Oxford University
Ethan Ligon, University of California at Berkeley
Craig McIntosh, University of California at San Diego
Three questions to…
Alain de Janvry
Alain de Janvry is a Professor at the University of California, Berkeley, and has been a Senior Fellow at the Ferdi since 2010. He is an expert in agricultural issues in developing countries, and coordinator of the workgroup "Index-based and loss assessment-based weather insurance contracts: towards a two-tier group individual approach" which was held at the Ferdi on 21 and 22 June.
What does this business forum organised by the Ferdi seek to achieve?
In my opinion, development economics exhibits two major market failures: the first involves access to credit, the second, access to insurance.
The past few years have been marked by tremendous progress in the credit sector. Microfinance represents a genuine revolution that has spawned a number of key institutional innovations. It has given millions of poor people (mainly micro-entrepreneurs and traders) who do not have the backing of financial or material guarantees an opportunity to access credit, thus enabling them to develop a sustainable economic activity.
In the field of insurance, and agricultural insurance in particular, a lot of work remains to be done. When developing countries experience an uninsured climate shock, the affected populations can be hit by severe economic repercussions. Conventional, loss assessment-based insurance mechanisms have endured widespread failure in developing countries, the main problem being that it is just too expensive to carry out systematic verification of the effective losses, both for insurance companies and for small-sized producers.
In the microfinance sector, however, work by researchers and professionals in the field has paved the way for setting up viable financial institutions despite the absence of guarantees. In the insurance domain, it is absolutely vital today to be able to identify profitable systems without having to carry out a systematic verification of losses.
There is another major difficulty relating to insurance claims. Producers in developing countries only have limited knowledge of these particularly complex issues, and have an innate distrust of insurance agents.
This workshop, organised by the Ferdi, is seeking to answer two key questions. Firstly, how can we improve the offer? In other words, which are the best products to offer that will be of equal benefit to producers and insurance companies alike? Secondly, what can we do to encourage more frequent use of insurance contracts, and to build up a climate of trust between insurance companies and insured persons?
Why is it so complicated to set up weather insurance contracts in developing countries?
There are several underlying reasons for the difficulties in setting up this type of contract. For instance, we encounter three main hurdles regarding the insurance contracts offer. Firstly, the weather data we have for these countries only covers fairly short timescales and, furthermore, is sparsely distributed in geographic terms (there are very few weather stations, and the ones that do exist are not recent). This makes it difficult to establish statistical laws on weather events.
Secondly, there is a serious lack of historical data on effective yields. Thirdly, local insurance companies have only limited experience of the farming sector.
With respect to demand, again, we come up against three major obstacles. Firstly, farmers have great difficulty understanding even the most basic reasoning behind insurance contracts, which consists in paying in advance for a product that will only be used at a later date. Secondly, they have only a very limited cash flow, which means that they are often unable to pay an insurance premium prior to the start of the production season. Lastly, coming back to a point I mentioned earlier, there is an evident lack of trust between farmers and insurance companies.
Taken together, all these issues represent a considerable number of concerns. Nonetheless, developing countries do also boast important advantages. One striking example is the number of traditional mutual support structures that exist, and that could be used as a basis for setting up group insurance contracts. Another advantage is that these countries possess the means for linking insurance contracts to credit facilities, thereby encouraging access to credit and enabling the eventual adoption of technological innovations that, while available, are too often under-used.
The past few years have seen significant advances in research into agricultural insurance. In your opinion, what are the keystone developments?
It's true; we have made enormous headway in this domain. Several field tests have been carried out (in Mali, India, Ethiopia, and China), which have provided valuable insight into ways of improving the products on offer. We now recognise two major types of index-based insurance contract. Firstly, weather insurance contracts, which involve indemnifying producers when climatic variables are above or below a previously set threshold. For example, producers are automatically compensated with a financial sum when rainfall levels are below the threshold necessary to produce sufficient yields. The second type is index-based insurance contracts based on mean regional yield, calculated per period. This insurance system avoids the need to verify yields on an individual basis. For instance, if the mean yields recorded are 20% below typical yields for the region in question, all the producers in the area are automatically indemnified.
We also place a special focus on the group concept. In the world of microfinance, the preferential adoption of group credit facilities (rather than individual credit agreements) has proven to be a particularly effective way of overcoming difficulties linked to the moral hazard and to information asymmetries. In the insurance sector, we consider that the setting up of group insurance contracts should be one of our top research priorities. The theory is that, in the event of shocks, the group receives a comprehensive indemnity, which it then distributes to its members in proportion to the losses suffered by each individual. So, to come back to your first question, outside of our two main objectives, this workshop also has more specific targets i.e. to establish exactly where we stand at present with regard to the specification of group contracts; and to draw up a list of the internal group rules that need to be developed and put into practice.
Interview conducted on 22 June 2011 by Claire Gillot, Communications Officer, Ferdi