This piece was first published on the International Research Institute for Climate and Society’s blog here.
The world’s 570 million smallholder farmers are among the most vulnerable to the impacts of current and future climate variability and climate change. They need support and science-based strategies to make their communities resilient to these impacts and enable them to have sustainable livelihoods no matter what the climate might bring. One adaptation strategy being supported by the Adapting Agriculture to Climate Today, for Tomorrow (ACToday) Columbia World Project is to increase the availability of affordable index-based insurance, a proven way to help farmers cover some of the financial losses caused by drought and other weather extremes that damage their crops.
Developing effective index-based insurance products requires good data as well as input and feedback from farming communities (see video). For more than a decade, the International Research Institute for Climate and Society’s Financial Instruments Sector Team has been developing tools to help in the design, implementation and support of index insurance projects in more than a dozen countries around the world. The ACToday project has accelerated this work by supporting the further refinement of these tools to better serve farmers on the ground, and working with the World Food Programme (WFP) and World Bank to significantly scale their insurance projects in Ethiopia, Senegal and other countries in Africa.
We interviewed Rahel Diro, a senior staff associate and ACToday insurance expert to give more details about these recent developments.
Can you describe some of the tools you’ve developed with partners and agriculture communities over the years, especially under ACToday?
IRI has been working with WFP and other major partners in Ethiopia for many years to develop reliable methodologies for designing and validating insurance indexes, which are based on weather/climate data (see sidebar). We’ve learned that one of the challenges to overcome is the potential mismatch between what farmers need and the timing of payouts triggered using climate information. In other words, sometimes the index doesn’t “see” the reality on the ground and doesn’t trigger a payout, even though crops might be suffering and farmers could greatly benefit from the payout. We’ve done a great deal of research over the years on this issue specifically, and one of the things we have learned is that the risk farmers face in some regions stems more from the agricultural suitability of the land rather than the accuracy of the index. We want to avoid those kinds of situations and offer index insurance only for those areas where it makes sense and can be useful to farmers in times of drought and other weather and climate-related risks.
Under ACToday, we have developed an online mapping and analysis platform to help WFP, World Bank and other partners easily evaluate the viability of growing a particular crop in a particular location, using crop models and other data. This platform, or “maproom” as we call it, helps experts visualize rainfall patterns over wide areas, among other things. They use it to filter out areas considered marginal for the crops being insured and exclude those areas from insurance programs. Now that such a tool exists, the partners can tailor it to any of the countries where they support index insurance.
What role have farmers played in the development of the maproom, and how does the tool evolve as more information is incorporated?
All of our research and experience in this field has shown that co-designing insurance with farmers is key if we’re to develop a product that truly addresses their needs. This begins from understanding farmers’ historical exposure to drought, and is why we go through intensive exercises where we ask farmers to tell us the drought years they remember from the past 30-40 years.
In the past, IRI would produce indexes and share drafts for review by experts, but the new ACToday platform allows for true co-design. This tool has enabled local partners to make the design choices themselves, and to understand the tradeoffs between selecting different parameters. They have been able to design the drought indexes themselves. The main criterion is how much the designed index captures the historical drought experiences of farmers. This means looking back and asking, “could we have predicted the past droughts with this data?” If the history is right, then the tool is far more likely to correctly predict the future. That’s why one of the main metrics for evaluating the index is how much historical payouts from the satellite-based index are matching with the historical drought experiences of farmers.
How do comparisons between satellite observations and farmer reporting help improve index insurance tools? What does a good fit or a bad fit look like?
This is imperfect information. We work predominantly in rural areas, where data is difficult to acquire. Context is key, and we rely on the local experience of agricultural extension agents. Farmers have their own biases, and they may not remember the details very well. For instance, they may remember they had food shortages 30 years ago, but they may not recall whether it was caused by pests, droughts or floods. But for index insurance, we need to be specific. That’s why ACToday has also worked with government experts in Ethiopia’s agriculture ministry and food security offices, to create a tool that helps them review the farmer-reported drought years and conduct data-quality assessments. The output from this tool is then used as a comparison dataset for assessing the accuracy of the drought index.
The efforts involved not just IRI and farmers but a team made up of individuals from governments, universities and development agencies. Why is it important to have buy-in from all these different groups in the interest of long term sustainability?
Having buy-in from stakeholders at all levels is crucial. We want farmers to have these resources, like index insurance, even after projects like ACToday end. Making sure these tools meet the needs and are useful to farmers, getting commitments from government officials to write in these programs in future climate plans, and convincing academics to invest in the continuing improvement of the data sets all help ensure not just that these programs persist, but that they continue to adapt as needs—and the climate—continue to change.
Where does this research go next? How do you think methods might have to adapt in the next country where this process is implemented?
In index insurance, there is always some margin of error, even if it is small. A very good index might have an 85% match on droughts and payouts; another area might only have a 67% but for that market and farmer business model, it may be acceptable. The design choices are being made at the local level though, and so decisions about how to improve the margins can be made within these communities. This has huge implications for developing products that meet the needs of the clients and puts the ownership of the products in the hands of local stakeholders. Both of these traits have important implications for sustainability of these products.
The approach outlined here has been adopted in other countries. In Zambia, we’re implementing a similar process throughout the country for a government-funded program. Farmers were consulted in every seventh village, and the co-design is currently under implementation. It’s our hope that we can continue to refine the process with local stakeholders.
