IRI@AGU: Farmers’ Perceptions as Data

As our climate continues to change, it becomes increasingly important to consider the effects this will have around the world, especially in developing countries where many small farmers already struggle.  For more than 10 years, researchers at IRI have been developing index insurance, which would provide a safety net for farmers who experience devastating climate events such as droughts and floods that severely impact their crop yields.  And while advanced technology such as satellite imaging can provide crucial data about where and when these events occur, researchers have found that input from farmers on the ground has increased the accuracy and value of the data used to develop the insurance program.

Bristol Powell– Farmer Perception, Recollection and Remote Sensing in Weather Index Insurance in the Developing World: an Ethiopia Case Study

Powell will be presenting on this research on Tuesday morning in Marriott Marquis – Marquis 3-4. Full details in our IRI@AGU schedule and in the AGU program.

What does it mean to a farmer to have access to an insurance product?

When farmers have access to tools that help them alleviate risks associated with climate change, their ability to resist the worst effects of weather shocks is strengthened. Access to these tools also builds confidence in their decision-making, allowing farmers to better take advantage of “normal” years. Essentially, this could mean that a farmer that is insured would be able to take financial risks that increase their productivity during a good season.  Ideally, insurance should be just one of a suite of tools that a farmer has access to in order to confront weather-related shocks.  For example, insurance works best with existing social protection programs, savings and microcredit programs, because insurance alone won’t address all the needs of a farmer.

Why is it insufficient to only rely on remote sensing data for a history of droughts and other weather events in a given area?

Satellite-derived environmental variables such as vegetation indices, precipitation estimates and soil moisture estimates, especially when analyzed together, are extremely effective and surprisingly reliable indicators of drought over time. However, knowing farmers’ historical perceptions of drought in a specific place helps complete the picture. Drought impacts don’t always correspond directly to amounts of rainfall, although that is the main driver.  Understanding the complexity of how rainfall deficits feel to farmers gives us a clearer understanding of the remote sensing datasets and their limits.

For 10 years the financial instruments team has been working with and requesting feedback from farmers on insurance.  What have you gradually learned over time and have there been any surprises with this most recent research?

I’ve worked with the Financial Instruments Sector Team at IRI for almost five years now, and I’ve spent a significant amount of time in the field, conducting what we call “participatory processes” that we use to design indexes. What I instinctively knew from the start is that comparing what people experience on the ground to remote sensing observations is extremely valuable work.  And I’ve learned over the years what kinds of methodologies are the most effective at both standardizing data collection processes and being flexible to localized needs.  I think this balance is the key to responsible scaling up of index insurance projects. I’ve also learned that transparency in the product design and financial education are key to building trust between clients (farmers) and project coordinators and insurance companies.

How do you plan to improve index insurance following this paper? Are there plans to incorporate climate forecasting? 

This paper isn’t meant to present ideas to help improve index insurance, rather it’s more about proving that there is value in farmer perception data, even when that data is noisy.  The next step for this research will be to see if data quality improves when farmers are incentivized, either through competition or through rewards. The next generation of index insurance will most likely link to forecast-based financing efforts, which is beginning to take off around the developing world.