Impact and lessons learned from the GIACIS project

In another article, we described the VICI product and what made it unique. Read about it here: How geodata-based insurance can help smallholder farmers cope with weather risks. 

In this article, we'll discuss some of the impact and lessons learned from the project. 

Introduction

GIACIS has been successfully implemented in Ethiopia and there are efforts to scale it to other regions in Africa.

In Ethiopia, the GIACIS approach is now part of the larger Resilience Enhancement Package (REP). Through this project, nearly 20,000 smallholder farmers received training and over 3,400 farmers enrolled in the insurance scheme, contributing a total premium of 780,000 Ethiopion Birr (ETB, 1 Euro was equivalent to approximately 39 Birr in 2020). Drought in the 2020 season meant that there were around 256,000 ETB payouts in total for 999 farmers. More than half of the insured farmers signed up again for the insurance in the 2021 season.

This approach is currently in Ethiopia the only officially approved and widely supported crop insurance scheme to cover small-scale farmers against drought. Kifiya Financial Technology, the business partner for the project in Ethiopia, continues to work on scaling up the project to include more farmers in Ethiopia in the coming years.

Meanwhile, ITC and VITO are collaborating on developing similar services in other regions of Southern Africa, with ITC being the key NDVI data provider.

Key lessons learned

From an operational perspective, key lessons have been learned in developing a geodata-enabled insurance scheme for supporting food security and economic resilience for smallholder farmers:

• Collaboration is needed across the entire data value chain. This is because a business model that depends on input data from elsewhere creates risks, such as delays or bugs that have cascading impacts on the ability to provide the service on time and of the expected quality. Therefore, the whole pipeline should be somehow involved, especially where people's money is concerned.
• Satellites and platforms have a limited life expectancy, and this affects the continuity of developed algorithms. This is because the source, quality, specifications and frequency change as data inputs change. Adapting an approach to new data with different specifications is not easy, which can disrupt products that are already in operation. It takes a long time to establish a stable algorithm for supporting insurance, and once it starts to function, the sensor or platform may be discontinued or changed.
• Brilliant technical design and savviness are not sufficient for the success of insurance schemes. The success of developing and scaling a geodata-based insurance product is not just based on technical viability but also external non-controllable factors. For example, the war in northern Ethiopia disrupted some implementations, while the COVID-19 pandemic restricted required travel and interactions.
• Smallholder farmers may not be the best clients for micro-insurance schemes. While smallholder farmers may have sufficient numbers required in private insurance schemes, in the majority of cases, they are faced with other constraints to sustain premium payments for 9 years and expect payout in the tenth year.

This explains why there is high sign-up in the early periods of such schemes, and many clients drop out over time as they don't realize the benefit. These smallholder farmers are faced with more immediate decisions to feed their families, pay for healthcare, and invest in the education of children, among others, from meagre resources. An investment in insurance premiums is not a priority in these cases.

Commodity buyers, input suppliers or other intermediaries could be the best clients for this kind of insurance unless public sector resources can make these into 'welfare' insurance schemes. The target could, therefore, be medium to large-scale farmers or those producing for markets and not for consumption.

• Local capacity building is important for the implementation of geodata-based finance schemes. There is a need to build local capacity for data processing to ensure ownership, scalability and sustainability of the product. While technical backstopping can always be provided, there is a need for local people to be involved in technical aspects so as to build trust and wider acceptance of the product.
• Bundling insurance schemes with other services increases the penetration and acceptance of agricultural insurance products. Farmers are faced with complex decision matrices, and therefore, schemes that embed or combine insurance products in products or services may stand a better chance of success. For example, these micro-insurance schemes can be bundled with prices of inputs such as seed or commodity sales contracts or farm loans.

Conclusion

Geodata-based insurance provides a basic safety net to protect smallholder farmers against drought. In doing so, it enables the expansion of financial service delivery to smallholder farmers. Such a scheme goes beyond just providing insurance; it offers a bundled credit facility to purchase agricultural inputs and reduces farmers' fears of investing in more and better agricultural inputs and technology. This ultimately leads to food security.

Working with other partners and stakeholders, there is scope for further development, such as schemes to include other services and to expand to new regions.