Anticipatory insurance with African Risk Capacity: a holistic benefit-cost analysis (August 2024)

Attachments

By Leah B. Poole, Vaibhav Anand, Alexa Gozdiff Spognardi, Anne Radday, Komal Rathod, Erin Coughlan de Perez

EXECUTIVE SUMMARY

African Risk Capacity has designed an anticipatory insurance (AI) product for drought in Malawi and Zambia; this report lays out the possible costs and benefits of such a product and a methodology for evaluating realized costs and benefits for the product, which is currently being piloted. The costs and benefits assessed are intended to be holistic, considering not just the obvious and immediate costs and benefits but also those that may be indirect, long term, or less immediately apparent. Costs and benefits were assessed with an economic model, supported by qualitative data collected from key stakeholders that led to the production of a theory of change. We conclude by proposing a monitoring and evaluation (M&E) framework that can be adapted and utilized to assess the realized costs and benefits in the event anticipatory insurance is implemented. This initiative is a result of a partnership between OCHA and ARC, with OCHA contributing expertise in scaling up coordinated anticipatory action, financing product development and stakeholder engagement costs, and investing in learning, such as the benefit-cost analysis presented in this study.

Key Outputs and Results Connected to the Theory of Change

Expected positive benefits of the AI product centered on reduced use of negative coping strategies by drought-affected populations and improved yields, which resulted from use of aid such as replanting using the AI payout immediately after drought affects the original crops. Critical assumptions identified in the theory of change that could potentially pose challenges or points of failure for the product included premium defaults and other administrative-related delays or errors, whereby individuals or entities fail to fulfill their financial or contractual obligations, disrupting the anticipated cash flow essential for success. Once a payout is triggered, timing and delays at every step of the implementation process can lead to misalignment between projected and actual outcomes. Relatedly, poor targeting systems have been identified as a significant barrier to taking timely action. Another point of caution arises from procurement issues, which may include challenges in sourcing necessary resources or services, hindering the ability to deliver aid to recipients. Additionally, there may be delays in the disbursal of funding and materials as well as issues with the efficacy of the farm-level actions (e.g., yields for the replanted foods). Qualitative interviews revealed a variety of suggestions for the design of an AI product, including the consideration of climate change in the selection of attachment points, complementarity to existing farm subsidies, inclusive targeting, public engagement, government coordination, avoiding procurement failures, and the effectiveness of replanting. One proposed approach to ensure that these assumptions hold is to consider pausing the product’s development temporarily to improve speed and test the effectiveness of certain actions at different timings. Alternatively, African Risk Capacity (ARC) could explore strategies that minimize potential points of failure in the product’s implementation, such as requiring countries to generate and maintain beneficiary lists in advance and providing administrative support to ensure that contractual errors do not disqualify the country from coverage. In order to ensure the product’s success, and a benefit-cost ratio greater than 1, it will be beneficial to analyze the various pathways available, especially during implementation, and select the one with the fewest identified points of failure. This approach could enhance the likelihood of achieving intended outcomes at the highest benefit ratio.

Key Results Connected to the Economic Model

The benefit of an anticipatory insurance (AI) product originates from two main sources: (1) its ability to provide forecast information to decision-makers for early action, and (2) its insurance mechanism that offers financing for these actions when it is likely most needed. Thus, the relative advantage of the AI product to a country depends on the country’s access to forecasts and its capacity for early action. The AI product is welfare improving in scenarios where a country lacks the capacity for early action—due to limited access to forecasts, financing, or the institutional capabilities required for implementation. However, for countries that can utilize forecast information effectively and have the capacity for early action, the incremental benefit of the AI product may be limited. For such countries, the AI product proves beneficial primarily when forecast accuracy is high, but the available early actions have a lower benefit-to-cost ratio.

The economic analysis highlights the importance of having a robust capacity in order to take advantage of forecasts and proactively manage risks. This capacity may include access to forecasts, a supportive institutional framework, predetermined standard operating plans (SOPs), training programs, financial arrangements, and an effective last-mile delivery infrastructure. Many countries may lack this capacity, and reallocating resources from ex-post aid and other priorities to develop this proactive capacity can be challenging. The current pilot of the AI product can help bridge this gap. It offers an opportunity for countries and stakeholders to assess their current capabilities and commit to develop the necessary infrastructure and processes.

Key Recommendations Connected to the Monitoring and Evaluation Framework

We recommend monitoring and evaluating any anticipatory products that are introduced to the market to assess whether the stated assumptions were realized. Benefit-cost ratios for specific actions can be calculated based on post payout evaluations, providing further data on the effectiveness and potential areas of improvement in the development of novel AI products.