Africa RiskView
African Risk Capacity's software that turns satellite rainfall estimates into a count of people affected by drought, then into a response cost, and serves as the trigger index for its sovereign policies.
Africa RiskView is the software African Risk Capacity (ARC) uses as the trigger index for its sovereign drought policies. It turns satellite rainfall estimates into a count of people affected, then into a response cost. The tool was developed in the late 2000s by the World Food Programme, then taken over by ARC, which made it the engine of its drought contracts. The model works in three stages. The first is agrometeorological: from satellite rainfall estimates, it computes for each grid cell a Water Requirement Satisfaction Index, the WRSI. The index comes out of work by the FAO, the Food and Agriculture Organization of the United Nations, and tells whether the crop got the water it needed at each stage of the season. The second converts that index into a number of people affected, using a vulnerability profile specific to the country. The third multiplies that number by a response cost per person. The output, in dollars, is the contract index: the policy pays when it crosses the attachment point, up to the limit. Most of the work happens before signing, in what ARC calls customization. A national technical working group picks the rainfall source, the reference crops, the planting windows, the zones and the vulnerability curve, and each of these choices shifts the trigger probability. That is where basis risk gets built in. A wrong planting calendar, a poorly chosen crop or a rainfall source that underestimates local showers is enough to keep the model silent through a real drought. The government only finds out at the end of the season. For a reinsurer, the appeal of the model is its transparency: parameters are fixed and known before the season, and the output can be recomputed. Thirty years of rainfall can also be replayed to estimate expected loss. The flip side is that contract quality depends almost entirely on the customization, which the buyer does itself.
Textbook case, invented figures: in a Sahelian country, the technical working group picks millet and a planting date of June 15. The rains come late, farmers plant on July 10, and the actual season is poor. The model, which counts water needs from June 15, reads a moderate drought: 400,000 people affected instead of the 900,000 counted on the ground, below an attachment point set at 600,000. The policy pays nothing, and the recalibration will only apply to the next season.
Africa RiskView, ARV, ARV model, African Risk Capacity model
Follow the research
One letter a week. First, the week's underwriting note, when one is published: a dated, argued decision, signed by Paul Bernardski. Then what the observatory published on cyber risk, political risk and insurance. No account, no tracking.
No spam, one-click unsubscribe, no data ever sold.