Our Programmes



Sign up to our newsletter.

Follow ODI

Disaster risk insurance: understanding and improving the reliability of products

Written by Lena Nur


Recently, part of the international community working on index-based insurance in agriculture came together in Dakar to discuss ‘lessons learnt and the way forward’. One clear lesson is that reliability is critical to the impact of insurance schemes.

Index-based insurance is deemed to increase coverage because premiums are cheaper. Pay-outs are based on a trigger linked to data from rain gauges, satellites or yield monitors. This data is used to approximate losses and determine the level of the pay-out. Ideally, this makes index-based insurance less expensive than ‘classic’ insurance products that address claims on a case-by-case basis.

In this context, index-based schemes face two key challenges: strengthening the reliability of pay-outs and making sure that these pay-outs accurately correspond to the relevant loss that has occurred. Otherwise, the schemes can become a burden to all parties involved. Consequently, transparency, communication and innovation around basis risk need to be increased.

Basis risk and the reliability of insurance products

Basis risk describes the possibility that insurance does not pay-out when it should. In the case of index-based products, there can be a mismatch between the index that has been applied and the actual losses that have been experienced. This can result in either shortfalls, when appropriate pay-outs are not triggered, or overpayment, when pay-outs are higher than necessary.

This is a serious challenge to the reliability of index-based schemes and can render compensation mechanisms ineffective. Basis risk may arise because of the design of an index or because of the way in which it captures losses at different locations. Or, it may occur because of a difference between expectation and practice. For example, an index may be calibrated towards a certain type of groundnut seed but if the insured farmer uses a different variety – one that needs more water or time to mature – the links between the index and the actual damage from a rainfall deficit will be flawed. As a result, the farmer may experience a shortfall in pay-outs.

What does this mean for the insured, insurance companies and delivery channels?

Index-based insurance is often touted as a solution to under-insurance in developing countries. However, a shortfall in pay-outs can have particularly detrimental poverty impacts when target groups – who in many cases are already likely to have a low income and few assets – do not receive appropriate compensation. This increases distrust of the products and raises the perceived opportunity cost of paying a premium. Similarly, insurers may be less willing to provide cover in contexts where they are at risk of over-paying or suffering reputational damage. Consequently, basis risk may be an underlying factor in low take-up rates and explains why some insurers are reluctant to offer index-based products.

What to do about it?

While insurers, brokers, reinsurers and modellers undertake research on basis risk, this is often not available to others. Definitions and understanding of basis risk vary, and the publication of quantified basis risk is scarce. This means its impacts on policy-holders are not well understood.

Technology may have a role to play. Indices used in microinsurance and sovereign risk pools – such as the Caribbean Catastrophe Risk Insurance Facility (CCRIF) and the African Risk Capacity (ARC) – are usually complemented by ground surveys and combined with other data. These indices can be reviewed and refined over time. Technological innovations for reducing basis risk and spot-checking claims, for instance using mobile phones, may improve the usefulness of these refinements.

For the remaining basis risk, specifically dedicated funds could be a way to assist those who, despite high losses, do not receive pay-outs. This was the case, for example, under the R4 Rural Resilience Programme that compensated insured farmers when the insurance trigger mismatched losses in Ethiopia and Senegal in 2015. Alternatively the level at which protection begins could be lowered, although this has the disadvantage of increasing the premium price.

In either case, it is necessary to clarify who will, can, or should pay for dealing with basis risk. At the very least, intensive communication and efforts to support financial literacy would help policy-holders better understand the potential risks and potential benefits their policy represents.

At the Global Index Insurance Conference, Craig Thorburn, Lead Insurance Specialist at the World Bank, highlighted the importance of communication – with clients and insurance supervisors – in strengthening consumer protection. An important step in this process is comparing the years perceived by farmers to have been bad years with those in which the index would have triggered a pay-out in the past. These ‘back tests’, said Thorburn, can be used to explain and illustrate products; offering farmers a more informed choice between affordable products with higher trigger levels and more expensive products that pay out more often.

Even with index refinements and technological enhancements, basis risk cannot be removed from index-based insurance products. Gaps in localised weather observations, lack of affordable data and limited technical capacities in many developing countries tend to reinforce it. Consequently, there is an urgent need to clarify ownership and management of basis risk; whilst continuing to improve communication, the translation of data and transparency about the reliability of schemes.