A new model that turns uninsurable risk viable

In an ideal world, insurance policies would be transparent, free of lengthy small print and would pay out immediately once a loss occurs, following a streamlined and effortless loss assessment process. 

 

In pursuit of this ideal, insurtech companies often pitch their services in the style of a sports forecast, citing onboarding times, rapid loss assessment and record-setting payout speeds.

 

US insurtech Lemonade, for example, claims the unofficial world record for a payout of just two seconds. This is the time the company’s AI-enabled system took to assess a claim, check the policy conditions, perform anti-fraud checks and initiate a transfer to the customer’s bank.

 

Other digital insurers are striving to match Lemonade’s headline-grabbing speed. While not quite as fast, many now settle claims within minutes or hours, a considerable improvement on the weeks and even months legacy settlements can take.

 

Customer expectations versus reality are well-reflected by the figures quoted by claim-automation engine provider Sprout.ai: while 21 per cent of insurance customers expect claims to be settled in hours, 43 per cent of them are waiting over two weeks on average and 33 per cent two weeks or more.

 

Enter parametric insurance

 

When it comes to payout speed, digital insurers face also competition from an entirely different model: parametric insurance.

 

Unlike traditional indemnity insurance, which requires the insured to suffer a verifiable loss, parametric insurance pays out based on a predefined event, such as an earthquake of a certain magnitude or a cancelled flight.

 

This model doesn’t reimburse actual losses. Rather, it provides a pre-agreed payout when certain objective parameters measured by independent third parties are met. For this reason, it’s also referred to as index-based insurance.

 

While this model echoes the inner logic of a wager, it’s setting the indices and thresholds right that can distinguish it from one.

 

The traditional indemnity and the new parametric models manage basis risk – the difference between actual losses and payouts – very differently.

 

Indemnity insurance mitigates basis risk through detailed policy terms, clearly defined deductibles and pedantic exclusions – hence all the small print.

 

The payout not being dependent on an actual loss suffered by the insured but on the occurrence of an event means parametric insurance relies on model accuracy to ensure that payouts align with real losses.

 

This makes the selection of the indices and mechanisms triggering payout critical. Leveraging a toolkit of accurate and relevant data and advanced analytics tools, parametric insurers must strike a delicate balance to avoid both under-compensation and windfalls.

 

As Satya Beekarry, Partner at accounting firm PKF Littlejohn puts it, “The future success of parametric insurance largely depends on how accurately and consistently it can model the actual losses that would arise from traditional insurance.”

 

An alternative or a gap filler?

 

Beyond basis risk, the other major challenge for parametric insurers is regulation – or the lack of it. In the UK and Europe, parametric insurance is regulated under general insurance rules, which may not accommodate its unique model.

 

Given parametric insurance’s strength in covering natural catastrophes, especially those uninsured by traditional players, it’s little surprise that Latin and Central America, with their huge underinsured agrarian populations and frequent extreme weather events, have become trailblazers in developing and implementing regulatory frameworks.

 

The largest recorded parametric payout of $85.4 million, for example, was made by the Caribbean Catastrophe Insurance Facility (CCRIF) following Hurricane Beryl in 2024.

 

In these regions, quick payouts based on objective triggers can be game-changers for farmers, who have historically lacked protection against natural catastrophes (“nat cat”).

 

But index-based insurance is rarely comprehensive. Hurricane cover doesn’t provide any protection in the event of drought or pest damage.

 

That’s why parametric insurance can work best when combined with traditional insurance, where it can complement the main policy as either an add-on or as a top-up when its limits are unlikely to provide sufficient compensation for the potential loss.

 

The bundling of traditional and parametric products is also the most likely future trend as the advancements in data analytics and AI expected to improve parameter setting will also streamline the underwriting, loss assessment and payout processes in traditional insurance.

 

More transparency, quicker payouts and better customer experiences in indemnity insurance are expected to dissipate some of the competitive advantages of parametric insurance.

 

What might the future bring?

 

In 2023, the global market for parametric insurance was estimated between $14.8 billion and $18 billion in premium volumes, representing just around 0.8 per cent of the global property and casualty (P&C) premiums, yet 13.7 per cent of global insured losses from nat cat events.

 

This disparity suggests that parametric insurance is currently underused relative to the scale of nat cat losses and has considerable potential for market growth.

 

New York State’s formal authorisation of parametric insurance as a legal category under state law on 12 January 2025 was undoubtedly a milestone on the journey that leads this innovative new product out of insurance’s grey area.

 

The new legislation also underscores the complementary nature of parametric insurance by mandating insurers to warn clients that parametric products are not meant to replace but only supplement traditional insurance products.

 

By law, insurers must also inform policy holders about the risk of not fully recovering their losses and point out to them that “The policy provides coverage based on parametric triggers and may not indemnify for actual losses incurred.”

 

While this is a critical legal safeguard, it may not sound like a strong selling point.

 

What will truly encourage adoption is a track record of positive outcomes – instances where well-calibrated indices have led to fast and fair payouts in transparent and seamless ways.

 

To achieve this, insurers will need more data integration, sophisticated data analytics and scenario modelling tools. The bar is set high, but the digital tools that can help clear it are increasingly available. The devil, however, will be in the details of deployment and application.