Parametric insurance: the future of risk transfer?

The past year has seen a rise in extreme weather events across the globe. Typhoon and hurricane seasons were stronger and more damaging than in previous years, and businesses need better solutions for their exposure to such events.

Traditional insurance options have left many players dissatisfied. The amount of time that elapses between a catastrophic event taking place and the insured entity receiving payment for a claim is, in most cases, subject to some delay, which adds further frustration, disruption and damage.

“With innovation, the insurance sector can improve its value proposition,” says Tanguy Touffut, CEO of AXA Global Parametrics. “You should receive your payments quickly, but there is too much bureaucracy involved.”

Things seem to be changing since the introduction of parametric approaches to risk transfer, but so far only on a relatively small scale.

“Weather risk is the main catalyser for parametric options,” says Stephen Moss, director at Risk Management Solutions (RMS), a catastrophe risk modelling company. “The general aim is to reduce financial volatility and uncertainty for clients.”

Driven by data

Parametric products are driven by data. They are characterised by index thresholds that are agreed in advance and designed to trigger automatically when a certain event happens. As opposed to traditional insurance, the pay-out is agreed in advance upon the occurrence of a specific event that hits a pre-established threshold.

“There is a clear benefit over traditional insurance,” argues Moss. “Parametric insurance is based on a specific need for a specific entity.”

The novelty lies in the risk being measured directly on location. If the trigger is how much (or little) rainfall within a given period, there is no need for insurance companies to explicitly verify the damage incurred by a client.

The outcome of risk is measured by a parameter or mix of parameters which then correlate to (direct or indirect) losses potentially affecting the client.

This type of insurance might not indemnify the client’s loss in its entirety but can cover a wide range of risk aspects that traditional products might not. Business interruption is one application, and a risk that has been in vogue in recent months.

Crucial to the concept is that a third party is involved that provides the relevant data to both insurance buyer and insurance provider. The providers can vary widely, depending on the metrics involved in the risk.

Transparency is a potential game changer. “The pay-out is triggered by a third party, not by the insurance company providing the cover,” emphasises Touffut.

The data acquired to model and calculate risk is provided by certified international monitor agencies that possess the tools to accurately analyse the specific area under scrutiny.

“We don’t govern the data ourselves, but we receive it and then we use algorithms to analyse them according to our clients’ needs,” says Eric Schuh, global head of property and casualty solutions at re/insurer Swiss Re.

Understanding the potential applications of data remain widely unexplored in the insurance markets, but the technology is available to enable a dramatic shift.

Using targeted devices, such as high-resolution satellite imagery, mean, for example, that for insurance purposes, geological- and weather-related data can be examined and re-examined down to square meter.

“The cost of data is coming down very quickly,” emphasises Kurt Cripps, managing director and global head of weather risk at Aon Benfield, a reinsurance broker also interested in taking the concept direct to insurance buyers.

“Parametrics are becoming more attractive to chief financial officers because third party independence is guaranteed, and the pay-out is made within 14 days,” adds Cripps.

The speed by which insurance buyers receive payments could be the most revolutionary aspect of parametrics. Time is saved by reducing transaction costs and loss assessment and adjustment checks, because claims do not need approvals anymore. The data decide instantaneously, without human meddling.

That also saves money and makes parametric offers quite competitive on the market.

Still, the price of parametric insurance differs widely. “It depends on the trigger level and how likely it is a specific event will take place. Parametric insurance against earthquakes in Japan will not be cheap because they are so common there,” argues Schuh from Swiss Re.

The price – like any form of insurance – is based on frequency and severity of the events against which businesses want to protect themselves. Good analytics are therefore vital.

“Analytics is the most expensive aspect of the insurance policy. We need to make sure there is a strong correlation between the data we use and the potential loss,” adds Touffut from AXA Global Parametrics.

However, parametric products also have a significant disadvantage when used solely for weather risk.

“The pay-out will only be activated if it hits a certain threshold, whereas a traditional product could have covered you even below that threshold,” points out Schuh. “This is the basis risk with parametrics and might result in a mismatch between economic loss and insurance.

“It could be advisable to have part of the loss triggered by a threshold that links back to an index and another part by a traditional product,” he adds.

Examples for expansion

Given that most if not all businesses worldwide incur some costs due to meteorological events, weather risk has been the focus for the expansion of parametric products.

“It has motivated the set-up of measurement devices where they were previously absent. The extension of networks can be encouraged in this way,” says Moss at RMS.

The sectors most affected by weather risk globally are the agriculture, energy, construction and transport industries.

Still, the focus is “less on the industry segments but about the perils that are covered”, explains Schuh.

Is parametric insurance therefore also being used for perils not related to the weather?

The answer is yes. Flight delay programmes have been developed by several insurance companies to reimburse clients on the spot if they have incurred a travel delay. Delays can but don’t have to be caused by weather events to trigger a pay-out.

Weather risk is the dominant focus because it is relatively easy to measure through parametric devices, compared to more nebulous perils like cyber or political risks. Data are plentiful, available in real-time for all manner of triggers, weather and geographies, and for underwriters’ purposes are backed up by a century of historical data.

Initially, parametric insurance was pitched for low-frequency high-intensity losses, such as major natural catastrophes. While this focus remains, it is also turning towards direction of lower-key events and microfinance.

A recent example is the pay-out of parametric insurance to businesses in the UK and Ireland after the recent “Beast from the East” winter storm hit those islands. “Water pipelines of insured clients burst, causing the claim to be triggered by the predefined temperature gap in the pipes,” clarifies Touffut.

Another talked-about example, StrategicRISK understands, involves the use of parametric insurance to protect against expected supply and demand risk for a new product release, for instance in consumer electronics and telecoms, covering a business that could face a shortfall or a surplus – provided enough data and the right pricing exist for the product in question.

Broader application of parametric products could make it affordable for a wider range of businesses across different parts of the world – some of which face a greater protection gap and have been less well-served by traditional covers.

“The expansion of the market is clear,” argues Cripps at Aon Benfield.

“We are diversifying what we are currently doing. Will we ever get to the point where we can sell this to the man on the street? Yes, but we are years away from that. It will happen eventually,” he says.

So, how does the geographical diversification of these products look like?

Parametric insurance is at an advanced stage in countries like the US, Australia and the UK. India, Mexico and China represent a market that is becoming ever more sophisticated, although their insurance sectors are less developed compared to other players.

Emerging markets in Africa are also making use of this technology because it might be the only way to insure small and medium-sized enterprises (SMEs) as traditional products might be too expensive or non-existent.

The way microfinance and SMEs are making their way into parametrics – in some cases pioneered first in developed markets, in others linked to emerging market concepts such as microinsurance – can lead to interesting dynamics.

“Insurance covers are sometimes bundled to credit, and banks are used as an intermediary,” explains Touffut. “A typhoon cover could be linked to an SME loan at a local bank. This is a good option in emerging markets, because it basically protects both the bank and enhances the company’s access to credit.”

Climate change and blockchain

Catastrophe bonds have become a popular asset class among investors seeking higher returns, over a fixed period, typically a few years. And what better diversifier than the weather?

Such bonds emerged on the market as a risk transfer alternative to traditional reinsurance cover. After Hurricane Katrina in 2005, insurance companies increasingly sought such alternative risk transfer for their peak catastrophe risks, as a supplement to traditional reinsurance. Parametrics has added another dimension to catastrophe bonds, creating hybrid products.

“Many catastrophe products are already parametric. They are not triggered by an indemnity but by an event. The capital markets need a believable mechanism that determines whether something should pay out or not,” says Schuh from Swiss Re.

Moss at RMS emphasises the crossover, with climate risk creating additional urgency. “There is a strong intersection between those two worlds,” he says. “Climate related issues are higher up on people’s agenda and have become a priority, especially for those countries that are already impacted by it the most.”

An estimate from Swiss Re put the natural catastrophe protection gap at $153bn globally for 2016. Despite innovative products, the truth is that most extreme weather events remain uninsured or at best under-insured. Parametrics could offer a way forward to close the protection gap.

Climate change is adding risk but also building momentum to find solutions globally. “There is large consensus that climate change will increase volatility of weather anomalies worldwide, although the impact will be quite different from one area to another,” adds Touffut.

Parametric devices used to measure risk can go from simple boxes at weather stations, to highly sophisticated data products. The endless volume and myriad sources of data mean opportunities that might blend client and third-party data to capture risks more precisely.

“It would improve the personalised needs and risk profile of the business involved,” thinks Touffut.

Another emerging technology could link nicely with the automatic nature of parametric triggers: blockchain. Could the rise of blockchain, using distributed ledger technology for so-called smart contracts, influence parametrics’ development?

“Instant pay-outs are the only way in which this can make sense for micropayments in the future,” says Schuh. This derives from the fact that if microinsurance policies rely on low premium levels, fewer manual processes mean less efficiency, meaning lower cost.

Touffut agrees: “In the retail segment blockchain and parametric are intertwined,” he says.

The abandonment of onerous manual processes brings obvious efficiencies, but – with at most a backseat role for human involvement – how can the parties involved safeguard against the fraud risk of data being somehow manipulated?

“When you claim the pay-out that is when the risk of data manipulation takes place,” says Schuh.

Exclusive use of trustworthy data sources can reduce this risk. That means due diligence work before the point of inception. Reputable data sources can be agreed upon by all parties involved. But sources also agree it will continue to be a prominent discussion point for the future.

Regulatory differences pose further questions for parametrics’ cross-border expansion. But these are not necessarily new barriers for insurers to content with.

“There are differences in regulation across jurisdictions and sometimes data cannot travel across borders, or maybe only on an anonymous basis,” points out Schuh. “This is a massive issue that has always been there in the insurance market.”

Cripps thinks the ground is fertile for expansion: “Parametric is actually the simplest product that we sell. It will become more and more popular – no doubts about that,” he says.

By Claudia De Meulemeester