Sanjiv Singh, Head, Marine & Specialty lines, GI Council

Extending parametric insurance to consumers provides insurers with an immediate, data-triggered option for claims payments.

The term ‘parametric insurance’ describes a type of insurance contract that insures a policyholder against the occurrence of a specific event by paying a set amount based on the magnitude of the event, as opposed to the magnitude of the losses in a traditional indemnity policy.”

Simply put, parametric insurance provides fast, lump-sum payments upon the occurrence of a pre-determined data metric (a parameter). By expediting the claims process policyholder’s insurance creates an opportunity to improve the policyholder’s experience, building loyalty and trust.

Parametric is not new. It has been used for almost 25 years in large-scale transactions such as catastrophe bonds in both the private and public sector.

What is new, however, is the extension of parametric to the consumer level. Only now is data sufficiently available and technology adequately mature to enable the immediate, data-triggered payments of parametric insurance.

Parametric for consumers

When used as a consumer insurance product, parametric must be structured so that it meets the definition of insurance.

The most common hurdle in classifying parametric as insurance is its inherent basis risk: the difference — potentially quite large — between the insured’s losses and the payment amount received. With disbursement amounts fixed in advance, the insured necessarily experiences either a windfall (payment exceeds the loss) or a shortfall (losses exceed payment), never a precise reimbursement.

Shortfall occurs routinely in all insurance. Although windfall is known to occur occasionally, it is not the intention of any insurance policy (whether parametric or not). Thus, the most defensible parametric policy is one that minimizes the chance of windfall.

Precedents to parametric

Many existing insurance products already incorporate the attributes of parametric: speed, transparent settlement criteria and fixed-dollar-amounts, even if they are not called parametric per se.

Here are a few examples of “parametric-like” insurance for consumers:

• Accident insurance (AD&D): Fixed-amount payment triggered by a specific accident.

Trigger data is available and verifiable from third-party medical records. Loss and payment are correlated because payment amounts are derived from well-established data on medical losses.

• Travel insurance: Specific compensation triggered by cancellations or delays.

Trigger data is based on publicly available, real-time flight databases and pricing. Losses and payments can be tightly correlated to the cost of alternative plans.

Advantages of parametric

Parametric is not a substitute or replacement for conventional indemnity insurance. For consumers, the primary advantages of parametric are:

• Speed of payout

Speedy payouts prevent policyholders from having to tap savings or credit to pay their losses To the insurance company, the deployment of immediate funds can prevent follow-on losses.

• Sense of certainty

The customer knows the precise amount to be received and under what circumstances. In contrast, indemnity settlements are inherently uncertain. Research consistently shows that human brains do not process uncertainty rationally. The greater the uncertainty, the more distrustful and adversarial customers behave. Parametric alleviates distrust by building in certainty to the outcome.

• Transparency

This is an element of trust. When trigger data is equally available to both the insurer and the policyholder, it reduces the perception of unfairness.

From the perspective of the insurer, the primary advantages of parametric are:

• Customer retention

As described earlier, parametric policies present an opportunity for insurers to surprise and delight customers, thereby building trust and loyalty with fewer cancellations.

• Capped liabilities

With parametric policies, insurers know the total amount of liability with a high degree of certainty. This is particularly material for high-severity, low-frequency events where uncertainties are high.

• Simplified underwriting

Fixed and capped payouts simplify underwriting. Individual risks can be priced and selected solely on the probability of the trigger. This accelerates sales by reducing the amount of data and time required to quote and bind a policy.

Risks suited to parametric

When determining which risks are suited to parametric, Lloyd’s states: “Any risk with a loss outcome can be a candidate for parametric insurance, provided that the data is reliable and third-party verifiable and that a statistically significant correlation between the loss event and the insurance payment is established.”

Data reliability is key to this definition. For consumer applications, the most credible data to use as a parameter is that from a transparent, indisputable source. Real-time availability and API connectivity also facilitate automation. Good sources for parameter data include:

• Public agencies
• Sensors, whether specific to the risk insured (e.g., auto sensors) or a nearby proxy
• Third-party open data

The influence of frequency

For frequent, data-rich events, parametric policies should be designed so that losses and payments are well correlated. The illustration earlier on travel cancellations is one such example. Some types of auto accidents might also be relevant.

On the other hand, rare events such as natural disasters are data-poor, but it is still possible to achieve parity between losses and payments, and thus minimize basis risk. Methods of doing so depend on whether a policy is “pure” parametric or partly parametric. These differences are highlighted below.

Parametric doesn’t eliminate the need for claims

Consider a “spectrum” of insurance products, from indemnity-only on one extreme, to “pure” parametric on the other. “Pure” in this case means that the payment is based solely on data with no other claims adjustment.

Many product possibilities combine some principles and advantages of indemnity insurance with those of parametric insurance. Imagine these examples:

1. Two-phase

This idea for a policy would pay the first portion of losses parametrically upon exceedance of a specified trigger, and the remainder through indemnification.

1. Multi-tranche

This idea for a policy would pay parametrically in multiple tranches: the first payment occurs without documentation or adjustment of loss, but payments in follow-on tranches require increasing degrees of loss itemization and claims adjustment.

1. Adjuster-triggered

This idea for a policy would pay parametrically upon being “triggered” by the adjuster. The adjuster ascertains whether the payment threshold has been met. This could be the case, for example, for a policy triggered by an on-site sensor that is accessible only by the adjuster to minimize the potential for tampering.

In these examples, the claims professional remains essential, but the adjusting process is less stressful: it’s less time-sensitive, and the policyholder has already had the positive experience of receiving an initial payment.

The future of parametric

Parametric has enormous potential to help close the protection gap — to begin covering severe, low-frequency risks that until now have gone largely unprotected — not just weather-related disasters, but cyber-terrorism, political unrest, pandemic, warfare, global economic downturn and more.

Meanwhile, the explosion of data from artificial intelligence and machine learning is multiplying the types of risks that can be covered by parametric insurance.

Imagine, for example, that a parametric layer of insurance is routinely bundled with indemnity policies so that customers receive immediate (even if only partial) payouts the same day as a severe storm, an auto accident or a widespread power outage. The magnitude of the payout need not be large.

When a customer experiences frequent positive outcomes from their insurer, they are more satisfied with the insurer as well as the claims professionals who represent it.

Acknowledgements:

(https://www.linkedin.com/pulse/parametric-moment-building-insurance-age-uncertainty-kate-stillwell)