|Vol 1., No. 5|
May 10, 2023
Scott Clearwater, Gridmetrics, Inc.®
As power grids world-wide undergo a rapid transition to renewable energy sources and the effects of climatic extremes become more frequent, power grids are becoming less and less resilient. Resilience in this context refers not only to power outages but also surges and sags away from the voltage reference setpoint that is ultimately delivered to end users. This lower resilience has a knock-on effect in terms of business productivity and quality of personal life. To compensate for these increasing losses, it makes sense to consider how power resilience insurance can be employed. In this article we will explore some simple power resilience insurance products that could be employed today.
The first step in creating an insurance product is having data to create the profitable terms under which the insurance can be sold. Today, data is available for power resilience insurance from Gridmetrics which has nearly 300,000 voltage sensors with about half the US population living within a kilometer of a sensor. Many businesses are also close to these sensors. Gridmetrics has created various power resilience indexes and metrics that could be used to create parametric insurance products and others could be custom-tailored. For example, one of the indexes, was discussed in the previous blog and is the Voltage-at-Risk Shortfall, VaRS (not to be confused with Expected Voltage Shortfall, EVS). VaRS is defined at a percentile level of historical data, for example, VaRS(1%) is the voltage level (from the reference voltage) of the lowest 1% of the data. The greater the value of VaRS the more frequently sags and outages in voltages are observed. As with storm insurance, with power resilience insurance we are interested in the presence of extreme events because these are the ones causing the most damage and therefore most motivated to pursue insurance solutions.
The next step in creating a power resilience insurance product is to identify a region with high Gridmetrics sensor penetration and choose a time frame for the historical data underlying the actuarial data for the insurance term sheet. Term sheets can be constructed using either ranges of the index or the number of times the index value for the region, VaRS, falls within a pre-determined thresholds in a given time period. These are the parameters for a parametric insurance product.
For the term sheet to be profitable it must have a premium greater than the event-probability-weighted payout for the events covered by the contract. For the purposes of illustration, we’ll use some representative numbers as shown in Table 1. For simplicity we will assume that the term sheet is the same for each metro area even though in practice the term sheet would be customized to compensate for the risk in each area. We will also assume the same insurance company covers all the metro areas.
|VaRS Thresholds at 1% Tail||Payout|
|VaRS < 0.012||0%|
|0.012 < VaRS < 0.013||25%|
|0.013 < VaRS < 0.014||50%|
|0.014 < VaRS < 0.016||75%|
|0.016 < VaRS||100%|
Now assume one year contracts for all the metro areas and an interest rate of 4% and a probability of VaRS < 0.012 = 95%. Then the expected actuarial net present value per premium amount equals 5% / 1.04 = 4.81%, which the insurance firm will use in its premium calculation. Any losses amounting to less than 4.81% of the premium will result in a profit for the firm.
We now explore twenty metropolitan areas to see what their payouts would have looked like given the actual calculated VaRS values for each metro area. The results in 2021 look profitable but 2022 looks like a year with excessive payouts and probably those cities would need to have their terms changed for the insurance firm to continue to run a profitable business.
|MSA||VaRS 2021||2021 Payout||VaRS 2022||2022 Payout|
|Average Overall Payout||50% / 20 = 2.50%||175% / 20 = 8.75%|
|Average Overall Premium||4.81%||4.81%|
The point here of this simple hypothetical case is that given the vast amount of power resilience data now available in one place it becomes feasible to construct novel insurance products based on this data and derivatives of it. Who would buy such insurance? Would it be utilities to protect against excessive fines from regulatory agencies because of poor delivered power resilience? Or could it be be businesses looking to protect against sub-optimal power delivery effects on their business? Both cases are possible.
Learn more about Gridmetrics actionable power intelligence solutions.