Parametric insurance is an increasingly popular solution for weather-related risks. It insures a policyholder against the occurrence of an event, not on the extent of physical damage, making it a less costly alternative for risk transfer. But lack of data has made it difficult to apply the parametric method to the most common and most expensive natural disaster in the US – flooding.
Nearly every single US county has experienced a flooding event. Despite their ubiquity, floods are exceptionally difficult to track, making parametric insurance for floods hugely challenging.
The conventional way to track floods is using stream or tidal gauges to measure water levels. But Peter Lacovara (pictured), a parametric insurance expert, said this method leaves much to be desired.
“The issue with tracking floods using a stream gauge is that it tells you the depth of the water in one specific location. But if you don’t have a gauge somewhere else, then you don’t know where the water is,” said Lacovara, who is also the head of commercial at Cloud to Street (C2S), a climate adaptation technology company that provides precise, near real-time intelligence on flooding.
“In the US, where we have more stream, river, and tidal gauges than probably any other country, they’re still very, very sparsely placed. In many cases, they’re miles apart,” he illustrated.
“Floods can happen in areas where you may not have a stream gauge. You may also have one or two stream gauges [in an area], but the flood covers potentially hundreds of square miles.”
C2S’ technology helps insurers better understand and underwrite flood risk and even monitor flood events as they happen. The New York-based tech firm leverages flood data from a combination of channels, including satellites, historical flood maps, and on-the-ground intelligence. It harnesses machine learning to interpret the data and give insights into the extent and impact of floods.
Lacovara explained how C2S differs from traditional flood modeling: “Conventional flood modeling is done using a stochastic method: you build a model that says water flows to the lowest point. When you want to determine what a flood looks like, you take a digital elevation model and push a bunch of simulated water through it, giving you an estimate of flooding. But if you’re not accounting for events like levees breaking or unexpected drainage, then the model will be totally wrong.
“Cloud to Street can go back in time through the records and show you exactly what that flood looked like. When we look at the flood history, we’re looking at the actual footprint of floods and where the water went, not where a model tells us the water will go.”
Better flood data can give insurers the ability to price risk and to understand the likelihood of a flood event. “When you have a catalog of tens of hundreds or hundreds of thousands of floods – real flood footprints – that gives us a rigorous way of pricing the risk,” Lacovara said.
A parametric product also needs an index value to determine whether an event occurred. Underwriters need to know if the flood took place and to what extent it impacted an area, which is something real-time satellite data from C2S can provide.
Helping Colombia’s farmers
C2S has partnered with parametric insurance platform Raincoat and global reinsurer Munich Re to roll out the world’s first at-scale parametric flood insurance program in Colombia. The national program will make coverage available to more than 100,000 smallholder farmers and marks the first parametric flood program created for climate adaptation.
C2S’ data helped to set the specific parameters for flood risk to create fair, comprehensive parametric policies. Insured farmers under the Colombia program can receive financial relief within days of a flood disaster without needing to file a claim because parametric insurance doesn’t require an in-person inspection of damages to trigger a payout.
For Lacovara, this makes parametric insurance an ideal solution for coverage gaps in developing countries. “Conventional flood insurance needs a lot of underwriting data. To price flood risk, you need to know what the home is built out of, how likely flooding is, and how likely a house is to be damaged.
“In the developing world, getting basic underwriting data can be very challenging. With parametric insurance, we don’t care about the underlying asset, whether it’s a house or a huge commercial factory. Parametric is only concerned about the likelihood of the event itself,” he explained.
The other challenge with delivering indemnity insurance is the cost and difficulty of adjusting claims after an event. In rural areas or developing nations that experience flooding, it could take days or weeks for a claims adjuster to inspect the physical damage and assess the repair cost. In the US, millions of dollars of outstanding insurance claims from Hurricane Katrina still haven’t been fully adjusted 17 years after the disaster, according to Lacovara.
“Many of those claims are from floods caused by broken levees in the New Orleans area. If the capability of satellite flood monitoring had existed, insurers could have settled those claims within two to four weeks,” he added.
Plugging gaps in US coverage
Thanks to tools like Cloud to Street, parametric insurance is an increasingly viable option for plugging flood coverage gaps in the US. “We have more insurance in the US than anywhere in the world, but we’re still massively under-insuring flood,” Lacovara told Insurance Business.
According to US consulting firm Milliman, just four percent of US homeowners have flood insurance. The policies are primarily provided by the National Flood Insurance Program, which operates under the Federal Emergency Management Agency (FEMA).
The gap persists because flood risk remains expensive and difficult to price for carriers, said Lacovara. For people in high-risk areas or who have already been hit by catastrophic flooding, traditional indemnity insurance can only offer cover at exorbitant premiums or not at all. Technology like Cloud to Street can improve underwriting data for conventional insurers while also expanding the viability of parametric insurance for all kinds of climate-related perils.
“C2S can support further understanding of where conventional flood zones defined by FEMA coincide with actual historical flood zones. There are areas not defined as flood zones and have frequent flooding, and vice versa,” Lacovara noted.
“There’s already a robust marketplace for earthquake and hurricane [parametric insurance], and a growing marketplace for convective storm and hail perils. Flood is the last bastion because of the data problem that we talked about. C2S can provide the entire ecosystem of information needed for insurers to expand flood insurance in this country.”