Asia Pacific insurers urged to address data gaps to improve nat cat risk pricing

This third part of (Re)in Asia’s ongoing data series explores the need for better data to price Nat Cat lines.

In multiple markets across Asia Pacific, insurers and reinsurers face ongoing challenges accessing detailed and granular data to inform how to price risk. At times, the problem is that the data providers need may not exist. At other times, the challenge is putting systems in place to access more and better data.

“What I find is, oftentimes, the data is not completely mined. And oftentimes, the data is not well preserved,” says Jay Guin, Executive Vice President and Chief Research Officer at Verisk, a data analytics and risk modeler. “We hear this adage ‘data is the new gold’. If you believe in that, why is there not more investments in curating and preserving data?”

The positive impact of better data stretches out to almost all insurance lines, from motor to property. As people become wealthier, they buy more expensive cars or better appliances, which also add up to more expensive payouts when events happen.

Using data more effectively can minimise the negative impacts for insurance companies. These dual challenges are particularly visible (and affect) natural catastrophe (NatCat) lines.

“We hear this adage ‘data is the new gold’. If you believe in that, why is there not more investments in curating and preserving data?“

Jay Guin

Executive Vice President and Chief Research Officer at Verisk

This is an area in which companies that set out to get more data often have to settle for outdated datasets. Even the commonly used Intergovernmental Panel on Climate Change (IPCC), which came out in 2021, with data from 2019, is not sufficient. The IPCC has a comprehensive data portal but the data in it is not necessarily localised or up to date.

“In some cases, global or even regional climate models can be wildly inaccurate, because the local data is either absent or was not quality checked,” says Benjamin Horton, a Professor in Earth Science at the Asian School of the Environment in Nanyang Technological University (NTU) and the Director of the Earth Observatory of Singapore. “My worries are that a variety of insurance companies don’t use the correct models.”

“In some cases, global or even regional climate models can be wildly inaccurate, because the local data is either absent or was not quality checked.”

Benjamin Horton

Professor at NTU and Director of the Earth Observatory of Singapore

The industry penchant for tapping into international data repositories can be problematic because “certain countries will not give their data, or certain countries just don’t have the money to basically standardise the data.”

In November 2023, Horton raised the issue of data at an insurance industry summit, essentially begging executives to use data better and offering up his institute and himself as a resource.

“Afterwards, my business cards sold like hotcakes. ‘We need to talk. We need to talk. We need to talk to anyone.’ And I would say, basically: ‘If you want better information, come to us. If we can’t provide it, I can find someone who can. Think about yourself and protect your clients’,” Horton says. “Do you know how many people have gotten in touch with me?”

The answer is none.

Data variability

“Risk insights are increasingly becoming important in better understanding risks,” said Vineet Kumar, who heads up the Cat Perils department in Asia Pacific at Swiss Re, pointing out that the company taps into 150 years of knowledge and curated data that is regularly examined by some 50 full-time scientists and 20 data analysts, all of whom tap into proprietary models.

“Data is key to developing products, pricing risks and giving confidence to the market,” said Kumar. “This will eventually lead to a flourishing insurance market with better and affordable products, which consequently could help narrow the protection gap.”

But even for Swiss Re, the quality of data available across Asia Pacific is not uniform.

“Particularly for the lines of business affected by climate change and nat cats, the Australia market probably stands out in terms of data availability, while such data in Southeast Asia, India and China is generally aggregated in nature with none to a few risk attributes and have plenty of room for improvement,” said Kumar.

“Particularly for the lines of business affected by climate change and nat cats… data in Southeast Asia, India and China is generally aggregated in nature with none to a few risk attributes and have plenty of room for improvement.”

Vineet Kumar

Head of APAC Cat Perils at Swiss Re

For the insurance industry, understanding the impact of climate changes such as those related to sea levels, may be incredibly important because, even if greenhouse gas emissions start dropping today and the world manages to pull back from the 1.5 degrees C warming threshold, climate change such as sea level rises will continue and coastal properties will be at greater risk.

“It doesn’t matter whether we’re on a high emission or a low emission track. Sea level rise is scenario independent for the next few decades, because the amount of warming of the ocean or melting the ice sheets depends on cumulative emissions,” Horton says.

Understanding, as much as possible, the local impact of those changes is even more important, as it translates the theoretical importance of better data into practical realities of protection. It is difficult to underestimate the importance of making decisions based on effective data models.

Horton recounts a story from his days working at the University of Pennsylvania. One alumnus had a multi-million-dollar house in the Hamptons, a coastal area in New York, and Horton was asked to do an environmental risk study. His team surveyed the property, ran some models of hurricane impacts and passed the information along. After seeing the results, the alumni sold the house (making a nice profit). Then, in 2012, Hurricane Sandy smashed the house.

The moral is that data is key to making smart business decisions.

“There are a lot of things we don’t know about climate change. It is very hard to predict where or when the next disaster is going to occur. But there’s also a lot of certainty,” Horton says.

Climate change is occurring earlier and more rapidly than expected.

New models

Knowing that floods or typhoons will happen is not the same as knowing when and where they will happen or how much damage they will cause. Accurately assessing risks and potential costs requires granular, location-specific data that is not always available, says Verisk’s Guin.

In the U.S., Verisk can access location data to the postal code level and correlate it with, for example, the local price of lumber or tie it to local inflation patterns, which can be different than national level inflation. In many places in Asia, that type of granularity is almost impossible to access.

“A lot is lacking in Asia in terms of the quality of the data,” says Guin.

A case in point is property data, the granularity of which can vary enormously. Guin says there is some basic information that Verisk would love to have across the board, including location supported by GPS information from phones as well as the characteristics of specific properties, building materials, the type of construction and value. Property use is also important information.

“That sounds basic, but it is not easy to put it all together,” says Guin, although he is quick to point out that there has been progress.

“All markets are not the same. Asia is huge. Japan is very different from India. And if you extend it to Oceania, Australia and New Zealand are different from Indonesia. There is quite a bit of variation,” Guin says. “We see improvements, but not at the pace we would ideally like.”

There are some things that insurance providers can do to tackle this challenge. One approach starts at the point of sale and make efforts to keep the granularity of data, says Guin.

“Are insurance (companies) capturing enough of the information about what they are insuring? And even if they are, do they have the internal IT systems to preserve that data and make it flow through all their other business processes?”

“What we see often is some of the granularity is lost even if it’s captured at the point of sale,” says Guin. “The portfolio data that we sometimes see that are used for inputting into the models, they are often aggregated,” says Guin.

Technology is not the issue.

“There’s no technological advantage in any region,” Guin says. The problem may be more cultural.

“Some of it is prioritisation of investments and the recognition that the data is the most valuable part,” says Guin.

“While I am seeing improvements in the collection of exposure information such as detailed latitudes and longitudes, there still are significant gaps when it comes to, say, terrain, climate hazard and claims data.”

Alexander Pui

Adjunct Fellow at UNSW, and Senior VP for Climate and Sustainability Advisory at Marsh

With all that said, there is progress, says Alexander Pui, Adjunct Fellow at the Climate Change Research Center, UNSW, and Senior Vice President for Climate and Sustainability Advisory at Marsh. He points to initiatives throughout the region that link up both industry and academia, including Singapore’s NTU and Tokyo University’s work around floods.

When all is said and done, he says, the use of data by the insurance industry in Asia is “currently still lacking but moving in the right direction”.

“While I am seeing improvements in the collection of exposure information such as detailed latitudes and longitudes, there still are significant gaps when it comes to, say, terrain, climate hazard and claims data, which is critical for helping to develop views of risks through loss experience ratings and catastrophe models,” says Pui.

With regulators in some jurisdictions like New Zealand, Australia, Hong Kong or Singapore increasingly focused on how institutions price future risks and the implications on business and society.

“This compels increasing transparency over the pricing process, identification of gaps in data and efforts taken to bridge these gaps,” says Pui.

And the ultimate result, he says, may be forward progress in terms of the industry’s use of data, which, he says, is “lacking, but moving in the right direction.”