The complex insurance math that tries to calculate the risk of pandemics

I'm typically pretty skeptical of insurance as an industry, though I begrudgingly understand that it does have its purpose. But I found this Wired article on pandemic insurance to be utterly fascinating. Focusing on the work of Nathan Wolfe, who began his career as a virologist before shifting to ways of using data to offset the economic impacts of potential pandemics, largely through reinsurance - that is, insurance purchased by an insurance company from another insurance company in order to protect itself from risks, in case a disastrous event leads to overwhelming payouts. As I learned:

Reinsurance is a staggeringly lucrative endeavor: Munich Re had $56 billion in revenue and $3 billion in profit last year. The market is large enough that its perennial competitor, Swiss Re, took in $49 billion itself.

B2B industries are wild. But what I found most interesting was the way in which Wolfe's business model struggled to quantify the risks of pandemics - how to take the impacts of potential public fears that affect behaviors that could ultimately businesses, and turn them into actual numbers that can be fed into an algorithm:

The model would need to capture something much more difficult to quantify than historical deaths and medical stockpiles: fear. The economic consequences of a scourge, the historical data showed, were as much a result of society's response as they were to the virus itself.

The group started building what became known as the Sentiment Index. Ben Oppenheim, head of the product team and a political scientist, had studied the work of Paul Slovic, a University of Oregon psychology professor who studied how human beings perceive and respond to risk. Inspired by Slovic's data-driven approach, they gathered their own information from around the world on how much various symptoms frightened people. To validate their measures, they also began tracking and studying how media coverage evolved around different types of outbreaks. Scarier diseases tended to generate more news stories.

[...]

The Sentiment Index was built to be, as Oppenheim put it, "a catalog of dread." For any given pathogen, it could spit out a score from 0 to 100 according to how frightening the public would find it. That number could then be used to help calculate the possible financial losses from an epidemic, everything from empty hotels to postponed mining projects. Madhav and her team, along with Wolfe and Oppenheim, also researched the broader economic consequences of disease outbreaks, measured in the "cost per death prevented" incurred by societal interventions. "Measures that decreased person-to-person contact, including social distancing, quarantine, and school closures, had the greatest cost per death prevented, most likely because of the amount of economic disruption caused by those measures," they wrote in a 2018 paper.

I never thought I'd be so interested in insurance - or reinsurance, for that matter - but here we are. This was great.

We Can Protect the Economy From Pandemics. Why Didn't We? [Evan Ratliff / Wired]

Image: Feldsherov Victor / Wikimedia Commons (CC 4.0)