April 26, 2021
Beware the Grey Swan
by Paul Carroll
We as an industry need to learn to prepare better for grey swans -- rare but very possible events -- both for our own sakes and for our clients.
As the U.S. steadily emerges from the pandemic — and we all hold out hope for India — the temptation is to dismiss it as a one-off, a once-in-a-century health disaster, a black swan. But the pandemic is actually what’s coming to be known as a grey swan — something that, while rare, relates to a known problem and that can be planned for, if we come to grips with the cognitive biases that blur our ability to see them.
As we’ve learned the hard way over the past 20 years, there are a lot of grew swans out there, so we as an industry need to learn to prepare better for them, both for our own sakes and for those of our many clients.
This report from Aon on dealing with grey swans’ effect on corporate reputations includes a daunting list of those that have been broadly ignored over the past two decades, starting with the 9/11 terrorist attacks — which somehow caught the world by surprise even though Islamic terrorists were known to want to strike in the U.S., even though plots had been uncovered to hijack and crash planes into high-profile targets or blow them up and even though terrorists had attacked the World Trade Center itself and tried to make it collapse eight years earlier (right across the street from my office at the time).
The dangers that a hurricane posed to the levees in New Orleans were well-known long before Hurricane Katrina devastated them. So were the perils of subprime mortgages — a member of the Fed’s board of governors saw the crisis coming so far ahead of time that he published an alarmist book in 2007 yet was largely ignored until after the Great Recession of 2008-9 began. The tsunami that caused the disaster at the Fukushima nuclear plant in 2011 was eminently foreseeable. And, of course, many had been predicting a pandemic for years before COVID-19 pretty much shut the world down starting last spring and killed millions — Bill Gates even got most of the particulars of COVID right in a dire TED talk in 2015.
Why do we keep missing these grey swans?
Drawing on the seminal work of behavioral economist Daniel Kahneman, the Aon report lists six cognitive biases that cloud our judgment on risks more complicated than “white swans” — which are common enough that we have clear data on them and routinely incorporate them in our risk management.
The biases are: the ambiguity effect (our minds don’t like options with unknown probabilities); normalcy bias (we underestimate the likelihood and severity of disasters); optimism bias (we underestimate the probability of being affected directly); the ostrich effect (we ignore negative information to avoid the anxiety that comes with decision-making); herd instinct (we align with the behavior of a group to avoid conflict); and status quo bias (we prefer to keep doing what we’re doing).
As the report explains, the ambiguity effect, normalcy bias and optimism bias “relate to our limitations as natural statisticians. We gravitate toward information that we can process and organize [while avoiding]… uncertain, ambiguous data…. To help us navigate through the storms of life, we tend to be optimistic about our chances. Despite knowing the health risks associated with smoking or obesity, for example, we believe that ‘it won’t happen to me,’ yet we buy lottery tickets equally believing that, ‘it might be me!'”
The ostrich effect, herd instinct and status quo bias “relate to managing our emotional state. Evidence that conflicts with our rosy view of the world is uncomfortable and unpleasant…. It is easier to go along with the majority than stand one’s ground and cause waves.”
So, how can we do better?
The report’s conclusion: “Effective risk management strategies will acknowledge these flaws openly and institute measures to combat their most harmful effects.”
It suggests considering, in particular, the possibility of “a large-scale cyber attack with physical consequences. Cyber physical risk is not new, but its threat is growing rapidly, as adoption of the Internet of Things (IoT) accelerates and increases the ‘attack surface’: the number of connected systems and devices through which an attacker can enter or extract data.”
That kind of attack certainly seems plausible — as the SolarWinds attack by Russia showed, nation-states have the ability to sabotage each others’ infrastructure, such as electric grids, pretty much whenever they want.
The report adds: “We could turn our minds also to the ‘green swan,’ the term coined by the Bank for International Settlements to describe black swan events related to climate change.” It’s always hard to trace a disaster to climate change — some will say the recent Texas freeze may stem from climate change’s tendency to cause more extreme weather; some won’t — but the likelihood of green swans is certainly increasing.
One caution: I think the evaluation of disasters after the fact is often Monday morning quarterbacking: “It’s obvious that we should have punted/shouldn’t have punted,” “should have passed/should have run,” “should have seen that that player would be a star/a bust,” etc. You can’t just prepare for one grey swan and hope that’s the one you should have headed off. You have to prepare for all the grey swans you can imagine — you don’t just fix the levees in New Orleans; you prepare for what hurricanes might do up and down the Gulf Coast.
That can be expensive. So, there has to be some real calculation involved based on the odds of an event, the likely cost of a disaster and the expense associated with avoiding all such problems — and the nature of grey swans is that none of these figures are easily quantified.
All we really know for sure is that grey swans are occurring faster than we’ve expected and have been far costlier — COVID-19 has cost trillions of dollars in the U.S. alone, and the devastation in terms of lives lost has been even greater. So, we’d do well to confront our biases and keep trying to make ever-more-realistic evaluations of the risks we’re facing.
P.S. Here are the six articles I’d like to highlight from the past week:
The industry is probably looking at several more years of accident year combined ratios above 100%.
Customers, risks, operations and the workforce all have been transformed over the last year. This makes strategic planning a challenge.
The cadence and granularity of data about travel behavior need to be enhanced. Geospatial analytics can be the engine.
Environmental, social and governance (ESG) is a chance for the insurance industry to do well by doing good.
As trends point to rapid adoption of alternative payment methods, insurers must determine how to meet B2B needs.
Employers and workers’ comp carriers must tread lightly; accepting a COVID claim can have a big impact, beyond the initial care and recovery.