Currently showing: Climate/natural disasters > Climate change

21 Dec 15 17:13

In the wake of a natural disaster, be it a hurricane, earthquake, tornado or flood, we take the necessary steps to harden our defenses and ensure the impacts of another event aren't felt as strongly. There are countless examples of this in recent years. 

After Hurricane Katrina, the levees systems in New Orleans were expanded and improved.  After the 2011 Tohoku earthquake and tsunami, improved flood defenses were built around the Fukushima power plant. After Hurricane Sandy, buildings in parts of the Northeast were raised.  All these responses, plus the countless others put in place in the wake of natural disasters, are implemented to protect communities and populations, and reduce our vulnerability to extreme events.  But while at the American Geophysical Union Fall Meeting in San Francisco, I listened to a talk which framed protective measures in a different manner: Do protective measures actually increase risk by lulling the population into a false sense of security, causing them to disregard their residual risk? 

The argument presented was compelling; the authors focused on the notion that humans are, by definition, going to respond to disasters, and that the responses in general fall into two categories. If the disasters are frequent, societies will often figure out mechanisms for adaptation, thereby decreasing their vulnerability. If events are infrequent, populations will forget about the most recent disaster within a few years, and fall back into the same routines. Unintended complacency, if you will. 

One method of adapting to frequent extreme events is, of course, to put in protection measures which will decrease the frequency of those events.  The authors presented the two scenarios, which involved two communities exposed to the same hazard (e.g. riverine flooding). After the first flood, one community takes a softer approach to adaptation, implementing behavioral changes, such as relocating lifestock, to mitigate the damages from subsequent floods. The other community takes a harder approach, putting in levee systems and barriers to prevent another analogous river flood from occurring again. The logical conclusion drawn from these two approaches is that the community that has implemented the grey infrastructure (levees and barriers) is at a lower risk from floods. However, the presenters found that while the flood losses in the hardened community are less frequent, when they do occur, they are far more severe, since too much time lapses between events, and the population "forgets" about the previous event. 

To me, this doesn't negate the case for protective measures; it would be ineffective and inefficient to only change behavior as a response to extreme events. For example, no one wants to constantly have to move their belongings to higher ground because of frequent flooding.  Rather it re-enforces the concept that a holistic approach to risk management is a necessity, and that education is key.  Protective measures that are put in place will only be ineffective if viewed as a panacea, and not one tool in a very large tool box.  It's critical that everyone, from individual homeowners to sovereign governments, understand that hardened infrastructure will not protect against every event.  Savings gained should be put towards financing and recovering from the more extreme events; insurance and reserve funds are just two ways the additional funds can be invested.

In the future, we could even see financial products that merge the installation of protective measures with insurance. Swiss Re, RMS, the Rockefeller Foundation, through the re:bound initiative are working towards the creation of resilience bonds. While the framework is still being developed, conceptually, a resilience bond would provide protection against financial damage if a disaster strikes while the installation of the protective measure, like a seawall, is underway. Once the protection measure is complete and the risk has been reduced, the structure would issue rebates to reflect the increased protection that for example, the seawall provides. 

Protective measures are truly only resiliency measures if the communities who benefit from them acknowledge that their benefit only goes so far.  It's not about planning for the "bad," you have to plan for the "worst."

Category: Climate/natural disasters: Climate change, Disaster risk, Resilience

Location: San Francisco, CA, United States


Fabrizio Carenzi - 23 Dec 2015, 4:21 p.m.

Very interesting, Megan!
"Do protective measures actually increase risk by lulling the population into a false sense of security, causing them to disregard their residual risk?"
Apparently.. yes, they do.
At least, sometimes.
It is what is called "risk homeostasis", a cacophonous word that describes the behavioral pattern that people exhibit when facing risk: "people typically adjust their behavior in response to the perceived level of risk, becoming more careful where they sense greater risk and less careful if they feel more protected".
The keyword is people "feel more protected", not actually "are" more protected; " it has been observed that motorists drove faster when wearing seatbelts and closer to the vehicle in front when the vehicles were fitted with anti-lock brakes".
So, if we have to fight the very nature of humans... it seems clear that a holistic approach is unavoidable

Honza Kerver - 10 Jan 2016, 8:51 p.m.

Hi Megan, would you see a role for re/insurers in pushing clients to use mitigation strategies before they underwrite a risk at all?

Megan Linkin - 11 Jan 2016, 2:15 p.m.

Hi Honza, thanks for the question. There's plenty of available capital in the market, and underwriters have the tools and skills to assess the risk posed by a portfolio or individual asset, so I don't believe that the presence or absence of resiliency measures will be the deciding factor as to whether or not to take a risk. However, the re/insurance industry sees resiliency measures as an important tool to reducing the vulnerability of structures to natural hazards such as quake and hurricane. The impact of resiliency measures can be considered during a risk assessment, and if these measures reduce the severity and frequency of loss, this can feed back into the underlying expected loss and ultimately the premium. Check out our ECA website (, where you can learn how natural catastrophe models, traditionally used for underwriting, are used to help clients assess the cost benefits of different resiliency measures

If you would like to leave a comment, please, log in.