CENTER FOR INSURANCE
POLICY AND RESEARCH
Extreme
Weather and
Property
Insurance:
Consumer
Views
July
Contents
Overview
Survey Methods
Demographics
Insurance Details
Extreme Weather and Insurance
Extreme Weather Beliefs and Insurance Action
Climate Risk Disclosures
Six Americas Segmentation
Naic CIPR : extreme Weather aNd property iNsuraNce
Executive Summary
Homeowners face significant impacts to their property from escalating
extreme weather amplified by potential climate impacts. As roughly %
of homeowners have homeowners insurance,
1
the rising likelihood of
extreme and catastrophic weather events makes monitoring the frequency
and impact of natural disasters a critical insurance regulatory function. This
makes climate change/natural catastrophe risk and resiliency a key strategic
priority of the National Association of Insurance Commissioners (NAIC).
To help guide NAIC efforts in this domain (e.g., consumer outreach), this
study sought to better understand how homeowners are making the
connection between extreme weather events and their insurance coverage.
Specifically, we looked at three main questions:
) To what extent do respondents see a connection between extreme
weather events and their broader insurance coverage?
) What is the relationship between what respondents believe about
extreme weather and what actions they have taken related to their
homeowners insurance?
) How are homeowners segmented by their climate risk perceptions,
which further depend on loss experience and insurance coverage?
We surveyed , U.S. homeowners from age to over in every
U.S. state and Washington, DC. Respondents were recruited through
SurveyMonkey’s Audience panel. The survey was conducted March –,
.
Overall, we find that:
·
Most homeowners in our survey do, in fact, have homeowners
insurance, though respondents making less than $, a year
were % less likely to have a policy than others with higher incomes.
Affordability issues were the most prevalent factor indicated for not
having coverage.
NAIC Dwelling Fire, Homeowners Owner Occupied, and Homeowners Tenant and
Condominium/Cooperative Unit Owner’s Insurance Report (). Accessed May , at
https://content.naic.org/sites/default/files/publication-hmr-zu-homeowners-report.pdf.
Naic CIPR : extreme Weather aNd property iNsuraNce
·
A majority (%) believe their homeowners policy covers flood, even
though flood is not covered under standard policies and only about
% of homeowners actually have flood insurance.
·
Two-thirds of respondents said their homeowners insurance costs
have gone up over the past three years, with “increase in natural
disasters such as hurricanes and wildfires” indicated as a key reason
believed to be driving the increase.
·
People living in Pacific, West South Central and Middle Atlantic
2
states
are most likely to report trouble getting or renewing homeowners
insurance due to an increase in natural disasters. Across all premium
levels, homeowners paying more for their insurance report
increasingly higher levels of trouble renewing insurance because of
wildfires or hurricanes.
·
More than two-thirds of respondents said they were aware of things
they could do to protect their property from extreme weather events to
reduce their risk, but this awareness does not necessarily translate into
action. Only half had actually made changes, though more than three-
quarters said they would spend their own money to fortify their home in
exchange for a reduction in their homeowners insurance premium.
·
Four out of five homeowners expressed interest in knowing what
insurance companies are doing to address changing weather patterns
and other climate-related risks.
·
Homeowners revealed conflicting attitudes about climate-related
hazard risk. While many respondents indicated reluctance to move
to areas with flooding, hurricanes or wildfires and % think extreme
weather events are happening more frequently overall because of
climate change, on average they believe the area where they live
is slightly less vulnerable to extreme weather events than the U.S.
average.
·
Survey respondents demonstrated a high level of worry related to
climate change, with % of respondents classified as “alarmed,” more
than double the national average.
·
Those having trouble obtaining homeowners insurance consistently
indicate being alarmed about climate change.
This report provides an overview of the survey methodology along with
brief findings from the survey responses.
For more information, please contact Lisa Groshong
at [email protected]g.
Lisa Groshong, Ph.D.
Communication Research Scientist
Tyler Gerson
Research Assistant
Jeffrey Czajkowski, Ph.D.
Director
Juan Zhang, Ph.D.
Postdoctoral Research Associate
Laura Kane
Communications Director
Jennifer Gardner
Data Coordination and Statistical
Analysis Manager
Pacific: Alaska, California, Hawaii, Oregon and Washington; West South Central: Arkansas,
Louisiana, Oklahoma, and Texas; Middle Atlantic: New Jersey, New York and Pennsylvania
CIPR
NAIC
Disclaimer
This study represents the opinions
of the authors and is the product
of professional research. It is not
intended to represent the position or
opinions of the NAIC or its members,
nor is it the official position of any
NAIC staff members. Any errors are
the responsibility of the authors.
Naic CIPR : extreme Weather aNd property iNsuraNce
Survey Methods
The extreme weather and property insurance survey was conducted March
–, , using a nonprobability sample of , U.S. homeowners
recruited through SurveyMonkey’s Audience panel. SurveyMonkey
calculated a margin of sampling error on the total results as +/– percentage
points at a % confidence level.
SurveyMonkey targeted respondents in their panel who previously identified
themselves as homeowners when asked, “Do you rent or own the place
where you live?” SurveyMonkey’s methodology does not distinguish
between stand-alone houses, condos, apartments, or other types of homes.
The survey’s questions explored experiences, knowledge and perceptions
related to property insurance among homeowners. Participants were asked
to answer questions about their home insurance, including their role in
choosing it, the approximate cost, and whether the price has changed.
Participants were asked about their experience with home damage due
to extreme events, whether they have altered or would alter their home to
protect it, and if they believe their insurance costs have changed due to
such events. The survey also queried their interest in insurance companies’
actions to address climate change. Demographic information included age,
gender, race, annual household income, and U.S. Census Bureau region.
Audience segmentation questions were drawn from Global Warming’s Six
Americas Super Short Survey (SASSY!), an audience segmentation tool
designed to cluster respondents according to their climate views.
3
Breanne Chryst, Jennifer Marlon, Sander van der Linden, Anthony Leiserowitz, Edward
Maibach and Connie Roser-Renouf (). Global Warming’s “Six Americas Short Survey”:
Audience Segmentation of Climate Change Views Using a Four Question Instrument,
Environmental Communication, :, –, DOI: ./...
The survey also queried their
interest in insurance companies’
actions to address climate change.
Naic CIPR : extreme Weather aNd property iNsuraNce
Survey Results
Demographics
Participants were targeted to capture a balanced distribution of homeowners
across gender and age according to U.S. Census data (U.S. Census Bureau
). The U.S. Census Bureau estimates that as of , about % of
households owned their own home.
4
Table presents the demographics of participants. Slightly less than one-
half of the respondents were women (%). Women tend to be slightly
overrepresented in surveys, but SurveyMonkey had trouble recruiting
enough eligible older female homeowners to balance this segment of the
sample. We suspect that limiting survey participation to homeowners may
have influenced the gender balance of our sample.
The sample was evenly split between participants under and over the age
of . The age distribution differs from the general U.S. population because
we limited our survey to participants over the age of . Approximately
% of the respondents’ households reported earning less than $,.
This is less than the general U.S. population, possibly because people with
lower incomes are less likely to own their own homes.
About % of the respondents came from the Northeast, Midwest and
West, with about twice that many (.%) from the South. Every U.S. state
and Washington, DC, were represented.
The U.S. Census Bureau estimates
that as of , about % of
households owned their own home.
Source: U.S. Census Bureau. Accessed May , , at https://www.census.gov/newsroom/
press-releases//homeownership.html.
Naic CIPR : extreme Weather aNd property iNsuraNce
Table . Survey Demographics
Respondent Demographics (n = ,) N
% of Survey
Participants
% of U.S.
Population
Gender (n = ,)
Female 1,177 47.4 50.8
5
Male 1,304 52.6 49.2
Age (n = ,)
18–29 291 11.7 13.6+
6
30
–44 947 38.2 19.5
45
–60 662 26.7 18.9
>60 581 23.4 22.8
Household Income (n = ,)
$0–$49,999 594 23.9 38.4
7
$50,000
–$74,999 363 14.6 17.4
$75,000
–$99,999 383 15.4 12.8
$100,000+ 949 38.3 31.4
Prefer Not to Answer 192 7.7
U.S. Census Region (n = ,)
Northeast (includes Middle Atlantic
and New England states) 514 20.8 17.1
8
Midwest (includes East and
West North Central states) 522 21.1 20.8
South (includes East and West South
Central and South Atlantic states) 929 37.6 38.3
West (includes Mountain and
Pacific states) 506 20.5 23.9
Race (n = ,)
9
White 2,163 87.0 72.0
Black 167 6.7 12.8
Asian 103 4.1 5.7
American Indian or Alaska Native 41 1.7 0.9
Native Hawaiian or Other
Pacific Islander 9 0.4 0.2
Other 59 2.4 5.0
10
Source: U.S. Census Bureau.
Accessed May , , at
https://data.census.gov/
cedsci/table?q=United%
States&tid=ACSSTY.S.
U.S. Census groups ages –, so
this table includes people aged
– but excludes those –.
Source: U.S. Census Bureau.
Accessed May , , at
https://data.census.gov/
cedsci/table?q=household%
income&tid=ACSSTY.S.
Source: U.S. Census Bureau.
Accessed May , , at https://
www.census.gov/popclock/print.
php?component=growth&image=//
www.census.gov/popclock/share/
images/growth_.png.
Total exceeds % because
respondents could choose more
than one option.
In addition, .% of the U.S.
population identifies with two or
more races.
Naic CIPR : extreme Weather aNd property iNsuraNce
Insurance Details
In addition to basic demographics, we examined whether consumers have
homeowners insurance and how much they currently pay for that insurance.
We also asked a knowledge question that investigated whether consumers
understood that their homeowners insurance policy most likely did not
include flood coverage.
Homeowners Insurance
The vast majority of respondents (%) reported having a homeowners
insurance policy, though this varied by income. Only % of respondents
with household incomes of less than $, reported having a policy,
as opposed to % of respondents with household incomes of $,
or more. Geographic differences also emerged, with the lowest levels of
homeowners insurance among respondents from East South Central (%)
and West South Central (%). Regions with the highest rates of homeowners
insurance include New England (%) and Pacific (%).
The respondents who replied “no” were asked, “What is the main reason
you do not have homeowners insurance?” Among this small group, the
most common answer was that it’s too expensive.
Cost is the most common reason homeowners
do not have insurance.
It’s too expensive. 43.7% 66
I don’t need it because my
home is paid o.
20.5% 31
I am not required to have it. 11.3% 17
I self-insure. 8.6% 13
I don’t trust insurance companies. 8.0% 12
Other (please specify) 8.0% 12
Most respondents (%) reported being the sole decision-maker in choosing
their homeowners insurance, though there were differences between men
Naic CIPR : extreme Weather aNd property iNsuraNce
and women. While % of men said they were the sole decision-maker,
only % of women said they were.
Price of Insurance
As of , the average annual premium for homeowners insurance covering
a home valued at $, to $, is estimated to be $ to $,.
11
Numerous factors influence the cost of homeowners insurance, most
notably real estate values and exposure to risks, such as flood, earthquake
or tornado.
More than half of consumers pay $, or
less for their homeowners insurance.
Annual Premium for Homeowners Insurance
Response
Percent Responses
Less than $1,000 per year 21.8% 508
Between $1,000 and $2,000 per year 39.8% 928
Between $2,001 and $4,000 per year 21.0% 489
More than $4,000 per year 9.5% 221
I don’t know 8.0% 186
Total 2,332
Consistent with our expectations, respondents’ reported insurance
premiums were correlated with income, suggesting that those with higher
incomes own more expensive houses that cost more to insure. More than
two-thirds (%) of respondents with income over $, reported
paying more than $, for their insurance, with % of those paying more
than $, a year. In contrast, more than half of homeowners earning
less than $, paid less than $, for their homeowners insurance.
The regions reporting the highest premiums were the Middle Atlantic (New
Jersey, New York and Pennsylvania) and West South Central (Arkansas,
Louisiana, Oklahoma, and Texas). Respondents in East South Central
(Alabama, Kentucky, Mississippi and Tennessee) reported the lowest overall
premiums.
Flood
We asked respondents, “Is flood covered by your homeowners insurance?”
This was included as a knowledge question to test whether people are
NAIC (). Dwelling Fire, Homeowners Owner-Occupied, and Homeowners Tenant and
Condominium/Cooperative Unit Owner’s Insurance Report: Data for . Accessed online
at https://www.naic.org/prod_serv/HMR-ZU-.pdf.
Naic CIPR : extreme Weather aNd property iNsuraNce
aware that flood insurance must be obtained through a separate policy
or if they believe their regular policy covers it.
More than half of respondents (%) indicated that they believe their
homeowners policy covers flood, though regional differences emerged.
Respondents in New England were least likely to answer “yes” (%),
followed by West North Central (%) and Mountain (%). Respondents
were most likely to answer “yes” in West South Central (%), Middle Atlantic
(%), Pacific (%) and South Atlantic (%). In a survey conducted by
the NAIC in that made it more clear flood was not covered under
standard policies, only % of homeowners indicated they had purchased
flood insurance.
Standard homeowners insurance policies exclude flood damage, so
homeowners must purchase separate policies, either through private
companies or the National Flood Insurance Program (NFIP), which is
operated by the U.S. Federal Emergency Management Agency (FEMA).
Based on FEMA and U.S. Census estimates,
12
only about .% of
homeowners actually have flood insurance.
To what extent do consumers see a connection
between extreme weather events and their broader
insurance coverage?
For this question, we examined whether consumers’ homeowners insurance
costs have gone up and what consumers believe prompted those increases.
We also asked whether homeowners have had trouble getting or retaining
their insurance and what they perceive to be the most pressing threats
facing their property.
Rising Costs
Among homeowners who have lived in the same house for the past three
years, % report that their premium has gone up either a little bit (%)
or significantly (%). Only % said the premium has gone down. The
remainder report that the premium has not changed (%) or that they do
not know (%).
Homeowners whose premiums went up were asked what they thought
was the most likely reason. The most common response was “insurance
companies trying to increase profits,” followed closely by “increase in
natural disasters such as hurricanes and wildfires.”
As of , there were ,, flood insurance policies in force among ,,
housing units.
Naic CIPR : extreme Weather aNd property iNsuraNce
Consumers cite a variety of reasons for why
they believe their premiums went up.
% Responses
Insurance companies trying to increase profits 26% 405
Increase in natural disasters such as hurricanes
and wildfires
26% 397
Increase in home value 23% 363
Changes to coverage 21% 328
Made a claim recently 4% 55
Total 1,548
Trouble Obtaining or Renewing Coverage
Among the respondents who believe their premium has gone up
because of an increase in natural disasters, about half (%) said they have
had trouble getting or renewing homeowners insurance because of natural
disasters, compared with less than one-third of respondents (%) overall.
People who pay higher premiums for their homeowners
insurance are more likely to have trouble getting or
renewing insurance because of wildfires or hurricanes.
Yes % No %
Less than $1,000 14% 86%
$1,000
–2,000 22% 78%
$2,001
–4,000 41% 59%
$4,000 or more 66% 34%
Across the entire sample, homeowners paying higher premiums had
statistically significantly more trouble getting or renewing insurance because
of natural disasters (% confidence level, p = .).
Notable geographic differences emerged related to this question. More
than one-third of respondents in Pacific (%; Alaska, California, Hawaii,
Oregon and Washington), West South Central (%; Arkansas, Louisiana,
Oklahoma and Texas) and Middle Atlantic (%; New Jersey, New York and
Pennsylvania) states reported challenges in getting or renewing homeowners
insurance because of natural disasters, while few respondents did in New
England (%; Connecticut, Maine, Massachusetts, New Hampshire, Rhode
Naic CIPR : extreme Weather aNd property iNsuraNce
Island and Vermont), East South Central (%; Alabama, Kentucky, Mississippi
and Tennessee) and Mountain (%; Arizona, Colorado, Idaho, Montana,
New Mexico, Nevada, Utah and Wyoming) states.
Threats
Survey respondents cited several top property threats facing their homes.
Across the entire sample, nearly one-third (%) believed that the top threat
was weather-related water damage, such as rain, melting ice or snow. About
one-quarter (%) saw wind damage from a tornado or hurricane as the
top threat. About % cited non-weather-related water damage, such as
plumbing or appliance issues. Less-frequent choices included earthquake
(%), theft (%), hail (%) and wildfire (%). A majority of homeowners
identified weather-related threats, such as water damage, hail, wind damage
and wildfire, as the most significant threats facing their homes-- except in
the West, where earthquake was identified as the top threat.
Weather-related versus non-weather-related property threats
perceived by American homeowners, by U.S. Census Region
Survey question: What would you say is the most significant property threat facing your home?
Source: NAIC’s consumer homeowner property insurance survey, conducted March -, ,
with , adults nationwide.
West
South
Midwest
Northeast
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Weather-related Non-weather-related
Water damage—weather related
Wind damage—tornado/hurricane
Hail
Wildfire
Water damage—plumbing/appliance
Earthquake
Theft
Other
Naic CIPR : extreme Weather aNd property iNsuraNce
Top threat perceptions varied dramatically by region. For example, weather-
related water damage, such as rain, melting ice or snow, was especially
salient in New England, where % of respondents cited it as their most
significant property threat. This threat was also rated as higher than average
in the Middle Atlantic (%) and East North Central (%). Similarly, while
one-quarter of respondents overall cited wind damage as the biggest
threat, this number was far higher in East South Central (%), South Atlantic
(%) and West North Central (%). While only % of respondents overall
saw earthquake as their most significant threat, % of respondents in the
Mountain region chose it. It was also higher than average in the Middle
Atlantic (%), New England, East South Central and South Atlantic regions,
cited by about % of respondents in each. Hail was far higher than average
in West North Central (%) and Mountain (%) states.
Top property threats perceived by American
homeowners vary by U.S. Census Region
Survey question: What would you say is the most significant property threat facing your home?
Source: NAIC’s consumer homeowner property insurance survey, conducted March -, ,
with , adults nationwide.
Northeast Midwest South West
39%
Weather-
related water
damage
32%
Weather-
related water
damage
35%
Wind damage
—tornado or
hurricane
23%
Earthquake
20%
Water damage
—plumbing/
appliance
27%
Wind damage
—tornado or
hurricane
25%
Weather-
related water
damage
19%
Weather-
related water
damage
17%
Wind damage
—tornado or
hurricane
15%
Water damage
—plumbing/
appliance
16%
Water damage
—plumbing/
appliance
18%
Water damage
—plumbing/
appliance
10%
The
10%
Hail
7%
Hail
13%
Wildre
6%
Earthquake
8%
The
7%
The
10%
The
Naic CIPR : extreme Weather aNd property iNsuraNce
What is the relationship between what consumers believe
about extreme weather and what actions they have taken
related to their homeowners insurance?
To explore this question, we looked at whether homeowners report being
aware of ways to protect their property and asked if they had taken steps to
do so. We also asked about their experience with threats to their property
and whether they would move somewhere that faced certain types of risks.
Respondents were asked to rate the vulnerability of their location and
consider whether extreme events are becoming more frequent. Finally, we
asked whether they would visit a website where they could learn about how
their insurance company is assessing and managing climate-related risks.
Actions to Protect Property
Most respondents overall (%) said they were aware of things they could do
to protect their property from extreme weather events to reduce their risk;
this represents % of men and % of women. However, this awareness
does not necessarily translate into action. Only half (%) of respondents
had actually made changes to their homes, though more than three-quarters
(%) said they would be willing to spend their own money to fortify their
home in exchange for a reduction in their homeowners insurance premium.
Many respondents (%) indicated they would be willing to spend between
$ and $, for a % to % premium reduction, suggesting that
homeowner willingness to invest in protective home improvements may
depend on perceived return on that investment.
Homeowners would invest up to $, in home protection
Up to $500
Between $501
and $2,500
Between $2,501
and $5,000
More than
$5,000
0% 10% 20% 30% 40% 50% 60% 70%
Naic CIPR : extreme Weather aNd property iNsuraNce
Perceptions of Future Loss Risk
Homeowners were asked if they believe recent weather events have affected
their risk of future losses to their home or property. About equal numbers
of respondents answered “yes” (%) and “no” (%), though another %
said they were not sure.
Respondents’ perception of risk seems to be growing. Using a slider bar
with values ranging from to , with being “extremely low risk,”
being “moderate risk” and being “extremely high risk,” participants
were asked to rate the level of risk they believe their property faced from
extreme weather events five years ago, today and five years from now.
Mean level of concern increased from five years ago to related to
today and looking into the future.
Risky Living
The majority of respondents (%) said they would not move to an area
that has experienced flooding or hurricanes, while even more (%) would
be reluctant to move to an area that has experienced wildfires. Women
were far less willing to move to an area with flooding or hurricanes (%
of women versus % of men said “yes”) or that had experienced wildfires
(% of women versus % of men said “yes”).
This is in line with recent research by the real estate website Redfin,
13
which
found that more than three-fourths (%) of Americans would hesitate
before buying a home in an area with increasing frequency or intensity of
natural disasters. Similar percentages would be hesitant about buying in
areas with extreme temperatures (%) or rising sea levels (%).
Risk Assessment: Worse Elsewhere but Extreme
Weather Events are More Frequent
On the whole, respondents seem to consider the area where they live
slightly less vulnerable to extreme weather events than the U.S. average. On
a scale of to , with being “extremely low risk,” being “moderate
risk” and being “extremely high risk,” the mean response was , just
below average.
Despite believing that their own area is less vulnerable to extreme events,
three-quarters (%) of respondents believe that extreme weather events
are happening more frequently overall because of climate change. There
was no significant difference between men and women. Again, the mean
Katz, Lily (). “Nearly Half of Americans Who Plan to Move Say Natural Disasters, Extreme
Temperatures Factored into Their Decision to Relocate: Survey.” Accessed May , , at
https://www.redfin.com/news/climate-change-migration-survey/.
Naic CIPR : extreme Weather aNd property iNsuraNce
response to the question “To what extent do you think climate change has
impacted the risks to your home?” was just below half, a mean response of
on a -point scale from “extremely low risk” to “extremely high risk.”
Insurer Climate Risk Disclosures
Homeowners were asked to imagine there was a website where they could
find out what insurance companies are doing to address changing weather
patterns and other climate-related risks.
Most respondents (%) indicated they would visit this website. The ,
respondents who answered “yes” were asked to what extent they would
use this information to decide which insurance company they do business
with. On a scale of to , with being “not at all” to being “a great
deal,” the median response was , and the mean was roughly . This
suggests that consumers would use this information to a moderate amount.
Most respondents said they would visit a website to find out
how insurance companies are addressing climate risks.
Information about insurer climate assessment and management is, in fact,
currently available on the California Department of Insurance (DOI) website.
This information is based on the Insurer Climate Risk Disclosure Survey,
which the NAIC adopted in . Currently, about , companies provide
information about their climate risk governance, climate risk management,
Yes
No
Naic CIPR : extreme Weather aNd property iNsuraNce
modeling and analytics, stakeholder engagement, and greenhouse gas
management. In , the CIPR released a report examining how insurers
across key characteristics assess and manage climate risks and how their
responses have changed over time.
The NAIC’s Climate and Resiliency (EX) Task Force, formed in , has been
charged with coordinating the organization’s domestic and international
efforts on climate-related risk and resiliency issues, including dialogue
among state insurance regulators and with industry, consumers and other
stakeholders.
How are homeowners segmented by their climate risk
perceptions, which may depend on loss experience, and
insurance coverage?
To explore this question, we have included a brief overview of audience
segmentation based on perception of climate change, constructed using
a validated survey instrument developed by researchers at Yale. We have
also included respondents’ personal experience with hazards.
Global Warming’s Six Americas
We used the Yale Program of Climate Change Communication and the
George Mason University Center for Climate Change Communication’s
SASSY! Group tool
14
to analyze how our survey respondents compare to
the national average in terms of their global warming beliefs, behaviors,
and policy preferences. Using national samples and machine learning
algorithms, the Yale group identified a subset of four salient questions
from an original set of questions.
The SASSY! framework uses these questions to investigate respondents’
climate perceptions:
) How important is the issue of global warming to you personally?
) How worried are you about global warming?
) How much do you think global warming will harm you personally?
) How much do you think global warming will harm future generations
of people?
Breanne Chryst, Jennifer Marlon, Sander van der Linden, Anthony Leiserowitz, Edward
Maibach and Connie Roser-Renouf (). Global Warming’s “Six Americas Short Survey:”
Audience Segmentation of Climate Change Views Using a Four Question Instrument,
Environmental Communication, :, –-, DOI: ./...
Naic CIPR : extreme Weather aNd property iNsuraNce
Using responses to these questions, the Yale group
identified six audience segments:
·
Alarmed: Those convinced global warming is happening, human-
caused, and an urgent threat. These people strongly support climate
policies.
·
Concerned:They think human-caused global warming is occurring
and that it carries a serious threat, yet they tend to believe that climate
impacts are still distant in time and space. Thus, climate change
remains a lower priority issue.
·
Cautious:Those who have not yet made up their minds.
·
Disengaged:Those who are largely ignorant about global warming.
·
Doubtful:Those who do not think global warming is happening or
believe it is simply a natural cycle.
·
Dismissive:Those who believe global warming is not happening,
human-caused, or a threat, and who espouse conspiracy theories and
hoax claims.
The true accuracy rate for the model ranges from % to % across the
six segments.
Naic CIPR : extreme Weather aNd property iNsuraNce
Over recent years, the national distribution of the Six Americas has shifted
significantly. For example, between and , the “alarmed” category
rose from % to % of the U.S. adult population, while those categorized
as “dismissive” decreased from % to %. “Cautious” and “doubtful” have
also declined.
Six Americas audience segmentation classified
more than two-thirds of survey respondents as
“alarmed” or “concerned” about climate change.
Respondents to our survey demonstrated a high level of alarm related
to climate change; % of the respondents in our sample were classified
as “alarmed,” about twice the national average. Meanwhile, the other
five groups were smaller than their national counterparts. Nationally, the
proportion of those classified as “alarmed” has been steadily growing over
the last five years, while the more skeptical segments have been declining.
However, our sample highlights an even stronger level of concern related
to climate change issues than the national average. Below is a comparison
of our sample’s segmentation to the national average.
We also analyzed the Six Americas audience segmentation across coastal
states by clustering our sample into five groups based on region. Instead
1085: Alarmed
579: Concerned
408: Cautious
32: Disengaged
203: Doubtful
158: Dismissive
Alarmed
.%
Concerned
23.5%
Cautious
16.6%
Disengaged
8.2%
Doubtful
6.4%
Dismissive
1.3%
Naic CIPR : extreme Weather aNd property iNsuraNce
of grouping states by U.S. Census region, the following analysis groups
states by coastal exposure, given that climate risk perception and loss
experience can vary substantially between coastal and non-coastal areas.
·
Pacific Coast: Alaska, California, Hawaii, Oregon and Washington
·
Gulf Coast: Alabama, Florida, Louisiana, Mississippi and Texas
·
Atlantic Coast: Connecticut, Delaware, Georgia, Maine, Maryland,
New Hampshire, New Jersey, New York, North Carolina, Rhode Island,
South Carolina and Virginia
·
Great Lakes Coast:
15
Illinois, Indiana, Michigan, Minnesota, Ohio,
Pennsylvania and Wisconsin
·
Not Coast: All other states
The aggregated bar chart shows the percent of respondents in each of
the Six Americas categories by coastal area. On average, respondents
from coastal states are more alarmed about the climate change risk than
those from non-coastal states. Half of the individuals in Pacific states (%)
were classified as alarmed, followed by % in Atlantic states. Fewer
alarmed individuals were found in Gulf Coast (%), Great Lakes (%),
and non-coastal (%) states. The differences between Pacific states and
non-coastal states and between Atlantic states and non-coastal states are
statistically significant. Respondents in Pacific states and Atlantic states are
also significantly less likely to be dismissive compared to those in non-
coastal states.
For our analysis, the Great Lakes Coast states are virtually not different from the Not Coast
states in terms of the climate change activities. However, we would like to stick to this
commonly used method to divide the geographical regions. Also, the Great Lakes Coast
may have more inland flooding than the Not Coast states.
1%
40%
30%
20%
10%
0%
Alarmed Concerned Cautious Disengaged Doubtful Dismissive
21%
40%
23%
30%
17%
21%
7%
6%
9%
8%
12%
Group Data vs. National Estimate (March 2018)
Group Data
National Estimate
Naic CIPR : extreme Weather aNd property iNsuraNce
Note: The asterisk (*) denotes a statistically significant difference compared to the
Not Coast states.
Our Six Americas audience segmentation groups survey respondents
based on their risk perceptions of climate change. Here, we examine
the relationship between climate risk perception and two factors: ) loss
experience; and ) the difficulty in obtaining homeowners insurance.
16
First- or Second-Hand Experience with Hazards
We asked whether respondents or someone they knew had suffered home
damage due to three common hazards: ) hurricane (HC); ) wildfire (WF);
and ) earthquake (EQ). As a respondent may answer “yes” to all three
questions, we divide our sample into seven exclusive groups: no loss
experience, loss experience of only one event (HC only, WF only, EQ only),
loss experience of only two events (HC & WF only, HC & EQ only, WF & EQ
only), and loss experience of all three types of events (HC & WF & EQ). On
average, % of respondents had no loss experience; % of respondents
had loss experience with all three events.
For further analysis, please refer to our academic manuscript, in which we further examine
how respondents’ climate risk perception varies across demographic, loss experience,
insurance, and other factors.
100%
80%
60%
40%
20%
0%
Pacific Ocean Atlantic Ocean Gulf Great Lakes Not
Coast Coast Coast Coast Coast
50%
*
5%
*
22%
24%
40%
24%
22%
20%
15%
18%
17%
9%
9%
9%
7%
10%
7%
8%
41%
41%
47%
*
Six Americas Segment by Coastal Area
4%
*
28%
13%
5%
*
SASSY Segment
Alarmed
Concerned
Cautious
Disengaged
Doubtful
Dismissive
Naic CIPR : extreme Weather aNd property iNsuraNce
Six Americas Level by Loss Experience
Next, we show how the average percent of alarmed respondents varies
across loss experience. We specifically focus on the alarmed category
because it has the largest portion (%) in our sample, and only alarmed
individuals strongly support climate policies.
The chart below shows that only having hurricane experience does not
effectively increase individuals’ risk perceptions of climate change, even
in coastal states. For individuals who have had loss experience of both
hurricanes and wildfire, climate risk perceptions significantly increase (except
Great Lakes Coast). Their average alarmed level is significantly higher than
that of the no-loss-experience group. For respondents who have had loss
experience of all three events, the average alarmed level further increases
and exceeds % in all areas. The single loss experience of a hurricane does
not significantly increase individuals’ response to climate change possibly
because the hurricane experience happened decades ago. The availability
heuristic, or availability bias,
17
theory suggests that individuals may primarily
rely on recent experience or information when evaluating a decision.
Note: The asterisk (*) denotes a statistically significant and higher percent compared to the
No-Loss-Experience group.
Tversky, A., & Kahneman, D. (). Availability: A heuristic for judging frequency and
probability. Cognitive psychology, (), -.
Pacific Ocean Coast
Gulf Ocean Coast
Atlantic Ocean Coast
Great Lakes Coast
Not Coast
Pacific Ocean Coast
Gulf Ocean Coast
Atlantic Ocean Coast
Great Lakes Coast
Not Coast
Pacific Ocean Coast
Gulf Ocean Coast
Atlantic Ocean Coast
Great Lakes Coast
Not Coast
38% 55%
30% 31%
35% 38%
34% 33%
28% 37%
38% 63%
30% 48%
35% 66%
34% 29%
28% 45%
*
*
*
*
38% 72%
30% 73%
35% 80%
34% 73%
28% 72%
*
*
*
*
*
Average Percent of Alarmed Respondents by Loss Experience
No Loss Experience vs. All of Hurricane, Wildfire, & Earthquake Experience
No Loss Experience vs. Hurricane & Wildfire Experience
No Loss Experience vs. Hurricane only Experience
Naic CIPR : extreme Weather aNd property iNsuraNce
Six Americas Level by Trouble Obtaining or
Renewing Homeowners Insurance
Here, we looked at whether individuals exhibit a higher alarmed level when
they have had trouble getting or renewing homeowners insurance due to
an increase in natural disasters. We expected that individuals who had ever
been unable to obtain or renew insurance coverage would be more worried
about climate change risk because they were not financially protected
against natural disaster losses. The chart below shows a consistent result. In
all areas, the average percent of alarmed respondents is statistically higher
for respondents with trouble (Yes group) than for respondents without
trouble (No group). The difference is even larger in Not Coast and Great
Lakes Coast states: Individuals in non-coastal areas were more afraid and,
thus, more concerned about climate change when they were turned down
by insurers due to increased natural disasters.
Note: The asterisk (*) denotes a statistically significant difference comparing the Yes group to
the No group.
63%
69%
43%
64%
29%
38%
73%
34%
73%
32%
Yes No Yes No Yes No Yes No Yes No
Pacific Ocean Coast Gulf Ocean Coast Atlantic Oceans Coast Great lakes Coast Not Coast
Average Percent of Alarmed Respondents by Trouble Buying Homeowners Insurance
Yes/No: Whether individual have trouble getting or renewing homeowners insurance
due to an increase in natural disasters?
Avg. Percent of the Alarmed
0.8
0.6
0.4
0.2
0.0
**
*
*
*
