Auto-insuranceexperts are falling into two camps regarding the future of telematics:
good drivers will choose telematics policies that will ultimately
be cheaper than classic policies,
or bad/new drivers will choose
telematics policies to build or
improve their experience before
switching to a classic policy.
Celent outlined these two hy-
potheses, termed “self-selection
hypothesis” (good drivers opting
for telematics) and “null hypoth-
esis” (bad drivers choosing telematics) in
a recent report, “Innovation in Focus: The
Great Telematics Experiment.”
According to the report, the null hy-
pothesis assumes that drivers and car
owner will prefer classis auto-insurance
policies. “Here, we assume that custom-
ers don’t want insurers spying on their
movements, that capturing the data is
expensive and that the data doesn’t pro-
vide significant advantages in pricing and
underwriting,” Celent states.
The firm adds that telematics policies
would, in this case, be a “stepping stone”
for some to ultimately get a classic policy
at a reasonable cost. Drivers with poor history, or who need to build up experience
to prove they can be good risks, would use
telematics policies as “a first rung on the
ladder” toward a classic policy.
Classically insured drivers, too, may
opt for a telematics policy if they have an
at-fault accident and want to prove they
can be a better driver.
Meanwhile, the self-selection hypoth-
esis contends that bad drivers will fail to
get rewards from telematics policies and
will ultimately opt for cheaper classic pol-
icies. “The classic policies then become
less profitable over time due to the popu-
lation of bad drivers,” Celent says. “The
premiums for classic policies necessarily
increase over time as the book of busi-
ness increases in risk and the good driv-
ers self-select to the telematics policies.”
So which side is correct? Celent says it
is still up in the air. “Actually this is the
beauty of this great experiment: the an-
swer isn’t clear yet,” the report states.
Celent offers comments on two fac-
tors that could tip the scale toward one
hypothesis over the other: the cost of
telematics-based insurance, and people’s
willingness to share their data.
Regarding the costs, Celent notes, “The
devices cost money. Transmitting the
data captured in this process costs money.
Storing the data costs money. Analyzing
the data costs money. In short, there are
an array of costs associated with captur-
ing and leveraging this data.”
But Celent adds that these costs
are dropping over time as tech-
nology evolves. Advances in both
consumer and auto technologies
“offer routes to low-cost, effec-
tive telematics-based solutions,”
Celent says, noting that over the
next decade, newer, cheaper mod-
els for delivering telematics-based
insurance should come to market.
As for sharing data, Celent says,
“Recent concerns over NSA ac-
tivities in America, over the use
of drones and the deployment of
closed circuit TV and road cameras have
highlighted privacy issues and concerns
in the general populace.”
However, Celent notes that the rise of
social networks and the “trend of ‘check-
ing in’ to locations on Facebook and
Foursquare suggest that some people
don’t mind sharing information for the
right rewards—whether financial or non-
The key for insurers, according to
Celent, is to offer rewards and benefits to
customers that they would be willing to
accept in exchange for less privacy. Regulators, meanwhile, would be tasked with
protecting consumers in this new environment while not stifling innovation in
Good Driver/Bad Driver:
Who Will Gravitate to Telematics?
By Phil Gusman, PropertyCasualty360.com
of carriers are effectively modeling and
profitably insuring terrorism risk today.
Still, though, he says, “There is a high
uncertainty regarding certain types of
terrorist attacks,” and he mentions NCBR
attacks as an example.
Regarding conventional attacks—such
as bomb attacks—Folkman says the range
of damage is much better understood,
and RMS has been able to validate its
models against historical explosions.
On the larger TRIA debate, Folkman
says RMS’ role is to “quantify risk as best
as we can. We don’t hold opinion for or
against” an extension. He notes though
that many industry members have stated
they would be hesitant to offer terrorism
coverage without the backstop.
Models give insurers a good under-
standing of the risk he says, but some in-
surers feel the magnitude of some types
of attacks could bankrupt the industry.
Terrorism | continued from p. 9