Page 44 - Environmental Report for EXL Services_Web Only

Helping Clients
Our Presence Highlights World is Transition to There is Strategic Accomplishing Transitioning Climate Accelerating Water Use & Cultivating use Structured GRI GHG List of
Across and Progress Influencing Climate- aligned Always a through to Climate Positive Waste Culture of
the Globe on Goals Our Journey World Better Way Framework Actions Aligned Business Commitments Change Management Shared Value Environmental Data Index Statement Abbreviations
Extreme weather events are predicted to
account for the largest amount of
compensation insurers’ will have to pay out
over the next 10 years. Such trends will
require modeling that assists insurance
providers to better assess climate risk at the
property level, as well as mechanisms that
help organize collaborative action among
insurance carriers, state regulators,
communities, and homeowners to reduce
climate risk. As it always has, a company
that anticipates business disruptions such
as climate change is better prepared to
mitigate any negative impact on their
business.
First Steps
Insurers have already taken steps in the right
direction, starting with limited solutions such
as overlaying topographical maps with the
results from existing models to underwrite
properties. Yet, this simple approach does not
leverage the data that insurance carriers can
quickly access. They have massive databases
from simulation models and satellites when it
comes to weather and climate. The problem is
figuring out how to use them to their full
potential. As a result, scientists have turned to
computer vision to identify hail-bearing storms.
So far, damage prediction has been done
manually. Insurance carriers would
undoubtedly profit from the forward-thinking
strategy of exploiting property attributes, both
in terms of location-level risk management and
prudential capital needs.
Quick Weather Data Feeds Drive Risk Control
Decisions
Insurers need to understand the gaps in the
understanding of physical processes and the
inadequacy of weather information required for
high-resolution modeling of extreme weather
events. This would allow models to better
factor in the year-on-year effect of global
warming on the local and global weather as
well as predict extreme weather risk. Insurers
could then assess climate exposure to their
book roll, which would lead to identifying
structural inadequacies on a single policy.
They could then enforce risk control measures
like requesting changes, and, in case of
non-compliance by the policyholder, increase
the premium or reject policy renewal.
Forecasting the data well in advance along
with analyzing the existing property condition
can be a good indicator of potential claims.
Adjusting policies to account for risk-control
measures that can be accounted for before the
extreme weather conditions intensify will only
benefit both the insurer and the policyholder.
For this all to work, underwriters and data
scientists will need to integrate weather data
Assurance 44
for Risk Aversion
with property insights derived from internal
and external sources. For example, weather
variables such as low-level moisture,
convective instability and vertical wind shear
all have been linked to hail formation. What
does the historical data indicate? Now factor
in data on building construction to predicting
the risk of hail damage.
Similarly, when predicting flood risk to a
property, most carriers rely on the flood
insurance risk map that divides the U.S. into
low/moderate/high flood zones. However, the
premiums should be based on the actual flood
risk to a property structure rather than the
zone it is located in. Although some carriers do
take into account the design of the structure,
age of the structure and elevation or if the
foundation is subgrade or not, the premiums in
most cases do not reflect actual risk.
Therefore, a view of the current property
condition plus plugging in real-time weather
forecasts can be leveraged for a more active
approach to claims management. For example,
the path followed by natural catastrophic
events and their intensity can be clubbed with
property attributes for the scheduling of SIUs
and fraud detection. In this way, properties
that might have poor roofs and have been in
the eyewall of the storm can be prioritized and
triaged accordingly.