Assurant 2014 Annual Report Download - page 53

39ASSURANT, INC. – 2014 Form 10-K
PART II
ITEM 7 Management’s Discussion and Analysis of Financial Condition and Results of Operations
Reserves for our various product lines are calculated using
experience data where credible. If suffi cient experience
data is not available, data from other similar blocks may be
used. Industry data provides additional benchmarks when
historical experience is too limited. Reserve factors may
also be adjusted to refl ect considerations not refl ected in
historical experience, such as changes in claims inventory
levels, changes in provider negotiated rates or cost savings
initiatives, increasing or decreasing medical cost trends,
product changes and demographic changes in the underlying
insured population.
Key sensitivities as of December 31, 2014 for short duration medical reserves include claims processing levels, claims under
case management, medical infl ation, seasonal effects, medical provider discounts and product mix. The effects of these
sensitivities can be summarized by adjusting loss development factors, as follows:
Claims and Benefi ts Payable
Short duration medical, loss development factors 1% lower* $ 401,200
Short duration medical, as reported $ 378,200
Short duration medical, loss development factors 1% higher* $ 358,200
* This refers to loss development factors for the most recent four months. Our historical claims experience indicates that approximately 84% of
medical claims are paid within four months of the incurred date.
Changes in medical loss development may increase or decrease
the MLR rebate liability.
Property and Warranty
Our Property and Warranty lines of business include lender-
placed homeowners, manufactured housing homeowners,
multi-family housing, credit property, credit unemployment
and warranty insurance and some longer-tail coverages
(e.g. asbestos, environmental, other general liability
and personal accident). Claim reserves for these lines
are calculated on a product line basis using generally
accepted actuarial principles and methods. They consist
of case and IBNR reserves. The method we most often use
in setting our Property and Warranty reserves is the loss
development method. Under this method, we estimate
ultimate losses for each accident period by multiplying
the current cumulative losses by the appropriate loss
development factor. We then calculate the reserve as the
difference between the estimate of ultimate losses and
the current case-incurred losses (paid losses plus case
reserves). We select loss development factors based on a
review of historical averages, adjusted to refl ect recent
trends and business-specifi c matters such as current claims
payment practices.
The loss development method involves aggregating loss
data (paid losses and case-incurred losses) by accident
quarter (or accident year) and accident age for each
product or product grouping. As the data ages, we compile
loss development factors that measure emerging claim
development patterns between reporting periods. By
selecting the most appropriate loss development factors,
we project the known losses to an ultimate incurred basis
for each accident period.
The data is typically analyzed using quarterly paid losses
and/or quarterly case-incurred losses. Some product
groupings may also use annual paid loss and/or annual
case-incurred losses, as well as other actuarially accepted
methods.
Each of these data groupings produces an indication of
the loss reserves for the product or product grouping.
The process to select the best estimate differs by line of
business. The single best estimate is determined based on
many factors, including but not limited to:
•the nature and extent of the underlying assumptions;
•the quality and applicability of historical data — whether
internal or industry data;
•
current and future market conditions — the economic
environment will often impact the development of loss
triangles;
•
the extent of data segmentation — data should be
homogeneous yet credible enough for loss development
methods to apply; and
•the past variability of loss estimates — the loss estimates
on some product lines will vary from actual loss experience
more than others.
Most of our credit property and credit unemployment insurance
business is either reinsured or written on a retrospective
commission basis. Business written on a retrospective
commission basis permits management to adjust commissions
based on claims experience. Thus, any adjustment to prior
years’ incurred claims is partially offset by a change in
commission expense, which is included in the underwriting,
general and administrative expenses line in our consolidated
statements of operations.
While management has used its best judgment in establishing
its estimate of required reserves, different assumptions and
variables could lead to signifi cantly different reserve estimates.
Two key measures of loss activity are loss frequency, which
is a measure of the number of claims per unit of insured
exposure, and loss severity, which is a measure of the average
size of claims. Factors affecting loss frequency include the
effectiveness of loss controls and safety programs and changes
in economic activity or weather patterns. Factors affecting
loss severity include changes in policy limits, retentions,
rate of infl ation and judicial interpretations.