AIG 2014 Annual Report Download - page 211

ITEM 7 / CRITICAL ACCOUNTING ESTIMATES

194
below are only for the years using an expected loss ratio approach. Actual loss cost trends in the early 1990s were negative for
several years whereas actual loss cost trends exceeded the figures cited below for 1997 through 2001. Loss trends may
deviate by more than the amounts noted above and discussed below.
Loss development factors: The percentage deviations noted in the table below are not considered the highest possible
deviations that might be expected, but rather what we consider to reflect a reasonably likely range of potential deviation. While
multiple scenarios are performed, the assumed loss development factors are a key assumption. Generally, actual historical
loss development factors are used to project future loss development. Future loss development patterns may be different from
those in the past, or may deviate by more than the amounts noted above and discussed below.
AIG’s loss reserve analyses do not generally provide a range of loss reserve estimates. A large portion of the loss reserves
from the Non-Life Insurance Companies business relates to longer-tail casualty classes of business, such as excess casualty
and D&O, which are driven by severity rather than frequency of claims. Using the reserving methodologies described above,
our actuaries determine their actuarial central estimates of the loss reserves and advise management on their final
recommendation for management’s best estimate of the recorded reserves. Subject matter experts from underwriting and
claims play an important part in informing the actuarial assumptions and methods. The governance process over the
establishment of loss reserves also ensures robust considerations of the changes in the loss trends, terms and conditions,
claims handling practices, and large loss impact when determining the methods, assumptions and the estimations. This multi-
disciplinary process engages underwriting, claims, risk management, business unit executives and senior management and
involves several iterative levels of feedback and response during the regular reserving process.
The sensitivity analysis below addresses each major class of business for which there is a possibility of a material deviation
from our overall reserve position. The analysis uses what we believe is a reasonably likely range of potential deviation for each
class. Actual reserve development may not be consistent with either the original or the adjusted loss trend or loss development
factor assumptions, and other assumptions made in the reserving process may materially affect reserve development for a
particular class of business.
Class of Business Loss Cost Trend Loss Development Factor
Excess Casualty
The assumed loss cost trend was approximately five percent in
the 2014 reserve review. After evaluating the historical loss cost
trends from prior accident years since the early 1990s, in our
judgment, it is reasonably likely that actual loss cost trends
applicable to the year-end 2014 loss reserve review for excess
casualty will range from 0 percent to positive ten percent. The
loss cost trend assumption is critical for the excess casualty class
of business due to the long-tail nature of the claims and therefore
is applied across many accident years. Thus, there is the
potential for the reserves with respect to a number of accident
years (the expected loss ratio years) to be significantly affected
by changes in loss cost trends that were initially relied upon in
setting the reserves. These changes in loss trends could be
attributable to changes in inflation or in the judicial environment,
or in other social or economic conditions affecting claims.
After evaluating the historical loss
development factors from prior
accident years since the early
1990s, in our judgment, it is
reasonably likely that actual loss
development factors will range
from approximately 2 percent
below those actually utilized in the
year-end 2014 reserve review to
approximately 3 percent above
those factors actually utilized.
Excess casualty is a long-tail class
of business and any deviation in
loss development factors might not
be discernible for an extended
period of time subsequent to the
recording of the initial loss reserve
estimates for any accident year.
Thus, there is the potential for the
reserves with respect to a number
of accident years to be significantly
affected by changes in loss
development factors that were
initially relied upon in setting the
reserves. These changes in loss
development factors could be
attributable to changes in inflation