Unum 2008 Annual Report Download - page 36

32




Key Assumptions
The calculation of policy and claim reserves involves numerous assumptions, but the primary assumptions used to calculate reserves are
(1) the discount rate, (2) the claim resolution rate, and (3) the claim incidence rate for policy reserves and IBNR claim reserves. Of these
assumptions, our discount rate and claim resolution rate assumptions have historically had the most significant effects on our level of
reserves because many of our product lines provide benefit payments over an extended period of time.
1. The discount rate, which is used in calculating both policy reserves and incurred and IBNR claim reserves, is the interest rate that we
use to discount future claim payments to determine the present value. A higher discount rate produces a lower reserve. If the discount
rate is higher than our future investment returns, our invested assets will not earn enough investment income to support our future
claim payments. In this case, the reserves may eventually be insufficient. We set our assumptions based on our current and
expected future investment yield of the assets supporting the reserves, considering current and expected future market conditions.
If the investment yield on new investments that are purchased is below or above the investment yield of the existing investment
portfolio, it is likely that the discount rate assumption on claims will be established to reflect the effect of the new investment yield.
2. The claim resolution rate, used for both policy reserves and incurred and IBNR claim reserves, is the probability that a disability
claim will close due to recovery or death of the insured. It is important because it is used to estimate how long benefits will be paid
for a claim. Estimated resolution rates that are set too high will result in reserves that are lower than they need to be to pay the
claim benefits over time. Claim resolution assumptions involve many factors, including the cause of disability, the policyholder’s
age, the type of contractual benefits provided, and the time since initially becoming disabled. We use our own claim experience
to develop our claim resolution assumptions. These assumptions are established for the probability of death and the probability of
recovery from disability. Our studies review actual claim resolution experience over a number of years, with more weight placed
on our experience in the more recent years. We also consider any expected future changes in claim resolution experience.
3. The incidence rate, used for policy reserves and IBNR claim reserves, is the rate at which new claims are submitted to us. The
incidence rate is affected by many factors including the age of the insured, the insured’s occupation or industry, the benefit plan
design, and certain external factors such as consumer confidence and levels of unemployment. We establish our incidence
assumption using a historical review of actual incidence results along with an outlook of future incidence expectations.
Establishing reserve assumptions is complex and involves many factors. Reserves, particularly for policies offering insurance coverage
for long-term disabilities, are dependent on numerous assumptions other than just those presented in the preceding discussion. The impact
of internal and external events, such as changes in claims management procedures, economic trends such as the rate of unemployment and
the level of consumer confidence, the emergence of new diseases, new trends and developments in medical treatments, and legal trends
and legislative changes, among other factors, will influence claim incidence and resolution rates. Reserve assumptions differ by product line
and by policy type within a product line. Additionally, in any period and over time, our actual experience may have a positive or negative
variance from our long-term assumptions, either singularly or collectively, and these variances may offset each other. We test the overall
adequacy of our reserves using all assumptions and with a long-term view of our expected experience over the life of a block of business
rather than test just one or a few assumptions independently that may be aberrant over a short period of time. Therefore it is not possible to
bifurcate the assumptions to evaluate the sensitivity of a change in each assumption, but rather in the aggregate by product line. We have
presented in the following section an overview of our trend analysis for key assumptions and the results of variability in our assumptions,
in aggregate, for the reserves which we believe are reasonably possible to have a material impact on our future financial results if actual
claims yield a materially different amount than what we currently expect and have reserved for, either favorable or unfavorable.
Trends in Key Assumptions
Because our actual experience regarding persistency and claim incidence has varied very little from our policy reserve and IBNR
claim reserve assumptions, we have had minimal adjustments to our persistency assumptions and claim incidence assumptions during
the years 2006 through 2008. Generally, we do not expect our mortality and morbidity claim incidence trends or our persistency trends to
change significantly in the short-term, and to the extent that these trends do change, we expect those changes to be gradual over a longer
period of time. However, we have historically experienced an increase in our group long-term disability morbidity claim incidence trends