Unum 2007 Annual Report Download - page 36

Management’s Discussion and Analysis of
Financial Condition and Results of Operations
34 Unum 2007 Annual Report
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 new claims will be established below or above the discount
rate assumption on existing claims to reflect 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 separately 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.