Dominion Power 2007 Annual Report Download - page 36

Management’s Discussion and Analysis of Financial Condition and Results of Operations, Continued
We utilize periodic site-specific base year cost studies in order
to estimate the nature, cost and timing of planned decommission-
ing activities for our utility and merchant nuclear plants. We
obtained updated cost studies for all of our nuclear plants in 2006
which generally reflected increases in base year costs. These cost
studies were based on relevant information available at the time
they were performed; however, estimates of future cash flows for
extended periods of time are by nature highly uncertain and may
vary significantly from actual results. In addition, our cost esti-
mates include cost escalation rates that are applied to the base year
costs. The selection of these cost escalation rates is dependent on
subjective factors which we consider to be a critical assumption.
We determine cost escalation rates, which represent projected
cost increases over time, due to both general inflation and
increases in the cost of specific decommissioning activities, for
each of our nuclear facilities. The use of alternative rates could
have been material to the liabilities recognized. For example, had
we increased the cost escalation rate by 0.5%, the amount recog-
nized as of December 31, 2007 for our AROs related to nuclear
decommissioning would have been $267 million higher.
E
MPLOYEE B ENEFIT P LANS
We sponsor noncontributory defined benefit pension plans and
other postretirement benefit plans for eligible active employees,
retirees and qualifying dependents. The projected costs of provid-
ing benefits under these plans are dependent, in part, on historical
information such as employee demographics, the level of con-
tributions made to the plans and earnings on plan assets. Assump-
tions about the future, including the expected rate of return on
plan assets, discount rates applied to benefit obligations and the
anticipated rate of increase in health care costs and participant
compensation, also have a significant impact on employee benefit
costs. The impact of changes in these factors, as well as differences
between our assumptions and actual experience, is generally
recognized in our Consolidated Statements of Income over the
remaining average service period of plan participants, rather than
immediately.
The expected long-term rates of return on plan assets, dis-
count rates and medical cost trend rates are critical assumptions.
We determine the expected long-term rates of return on plan
assets for pension plans and other postretirement benefit plans by
using a combination of:
‰Historical return analysis to determine expected future risk
premiums;
‰Forward-looking return expectations derived from the yield
on long-term bonds and the price earnings ratios of major
stock market indices;
‰Expected inflation and risk-free interest rate assumptions; and
‰Investment allocation of plan assets. The strategic target asset
allocation for our pension fund is 34% U.S. equity securities,
12% non-U.S. equity securities, 22% debt securities, 7% real
estate and 25% other, such as private equity investments.
We develop assumptions, which are then compared to the
forecasts of other independent investment advisors to ensure rea-
sonableness. An internal committee selects the final assumptions.
We calculated our pension cost using an expected return on plan
assets assumption of 8.75% for 2007, 2006 and 2005. We calcu-
lated our 2007, 2006 and 2005 other postretirement benefit cost
using an expected return on plan assets assumption of 8.00%.
The rate used in calculating other postretirement benefit cost is
lower than the rate used in calculating pension cost because of
differences in the relative amounts of various types of investments
held as plan assets.
We determine discount rates from analyses of AA/Aa rated
bonds with cash flows matching the expected payments to be
made under our plans. The discount rates used to calculate
pension cost and other postretirement benefit cost were 6.20%
and 6.10%, respectively, in 2007 compared to 5.60% and 5.50%,
respectively, in 2006, and 6.00% for both discount rates in 2005.
Higher long-term bond yields were the primary reason for the
increase in the discount rate from 2006 to 2007. We selected
discount rates of 6.60% and 6.50% for determining our
December 31, 2007 projected pension and postretirement benefit
obligations, respectively.
We establish the medical cost trend rate assumption based on
analyses of various factors including the specific provisions of our
medical plans, actual cost trends experienced and projected, and
demographics of plan participants. Our medical cost trend rate
assumption as of December 31, 2007 is 9.00% and is expected to
gradually decrease to 5.00% in later years.
The following table illustrates the effect on cost of changing
the critical actuarial assumptions previously discussed, while hold-
ing all other assumptions constant:
Increase in Net Periodic Cost
Change in
Actuarial
Assumption
Pension
Benefits
Other
Postretirement
Benefits
(millions, except percentages)
Discount rate (0.25)% $ 13 $ 5
Rate of return on plan assets (0.25)% 12 2
Healthcare cost trend rate 1% N/A 20
In addition to the effects on cost, a 0.25% decrease in the
discount rate would increase our projected pension benefit obliga-
tion by $117 million and would increase our accumulated post-
retirement benefit obligation by $43 million at December 31,
2007.
A CCOUNTING FOR G AS AND O IL O PERATIONS
We follow the full cost method of accounting for gas and oil E&P
activities prescribed by the Securities and Exchange Commission
(SEC). Under the full cost method, all direct costs of property
acquisition, exploration and development activities are capitalized
and subsequently depleted using the units-of-production method.
The depletable base of costs includes estimated future costs to be
incurred in developing proved gas and oil reserves, as well as cap-
italized asset retirement costs, net of projected salvage values.
Capitalized costs in the depletable base are subject to a ceiling test
prescribed by the SEC. The test limits capitalized amounts to a
ceiling—the present value of estimated future net revenues to be
derived from the production of proved gas and oil reserves, dis-
counted at 10 percent, assuming period-end pricing adjusted for
any cash flow hedges in place. We perform the ceiling test quar-
terly, on a country-by-country basis as applicable, and would
recognize asset impairments to the extent that total capitalized
costs exceed the ceiling. In addition, gains or losses on the sale or
other disposition of gas and oil properties are not recognized,
unless the gain or loss would significantly alter the relationship
34 Dominion 2007 Annual Report