3a. In general, we believe that a reasonable actuarial funding (or cost) method should
include a projection of salary and other factors in determining a plan's annual funding costs consistent
with current GASB requirements. We understand that currently GASB accepts six different actuarial
funding methods for financial statement reporting (entry age, frozen entry age, attained age, frozen
attained age, projected unit credit and aggregate). We also understand that each of these six methods
project automatic COLAs, future salary increases and future service credits to calculate plan liabilities.
Therefore, to the extent that these items are projected as part of a plan's actuarial funding method, we
agree with the that automatic COLAs, future salary increases and future service credits should be
included in the calculation of the total pension liability and the service cost.
3b. With respect to ad hoc COLAs, to the extent that a plan has established a consistent
method of providing ad hoc COLAs, we agree with the Preliminary View that such ad hoc COLAs
should be included in the plan's actuarial funding method. For this purpose, we would view an ad hoc
COLA as substantively the same as an automatic COLA if it has been consistently provided in both
timing and amount over a period of at least nine years, which reflects the average length of three
consecutive bargaining contract periods.
3c. We disagree in part with the Preliminary View that a single "blended" rate as defined in
Chapter 4, Paragraph 14 should be used to determine plan liabilities. The states that the long-term
expected rate of return on plan investments should be used to the extent that current and future plan
assets are projected to be sufficient to make benefit payments. Chapter 4, Paragraph 17 describes
future plan assets as follows:
The projection of future employer contributions for this purpose should
reflect a reasonable expectation of future employer contribution levels
for current employees and should consider factors such as the employer's
stated contribution policy and recent contribution pattern.
As we suggested in our response to Issue 1, an employer's stated contribution policy should be
built with a view as to how future contributions will be sufficient to provide the defined benefits.
Therefore, in our view the blended rate would never apply given an employer's stated funding policy
that is designed to in fact ultimately provide the benefit.
This leaves the ambiguous phrase "recent contribution pattern" as relevant to whether the
blended rate should be used. However, we do not believe that past practice is indicative of future
contribution patterns. Consider, for instance, a plan that is 40% funded (using the expected rate of
return on assets) and 25% funded (using the blended rate). The funded status has been driven by a
combination of low investment returns and a recent employer practice of contributing less than the
ARC. However, the employer and members agree in good faith to a combination of increased
employer contributions and certain benefit concessions. As a result, future contributions are expected
Director of Research and Technical Activities,
Project No. 34
September 14, 2010
to provide for all future benefits. We believe the expected rate of return on plan assets should be used
in this instance to determine plan assets despite the fact that there has been a recent pattern of
insufficient employer contributions. Shortfalls in meeting the funding plan should be disclosed. The
actuary should set the "expected rate of return" assumption by considering actual plan assets (as well
as other factors). If, for example, the system is in "pay-as-you-go" status, the rate should be based on
the return on general employer assets. In any event there would not be a "blended rate," but instead a
rate reflecting the actuary's judgment of the situation at hand.
As we have indicated, we believe that the assumptions used for financial statement reporting
purposes should be consistent with the assumptions used under a stated funding policy. As a result, we
believe that the discount rate should be the long-term expected rate of return of plan assets, and that
there should be no disconnect between the employer's funding costs and accounting expense. We
believe the actuary should take into account certain factors such as any expectation that the plan assets
may deplete or the employer may pay plan benefits on a pay-as-you-go basis when setting the plan's
expected rate of return on plan assets. We think connecting blocks of plan liabilities to the municipal
bond yield has no better theoretical underpinnings than connecting the same to a diversified portfolio
yield. The municipal bond yield will simply produce more volatility and confuse the stakeholders as to
what an appropriate funding level is.
3d. The Preliminary View is to use the entry age actuarial cost method applied on a levelpercentage
of payroll basis to determine the total pension liability and the service-cost component of
pension expense. We realize that a consistent method by which plan costs are funded and expensed
reduces complexity, enables the user of financial statement information to appropriately interpret
results, and eliminates any potential bias in reporting financial information. We understand that a
majority of plans currently use the entry age method. If the Board adopts entry age method as the
standard for financial statement purposes, we would anticipate that many additional employers would
switch to entry age method for ongoing funding purposes. These plans could, however, experience
significant disconnects between the method they are using and the entry age method. We feel the
Board should carefully consider this issue when drafting any transition rules from the current standards
to the new standards, allowing for a longer transition period to accommodate those employers who
may change funding methods.