A Matter of Equity

A. A. Brooks 1/11/2006

 

The CORRE proposes an across-the-board adjustment of about 1.3 % per year since retirement and lowering the spousal option reduction to 2% for all past retirees as it is for recent retirees. CORRE has not proposed any changes in the basic multiplier factor and is therefore concerned primarily with adjustments for increases in cost-of-living since the employees' retirements. Recently responding (at the request of Sen. Lamar Alexander) to a letter from a retiree, Mr. D. R. Erbschloe (DOE/Hqts Office of Science) stated (see letters):

"In 2004, UT-Battelle commissioned a Benefits Valuation Study that, according to the UT-Battelle Chief Financial Officer, shows the defined benefit plan to be approximately 14 percent higher than the comparators' plans. These comparators include several universities, Argonne National Laboratory, Pacific Northwest National Laboratory, Battelle Memorial Institute, and private industry giants such as General Electric, Motorola, Merck, Honeywell, and Dow Chemical. "

 

This statement, which implies equity in terminal wages, contradicts all of the CORRE claims and all of the data available to the public. DOE has extended the scope of the problem to include, not only the post retirement plan but also the pre retirement wages which strongly affects pension levels. Therefore it needs to be analyzed and challenged if not supported by the facts. As a first step in this I shall analyze a common model for the equity of pension plans and their immediately preceding salaries.

The equity of a pension plan is dependent on the parameters of its payment model:

RPE = MWA x Y x M x PPAC/CPIC,   where

RPE = Relative Pension Equity

MWA = Monthly Wages Averaged over Last N Years; N is about 5

Y = Years of Service; typically from 10 to 60 years

M = Multiplier Factor; often in the range of 0.012 to 0.025 (Oak Ridge; 0.012 to 0.014)

PPAC is the annual Pension Plan Adjustment Compounded over the retirement period

CPIC is the annual Consumer Price Index Compounded over the retirement period

The compounding is given by the repeated product of (1+PPA) or (1+CPI) for each retirement year

 

ASSUMPTIONS: In the absence of reliable data and in order to compare Oak Ridge to some hypothetical DOE site, some assumptions must be made as follows:

1) Monthly Wages Averaged (MWA) - We shall assume that the Monthly Wages Averaged over N years is the same for Oak Ridge and the comparison site. This assumption has several contributing assumptions and they are not at all evident. These are: a) actual job functions are equivalent, b) wages, including any cost of living considerations and local factors, are equivalent and based on job function and performance. Oak Ridge is known for the quality of its varied work but not for its generous wages. Personal experience and contacts with other DOE sites confirm this reputation. We shall assume that competition has kept the Terminal Salaries at our two sites equivalent and assume 1.00, placing our calculations on a per dollar basis. This assumption is subject to change if data is made available.

2) Years of Service (Y) – We assume that both sites have the same policies although it is known that ORO did not allow some years of Manhattan Project service.

3) Multiplier Factor (M) – We assume the current value for Oak Ridge (0.014) and the average of the five remaining DOE site for our comparison site (0.0222).

4) Pension Plan Adjustment Compounded (PPAC) – Having been refused information by the Oak Ridge plan administrator; we shall assume the experience of the author and the comparison site received adjustments (See CORRE slides (#19) equal to the Social Security Cost of Living Adjustments, SSA COLA. Many pension plans have an automatic cost of living adjustments based on the BLS CPI or the SSA COLA. These include the following: a) the Social Security System, b) the Federal Government including the retired military and the DOE, c) TVA, d) etc. Three of the five comparison sites have automatic adjustments and the other two have ad hoc adjustments.

5) Consumer Price Index Compounded (CPIC) – Based on the Bureau of Labor Statistics Consumer Price Index, BLS CPI.

Equity Computations

Params

ORO

DOE Aver

Comments

 

 

 

 

 

 

 

DOE Average = LBNL, LLNL,LANL, Sandia, PNNL averaged

MWA

1.000

1.000

Terminal Wages, assumed intersite equity

 

Y

28.000

28.000

Years of Service, AAB

 

 

M

0.014

0.022

Pension Multiplier, from CORRE Presentation Slides

PPAC

127.600

185.100

Cumulative Pension Plan Adjustments, = AAB

CPIC

186.000

186.000

Cumulative Consumer Price Index

RPE

0.269

0.613

Relative Pension Equity = MWA*Y *M *PPAC/CPIC

Note: The author apologizes for use of personal data (AAB) but he was denied access to the pension adjustment history.

Conclusions:

It should be immediately stated that the above is intended to compare the Oak Ridge site against a DOE average site and should not be used to establish the equity of today's pension compared to the retiree's terminal wages as this involves additional considerations. However the comparisons of the RPE figures are directly applicable to the question of inter-site equity.

The Oak Ridge Site Relative Pension Equity is only 44 % of the equity at the DOE average site.

This does not seem consistent with the DOE claim of Oak Ridge being 14% above a broad group average. Surely the DOE internal consistency should be better than a factor of two.

I acknowledge that such a computation as this should be carried out with the best data available and will be glad to repeat it when and if DOE will supply the data they allude to in their correspondence with Charlie Price.

 

 

******************************************************************************************************************************

 

The DOE Pension Equity Model

 

Based on a recently, received copy of a DOE letter (D. R. Erbschole) to CORRE (D. E. Reichle) dated 12.23. 2005:

"The benefit provided by retirement plans (defined benefit and defined contribution)' when combined with Social Security payments, on the average replace in excess of 90 percent of a retiree's final salary for retirees who take full advantage of the retirement plans, have a career of at least 35 years, and retire at age 65."

The example given is for 28 DOE contractor employees for the year 2003 both hourly and salaried. It is interesting to note that the DOE to Price letter list several large organizations but does not associate them with DOE.

Translation of this information into a formula for comparison with the above, gives (on a per dollar basis):

RPE = (Y x M) + SSP/MWA,   where

SSP= the Social Security Payment

And the rest have the same meaning as above.

The need to know the Social Security payments which vary with the lifetime wages makes this model difficult to use. Using the 2003 parameters of the DOE example: Y x M = 35 x 0.014 = 0.49 or only half of the final wages leaving the other half to be made up by Social Security, an unlikely event.

However if one corrects this formula for the passage oft time, one gets:

          RPE = (Y x M)/CPIC + (SSP/MWA) x (PPIC/CPIC)

From this it is clear the DOE contribution grows smaller and the SSA contribution remains relatively constant. This will remain true as long as DOE only makes partial adjustments.

From the DOE model one can claim that the Oak Ridge Pension plans are perhaps equitable on your retirement date BUT it doesn't last long. It is quite clear that there are aspects of the DOE plan which are not equitable, namely, the retirement years. Also, the DOE model does nothing to explain the high multiplier factors of the southwest laboratories.

Simply put, just because a pension plan is equitable on your retirement date doesn't mean it is equitable twenty years later.