Pennsylvania Law Weekly
Monday, June 3, 2002
One of the thorniest areas for courts faced with making decisions about equitable distribution is how to value pensions. In a recent decision, the state Superior Court provided much-needed direction on the use of actuarial standards in fixing a "normal" retirement age to be used in pension valuation.
In calculating the present value of a pension, the choice of the retirement age is an important factor. In most cases, the normal retirement age is used, which is the earliest date the individual can retire with an unreduced pension. However, suppose the individual is past the normal retirement date? The question then arises what retirement age to use. One obvious answer is to use the current age, if the employee is beyond normal retirement age. The reason for this is that the value of the pension may be seen to be equivalent to its purchase price, what it is worth on the market. This is the finding of the Superior Court in Palladino v. Palladino, 713 A.2d 1072 (Pa. Super. 2001).
The Actuarial Standards Board of the American academy of Actuaries has addressed the issue. Actuarial Standard of Practice No. 34 (Actuarial Practice Concerning Retirement Plan Benefits in Domestic Relations Actions), listed the purchase price as one of the methods in valuing pensions (Section 184.108.40.206. If an individual is 50 years old and the benefit is payable immediately, a private carrier will use age 50 as the commencement date in calculating the value of an annuity. Thus, it is reasonable to use the current age, when the participant is beyond the normal retirement age, since the pension is payable immediately.
In DeMarco v. DeMarco, 787 A.2d 1072 (Pa. Super. 2001), the employee was 51 at time of trial and the normal retirement age for the man's profession was age 50. The husband is a police officer with the City of Pittsburgh. The valuator used age 50 as the retirement age. This made no sense, since the employee had not retired at age 50. The Superior Court found that use of retirement age 50 was unreasonable:
"The trial court concedes its selection of age fifty was an arbitrary age chosen to maximize the value of this asset," the court wrote.
Age 51 would be a reasonable retirement age in a pension valuation, under the theory of market worth. However, husband argued that the pension should be valued using the average retirement age, which he claimed to be 65. The Superior Court accepted this argument, and found that the retirement age to be used in valuing the pension should be based on the average age of retirement, as well as other related factors. This is also a method indicated in the Actuarial Standard of Practice (3.3.3.d). Thus, when the employee is beyond the normal retirement age and the retirement age is in dispute, DeMarco indicates the use of "statistical data regarding average age of retirement from the company or industry with which the employee/spouse is affiliated."
Note that in DeMarco, the issue was delayed retirement, where the employee spouse argues for use of delayed retirement age. In the opposite situation, the non-employee spouse may argue for the use of an earlier "normal" retirement age. This typically arises in "age and service" formulas.
Suppose the normal retirement age is 65, but an employee may retire with an unreduced pension under a "rule of 85" at age 55 with 30 years of service (age plus service equals 85). At the time of the valuation, assume the employee is age 50 with 25 years of service. Use of age 55 as the normal retirement age, rather than age 65, results in a much higher present value of the pension.
Absent of plan turnover statistics, the assumption of five years of future service followed by retirement at age 55 is purely speculative. This speculation leads to a biased result, intended to "maximize the value" of the pension, rather than to determine an objective present value. While the DeMarco court did not specifically address this issue, it did address the use of plan or industry statistics, and did find fault with a biased result.
Actuarial Standard of Practice No. 34 does address this issue, in section 3.3.2.b. Regarding age and service retirement, the Actuarial Standards Board recommended the use of either the "immediate termination approach," which assumes no future service (and thus retirement age 65), or the "continued employment approach ... in accordance with selected ... turnover... assumptions." The practice standard then goes on the state that turnover assumption should reflect the turnover experience of plan participants. Thus, the use of the earlier retirement age (55) based upon future service needs to take into account the likelihood of the employee working the additional five years, and then retiring.
The higher present value using age 55 must then be multiplied by the probability (less than 1) of the employee working the additional 5 years, and then retiring. Thus, there is an additional discount factor in assuming the earlier age and service retirement. The Standards Board did not recommend the age and service retirement age, absent of turnover statistics. Without such statistics, use of age 55 (rule of 85) is a speculative, biased result, intended to "maximize the value" of the pension. Note that the term in quotes is taken from DeMarco. Although DeMarco addressed different circumstances, the principal is the same.
The DeMarco decision did leave one situation open. The trial court used a mixed method of present value and deferred distribution in the DeMarco case. Although a present value was calculated, there were insufficient funds for immediate offset. The present value was used, along with other assets, in order to determine what percentage went to wife. The wife in DeMarco received 50 percent of the marital portion of the pension, based on the present value.
But the Superior Court has ordered the pension to be revalued, using the average retirement age, which will change the present value: "If the value placed on the pension is changed, Wife's proportional share of the pension is changed as well. Thus, we vacate the trial court's equitable distribution order in its entirety."
Suppose it is established that the average age of retirement for a Pittsburgh police officer is 58, and the pension is valued using age 58 as the retirement age. The wife's portion of the pension under a domestic relations order would then be determined using this present value, assuming the same mixed method of distribution is used. What if the police officer does not retire at age 58? Under an ERISA pension, this does not matter, since the alternate payee (the officer's spouse) can commence as soon as the participant's earliest retirement age, whether or not the employee retires.
However, in a non-ERISA pension, a spouse cannot commence benefits until the employee retires. If the officer does not retire at age 58 (assuming this is the average retirement age), the present value of the benefit the spouse receives will be less than that used to determine her portion. This, in effect, results in the dissipation of a marital asset.
One solution to this issue is to have a sliding scale that will increase the wife's portion, based upon delayed retirement age. This exact issue has been dealt with the California case Cornejo vs. Cornejo, 916 P.2d 476 (Cal. 1996). If the participant delays retirement in a non-ERISA pension when there is a deferred distribution, the alternate payee must receive an immediate cash payment equal to the present value of his or her benefit under the court's distribution order, assuming normal retirement age.
If the participant lacks sufficient funds for the payment, Cornejo essentially forces the participant to retire. The use of the sliding scale allows much more flexibility. The Pennsylvania Superior Court, in DeMarco, was not entirely clear on this issue. Thus, in cases like DeMarco, it may be reasonable to use a sliding scale for delayed retirement, until the issue is definitively addressed by an appellate court.
Although DeMarco did not seal this one issue when there is a mixed distribution scheme and deferred distribution with a non-ERISA pension, the decision, in conjunction with actuarial standards, has clarified the issue of use of retirement age in pension valuation.
Mark K. Altschuler is an actuary and president of Pension Analysis Consultants, Inc. in Elkins Park. He is a contributing author to Valuing Specific Assets in Divorce, edited by Robert D. Feder, which covers issues in choice of retirement age and actuarial standards. He has spoken on the topic of age and service retirement before the Pennsylvania Bar Association.
Reprinted with permission from Pennsylvania Law Weekly