Governmental DB plan - hypothetical problems

[Belgarath]

The following situation is of course purely hypothetical. In such hypothetical situation, the services of an actuary would be engaged, as would legal counsel if and as necessary. But in wandering through this exercise to educate myself, would appreciate any input from the DB experts. I could see this being a very difficult VCP filing just due to the potential volume of calculations/payouts that might need to be corrected.

In this hypothetical situation, suppose there is a Governmental plan (state level, not federal, if that matters), currently using volume submitter document. Going back an as yet undetermined number of years, (could be a lot of years) the plan document specified actuarial equivalence assumptions using a sex-distinct table, rather than a unisex table. (As many of you may know, I'm not a DB person, so I'm very hazy on this - is it possible that a table that appears to be sex-distinct to the non-actuarial mind, such as GA ’51 project age 65 (male) is in fact acceptable, whereas the 1951 GAM is not - or are these really the same thing just with different title)?

Also assume that regardless of the specified table, whether acceptable or not, all alternate benefit form calculations and payments have definitely been made on a sex-distinct rather than unisex basis. The plan never had an actuary review any of the calculations for alternate forms of payments - actuary only reviewed funding assumptions/contribution requirements.

First, am I correct that the fact it is a governmental plan doesn't matter, and that unisex rates are required for all alternate forms of payment? (I seem to recall that is what the Norris decision was all about)

Second, is there a choice amongst various acceptable actuarial equivalence rates/assumptions that may be allowed, or is there only one acceptable table/set of rates?

Assuming that there is more than one acceptable rate/assumption(s) allowed, is it permissible, if submitting through VCP, to select rates/assumptions that minimize the additional cost to the employer to correct?

Maybe there is no across the board answer to this one, but if sex-distinct rates were used, I'm under the impression that this would typically result in males getting a higher payout than females for the same accrued benefit at the same ages, due to the shorter life expectancy? But perhaps that isn't necessarily true in all situations? Also, is it possible that an allowable unisex table might produce higher payouts for ALL participants than a given sex-distinct table, but that the amount of the increase is just smaller for women than it is for men, or vice versa?

Any off the cuff thoughts/observations are appreciated. Please don't take a lot of time, as this is hypothetical, and if such a situation were ever encountered in real life, the services of an actuary would be engaged immediately. Thanks!

[tymesup]

A plan is allowed to use a male table or a female table or a unisex table to determine actuarial equivalence. What it is not allowed to use is one table for males and another table for females.

[Belgarath]

Thank you - that clears up a lot of questions (for me, at least) and allows the focus to be on the required corrective calculations/distributions/overpayment recovery/etc., whatever those might be.

[My 2 cents]

A plan is allowed to use a male table or a female table or a unisex table to determine actuarial equivalence. What it is not allowed to use is one table for males and another table for females.

in 1978, in Los Angeles Water and Power versus Manhart (a governmental plan case where longer female longevity was used to justify higher costs being charged to female employees), the U. S. Supreme Court ruled, 6-2, that it was illegal to take into account the differences in mortality between men as a group and females as a group with respect to any aspect of employee compensation or benefits. Consequently, no employment-related benefits can use different factors for males and females. This is a matter of the Civil Rights Act and the 14th amendment to the constitution. Subsequently (i.e., for nearly 40 years), no employee benefit plan has been permitted to apply different factors in determining such things as early or late retirement benefits or conversion to an optional form of benefits. Defined contribution plans have not been permitted to divide account balances by sex-distinct factors to calculate the amount of annuity payments that can be provided by an account balance. Sex-distinct annuities can be purchased from insurance companies, so long as the actual benefits payable are independent of the differential the insurance companies may be charging.

Speaking as a non-lawyer: This is not an issue of an ERISA or IRC violation where a corrective filing can fix things. It is outright sex discrimination. This probably should be resolved under a class action lawsuit.

If the sponsor has availed itself of legal advice, one presumes that this practice could not have been brought to the attention of the legal advisor. This practice could not possibly have been condoned by a competent legal advisor.

As for the plan, if it defines actuarial equivalence using the 1951 Group Annuity Table for males, that would be fine if applied on a unisex basis.

[david rigby]

While reviewing the Manhart case, it is recommended that you also review the "Norris" case (1983? 1988?).

BTW, both were governmental plans.

[Belgarath]

"Speaking as a non-lawyer: This is not an issue of an ERISA or IRC violation where a corrective filing can fix things. It is outright sex discrimination. This probably should be resolved under a class action lawsuit."

So do you think that if not corrected there are no plan qualification implications? Or, would you say that it must be corrected to retain plan qualification, but there are additional issues unrelated to ERISA/IRC, as you said above?

P.S. reviewed a synopsis of the Norris case, as David suggested.

[My 2 cents]

To the extent that the situation is not actually hypothetical, my vote is that everyone who was ever adversely affected (at least since the Norris decision) must be made whole, even if the person died 30 years ago. This was really common knowledge then and every day since. No excuses, no time limits, no whining about the cost of fixing it. Make everyone whole.

This is over and above (although probably a necessary part of) any action needed to deal with plan qualification issues.