Increase Normal Retirement Age from 55 to 62

[Dennis Povloski]

Plan's normal retirement age is 55, under the new NRA rules, the plan must increase NRA from 55 to 62 (there is nothing to base the earlier retirement age on). The plan has a flat benefit formula with fractional accrual.

It would make sense to me that you have to preserve the accrued benefit, so actuarially increase the age 55 accrued benefit to age 62.

The part that's a little harder for me to wrap my brain around is how to deal with the fractional accrual.

Now that there are more years to retirement, the denominator in my accrual fraction is larger, which means that each year a smaller piece of benefit is accrued. It seems to me that the accrued benefit is now larger than what it would be if I just did a straight calculation using the new accrual fraction, so the accrued benefit would remain unchanged until the point where the new accrual fraction causes the accrued benefit to increase.

Am I on the right track? If so, is there any problem with this "wear-away" type affect?

It has been suggested to me that in conjunction with the change in NRA, that the plan formula also be changed to a unit accrual-type formula.

Thanks!

Dennis

[Guest TooMuchFreeTime]

The "safe harbor" age is 62. If the plan has anything 62 or higher, the IRS will leave them alone.

What happens at 55-62 is that you must justify the use of an age <62, but the IRS will give the employer significant deference in their explanation. In the case of the plan at issue, it sounds like the employer wouldn't have anything to point to to justify a NRA of 55. Notice 2007-69 does a nice job of laying out the new regs without causing too many headaches.

So, quick summary...

62+ You're fine.

55-61: You'd better have a reason, but Service will give deference to your explanation.

<55: You'll need a reason, and you'll have to defend it.

Section 1.401(a)-1(b)(2)(iii) of the 2007 regulations provides that, if a plan’s normal retirement age is between the ages of 55 and 62, the determination of whether the age is not earlier than the earliest age that is reasonably representative of the typical retirement age for the industry in which the covered workforce is employed is based on all of the relevant facts and circumstances. The preamble to the regulations provides that it is generally expected that a good faith determination made by the employer (or, in the case of a multiemployer plan, made by the trustees) that the typical retirement age for the industry in which the covered workforce is employed is an age between age 55 and age 62 will be given deference, assuming that the determination is reasonable under the facts and circumstances.

As for whether you would need to preserve the current benefits as projected to 55 to avoid any cutbacks, The new regs (at 1.411(d)-4, Q&A-12) provide an exception to the cutback rules of 411(d)(6) for an amendment that changes the NRA from an impermissibly low one to a permissible one. In other words, everything gets projected out to the new NRA (62, 65, whatever).

[FAPInJax]

I might consider various document options.

1 Convert the benefit at 55 to 62 and use a floor for future accrued benefits. (This is your suggestion). The formula will work and the participant may not experience any accrual for several years.

2 Another option might be to rewrite the accrual language such that the accrual if for the difference between the grandfathered converted benefit at 62 (which becomes a constant) and the projected benefit. This has the advantage of permitting continued accruals in future years but am unsure whether this type of provision is acceptable.

[ak2ary]

There is an old revenue ruling (from like 1977 or so ) dealing with breaks in service and the fractional rule that, I think, would provide the logic for use here

Participant's NRA is changed from 55 to 62

Participant is age 50

First began accruing at age 40

In essence, using the logic of the revenue ruling, after the change, he will accrue the remainder of his projected benefit prorata over his remaining service after the change. So you would accrue 1/12 of thew difference between his projected benefit and the act equiv of his age 50 acc bft each year from 50 to 62

Acc bft = [{Proj bft - act equiv of acc bft at date of amend} x {yrs since amend/yrs from amend to NRA}] + act equiv of acc bft at date of amend

[Andy the Actuary]

There is an old revenue ruling (from like 1977 or so ) dealing with breaks in service and the fractional rule that, I think, would provide the logic for use here

I believe you may be referring to IRS Rev. Rule 81-11

[Andy the Actuary]

So, quick summary...

(a) 62+ You're fine.

(b) 55-61: You'd better have a reason, but Service will give deference to your explanation.

© <55: You'll need a reason, and you'll have to defend it.

Could someone provide an example of how you would provide an explanation for (b) without having the evidence to support ©?

Also, there are numerous occupations where there will be no industry information. For example, suppose we have a research physician who lives off of grants and whose focus is Hutchinson-Gilford Progeria (premature aging). While you may find industry specifics on academic researchers, they will not represent the typical age at which physicians in his occupation retire. What about if you write the professional body such as the American Society of Dingbats, Wombats, and Claudican Crustaceans and they write back they have no retirement statistics. It seems as if you're caught in (a). In fact, unless the occupation carries an age-related hazzard or physical strain (police, fire, lumberjack), it may be unlikely to find statistics that support an earlier "typical retirement age."

[Belgarath]

I agree with Andy. My experience with small plans is that those professions where the owners make a tone of money, and therefore use a lower NRA to load up contributions/deductions are precisely those professions where in general (of course there are exceptions) they do NOT retire at age 55, or even 62 for that matter.

I'm not sanguine about the amount of "deference" the IRS will actually grant when looking at a given plan.

[AndyH]

" What about if you write the professional body such as the American Society of Dingbats, Wombats, and Claudican Crustaceans and they write back they have no retirement statistics. "

Thanks for making me look up wombat and expand my vocabulary. I already knew what a dingbat was. Somebody else can look up "Claudican"

wom·bat –noun any of several stocky, burrowing, herbivorous marsupials of the family Vombatidae, of Australia, about the size of a badger.

Many of us could become those on April 1.

[GMK]

Perhaps the intended word was "claudicant," which means limping.

Go Badgers.

[Andy the Actuary]

" What about if you write the professional body such as the American Society of Dingbats, Wombats, and Claudican Crustaceans and they write back they have no retirement statistics. "

Thanks for making me look up wombat and expand my vocabulary. I already knew what a dingbat was. Somebody else can look up "Claudican"

wom·bat –noun any of several stocky, burrowing, herbivorous marsupials of the family Vombatidae, of Australia, about the size of a badger.

Many of us could become those on April 1.

The word is claudicant but my left index finger isn't working so well and missed the "t." Claudicant = cripled.

[Guest merlin]

Whether it's claudican or claudicant, cripled or crippled matters not. It still deserves a "Wow"!

[mwyatt]

Correct me if I'm wrong, but my reading here is that if you have an NRA under 62, you need to file w/ the IRS for a ruling (at a modest $9,000 user fee - GMAFB). What are folks doing in the small plan market? Of course the fact that Schedule Bs are no longer filed with 5500-EZs, I wonder exactly how the IRS is planning on figuring out how Citrus Valley part 2 is going to be discovered on their part.

[Mike Preston]

First, with the advent of 415 limits plateauing at 62, almost all of my plans have a 62 or higher NR. But those that are 55 or above are entitled to deference are they not? Probably not going to file them. Under 55? What's 9k to them?

[mwyatt]

Mike, actually the switch to essentially unit credit funding under 430 (as long as you have the benefit accruing in the first 10 years using 133 1/3 v. fractional accrual) will indeed lessen impact. My first impression when this came out was that if you were under 55 NRA, you had to go to the IRS; between 55 and 61 just have good reason but filing not required. Is that correct or is it saying if you have under 62 NRA you need to file (at the $9k rate - BTW, I've gotten in battles with clients who have saved close to 6 figures on taxes over slight bills - for some reason they have the impression that we work pro bono for rich people - can't even imagine what would happen presenting a $9k bill for filing fee).

Think that the real impact of this reg was more the "have your cake and eat it too" crowd who tried to push too far on the cash balance scheme. Liked the concept, but didn't like the interest credits (at least at the time when markets went up). So you would put in a CB plan w/ NRA of say your age as of entry date and allow for distributions at NRA, so that the contribution goes in, and then immediately went out to an IRA for higher investment gains.

[Mike Preston]

I agree that the reg's primary purpose is to snare those CB plans that wanted to distribute as soon as possible. But I still think the reg is clear enough and that the entitled to deference clause will be adhered to by the IRS. What can I say, I'm an optimist.

[Andy the Actuary]

I agree that the reg's primary purpose is to snare those CB plans that wanted to distribute as soon as possible. But I still think the reg is clear enough and that the entitled to deference clause will be adhered to by the IRS. What can I say, I'm an optimist.

And the pessimistic view, "The road to hell is paved with good intentions?"

[Andy the Actuary]

As I get into this issue, I am easily confused and conflicted. Example:

Small physician plan established both practice and plan in 2005 with immediate participation. NRA is 52 and plan has been funded with this target. Assume benefit is 10% of AAC per year of participation. So, at 12/31/2008, participants have accrued a 40% benefit payable at age 52. We now amend the plan for an NRA of 62 and in so doing, grandfather the existing participants so that (1) they may retire on or after age 52 with the same benefit they would have had but with an actuarial increase and (2) there is no provision to retire in-service prior to age 62. Thus, as of 12/31/2009, participant's have accrued a 50% benefit which, if no longer employed, may be taken at age 52 or at a later date with actuarial increase.

Question (1): Since we must value the plan to provide for the greatest subsidy, we continue to target all benefits going out at 52. Is this correct assuming such age provides the greatest subsidy?

Question (2): An employee terminates after 2008, do we value the deferred lump sum to 52 or the actuarial increase of the age 52 benefit to age 62?

In short, apart from differences in actuarial assumptions and the right to an in-service distribution, are we still in the same place?

[Mike Preston]

I think the assumption as to when a distribution is assumed to take place is still within our purview, absent the circumstances where the law takes it away. So, on (1), I'd say yes, IMO. At least, yes based on the fact pattern you've given. On (2), it is up to you subject to the first sentence of this response.