DB valuations as of first day of plan year

[Cynchbeast]

In general, our DB adoption agreements have the valuation date as the last day of the plan year, and our actuary runs the actuarial valuation as of the first day of the plan year. There seem to be some inconsistencies and I am trying to make sense of them. Assuming plan has calendar year:

1. A person is 40% vested on 01/01, and with an additional year of service is 60% vested on 12/31. Actuary shows him as 40% vested on his valuation reports but 60% vested on participant certificate.
2. A person was hired mid-year 2012 and enters plan 07/01/13. The actuary did not pick this participant up on valuation because he was not participant as of 01/01/13. That means he is not shown as accruing any benefit even though he meets requirements for accrual (participant with 1,000 hours).

We used to use the last day of the plan year as the valuation date, but moved it to the first primarily because we don't yet have year-end assets when determining contribution range for sponsor. I though the valuation date of 01/01 would be used simply for valuing the assets, not for determining benefits. How should these inconsistencies be dealt with?

[My 2 cents]

If the plan covered 100+ participants, there would be choice but to perform the actuarial valuation as of the first day of the plan year. It is my understanding that the plan provision defining Valuation Date is used for top-heavy testing purposes and would not have any control over the date to be used by the enrolled actuary as the valuation date for purposes of the annual Actuarial Valuation.

In preparing the participant benefit statements or in performing benefit determinations, follow administrative procedures and basically ignore how the actuarial valuations are being performed (but, except for the use of snapshot data in the actuarial valuation, there should be consistency between them).

[Effen]

I have never seen a document that defines a valuation date - I am not even sure what it would mean or why you would do it, other than for top-heavy testing as 2 cents mentions.

You just need to change your thinking. DB plans are not like DC plans. In your case the valuation is run as of the beginning of the year (BOY). Think of this like a "snap shot" date. Who is in the plan on that date gets included in the valuation. The actuary makes various assumptions about compensation changes, termination rates, mortality and determines the value of the benefits they think will be earned during the current year. The only thing that is generally "real" are the accrued benefits as of the valuation date (BOY). What really happens during the year creates gains/losses in the next valuation. If someone quits or is hired mid year, it does not impact the valuation.

If you are using valuation software to produce benefit statements, you would produce the statements at the BOY based on real data from the prior year. Therefore, the person hired mid year just wouldn't get a statement until the next valuation date.

It sounds like you might be having problem with the timing. If you aren't getting your BOY 13 valuations until mid 2014, then you either need to get the data to the actuary sooner, or you need an actuary who is more responsive. If the 13 BOY vals were done my mid 2013, would you still have the same concerns about the benefit statements? Doing EOY vals might help, but if timeliness is already a problem, they will just compound the problem by forcing a shorter window. Plus, they add complexities in other areas.

[david rigby]

Agree with prior comments.

I'm not sure the plan document and/or adoption agreement could legitimately specify the valuation date (as most actuaries use that term).

The ValDate is part of the plan's funding method (not to be confused with a "funding policy"). It's not a plan provision.

[John Feldt ERPA CPC QPA]

As noted above, some DB plan documents define it for top heavy "valuation" purposes.

[Cynchbeast]

Thank you all for the very helpful comments - especially Effen. What has me a bit confused is that our actuary seems to base some things on BOY and others on EOY:

• He uses asset value as of BOY
• He is including only those who are participants as of BOY
• He uses hours worked and compensation for entire year such that funding for the year is based on accruals required for the year

Omitting mid-year entrants would make sense if 2013 funding were based on a BOY snapshot and accruals for 2012. But It is not; our actuary determines accruals and funding for 2013 based on what happened during the entire 2013 year, yet ignores the participants that entered mid-year. We are still trying to determine if his then gives the participants in question 1 1/2 yr worth of accruals in the next val.

So the sponsor is funding for some participants and not others. If he has a large number of mid-year entrants that are omitted from val, he could be unknowingly underfunding plan by fair amount.

Just seems very inconsistent.

[My 2 cents]

For a beginning of year valuation (required for all defined benefit plans with 100+ participants and acceptable for all other defined benefit plans), the funding liability ("Funding Target" or "Target Liability") is based on the accrued benefits as of the beginning of the year, valued as of the beginning of the year. There is a normal cost component in the minimum required contribution ("Target Normal Cost") based on the expected increase in accrued benefits from the beginning of the year to the end of the year, valued as of the beginning of the year.

People who are not yet participants as of the beginning of the plan year (assuming that the actuarial valuation is as of the beginning of the plan year )are not taken into account, and there is no recognition of them for purposes of this year's Target Normal Cost. If they enter by the beginning of next year, their Funding Target for next year's valuation will be based on their accrued benefits as of the beginning of next year, and their Target Normal Cost fore next year will be based on the benefits expected to be earned by them next year. If they enter and terminate before the end of this year, if there are any unpaid vested benefits, they would be included in next year's Funding Target (otherwise they would not be recognized in any actuarial valuations.

This is the way defined benefit plan funding is required to be handled under the Pension Protection Act of 2006 (effective for most defined benefit plans since 2008).

[Effen]

"He uses hours worked and compensation for entire year such that funding for the year is based on accruals required for the year"

There was a time when people would do BOY valuations using EOY census data. The theory was/is that the "assumptions" just happened to match reality. The IRS has been fairly clear that this is not an acceptable method, however you still see some fighters left on the island who don't know the war is over.

He could be doing it right as 2 cents points out, or he could be doing it "wrong" if he is really using EOY census data on a BOY valuation. You should just ask him why you can't get valuation results earlier and see what he says.

[Cynchbeast]

This really becomes messy as we now have several clients with aggregated DB and PS plans. So how does the actuary cross test with the DB valued as of BOY and PS at EOY? This is really how this all came to light because we have a plan with two mid-year entrants for whom he did not show any PS allocation.

[John Feldt ERPA CPC QPA]

You run the valuation at the beginning of the year. At the same time, you run the testing for the prior year.

In your beginning of the year valuation, you also mention any 401a26 problems that might occur in the current year, you mention any 401(a)(4) problems that might occur in the current year. If the plan is not PBGC covered, you watch out for and mention any possible deduction limit issues that could apply in the upcoming year. For each of these you discuss any needed changes or make suggestions.

In your case, did the new entrants cause a 404 limit issue or something - what was the problem?